![1
CHAPTER
A review on Internet of
Multimedia Things (IoMT)
routing protocols and
quality of service
Dinesh Singh, Ashish Kumar Maurya, Rupesh Kumar Dewang, and
Niharika Keshari
Department of Computer Science & Engineering, Motilal Nehru National Institute of Technology
Allahabad, Prayagraj, India
1.1 Introduction
During the last decade, Internet of Things (IoT) has grown very fast and connected
numerous devices worldwide. IoT can provide anytime, anyplace connectivity for
anyone and anything [1,2]. Presently, 23.8 billion units of devices are connected to
the Internet, and this amount is expected to jump 41.2 billion units by 2025 [3].
The IoT devices have limited memory, size, energy, and computing capabilities [4].
Thus, these IoT connected devices highly depend on reliable communication network
and efficient routing protocols [5]. Technical developments in IoT operating systems
[6], IoT enabling platforms [7], 5G IoT [8,9], wireless sensor networks (WSNs) [10,
11], software defined networks (SDNs) [12], mobile ad hoc networks (MANETs)
[13–16], vehicular ad hoc networks (VANETs) [17,18], fog/edge computing [19,20],
cloud computing [21], are facilitating IoT to comprehend to connect anything and
anywhere [22].
The massive growth in multimedia on-demand traffics such as images, audios,
and videos, has evolved the concept of Internet of Multimedia Things (IoMT) from
IoT [22]. IoMT devices produce massive amount of multimedia data and are more
constrained than IoT devices. They need more processing power, massive memory
storage, adequate bandwidth, and consumes more power to support the underlying
application efficiently [22,23]. The comparison between key data characteristics of
IoT and IoMT [22] has shown in Table 1.1. The IoMT has diverse applications like
smart grid, smart cities, industrial IoT, smart homes, health IoT, smart farming and
agriculture, traffic monitoring, soil health monitoring, and satellite control systems.
Most of these applications are interactive applications that involve reliable and timely
delivery of data. Hence, it demands efficient routing protocols, and enforces stringent
quality of service (QoS) parameters [23]. Due to transmission of unstructured, bulky
and multimedia data over the network, IoMT needs revision of existing routing pro-
Internet of Multimedia Things (IoMT). https://doi.org/10.1016/B978-0-32-385845-8.00006-X
Copyright © 2022 Elsevier Inc. All rights reserved.
1](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-20-320.jpg)
![2 CHAPTER 1 A review on IoMT routing protocols and quality of service
Table 1.1 Comparison between key data characteristics
of IoT and IoMT.
S. No. IoT data Multimedia data
1 Linear data Gigantic data
2 Limited bandwidth Adequate Bandwidth
3 Less memory storage Massive memory storage
4 Limited processing Excessive processing
5 Delay tolerant Delay sensitive
6 Less power consumption More power consumption
tocols of IoT which focus on energy aware computing, efficient feature extraction,
and optimizing routing criteria to minimize delays [23]. The quality of service per-
formance in IoMT is determined by various metrics such as bandwidth, packet loss
ratio, delay, throughput, resource management, and energy conservation [23]. The
quality of service in technical white paper [24] is defined by Microsoft as: “Network
QoS refers to the ability of the network to handle this traffic such that it meets the
service needs of certain applications.” The frequent packet loss and redundant packet
delivery degrade the quality of constrained multimedia applications in IoMT network.
The massive amount of data produced in IoMT network from many heterogeneous
sources creates the processing overburden on the routing devices, resulting in packet
loss.
The multimedia traffic is growing very fast which increases newer challenges for
computing, sharing, storing, and transmitting the multimedia data. Computing data
in IoMT needs novel methods for fog/edge and cloud computing devices. Similarly,
for storing multimedia data, different compression/decompression methods are re-
quired in IoMT [22]. For example, utilizing a standard IoT routing protocol called
RPL (Routing protocol for low-power and lossy networks) [25] in IoMT deployment
scenarios, requires more improvements in terms of fault tolerance, energy-awareness,
delay-awareness, and load balancing [22]. The Green-RPL [26], Context-Aware and
Load Balancing RPL (CLRPL) [27], and free bandwidth (FreeBW)-RPL [28] are
some modified versions of RPL protocol for IoMT. In this chapter, we give a com-
prehensive review on routing protocols and quality of service in IoMT.
The remainder of the chapter is organized as follows: In section 1.2, we illustrate
the working methodology of routing protocols used in the IoMT network. The QoS
routing protocols are explained in section 1.3 of this chapter. The conclusion and
future directions is presented in section 1.4.
1.2 Routing protocols in IoMT
Wireless Sensor Network (WSN) performs a crucial role in assisting IoMT in diverse
application spectrum. The energy constraint of nodes in WSN somehow restricts it to
fulfill the expectations of IoMT applications. The efficient routing protocols save the](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-21-320.jpg)
![1.2 Routing protocols in IoMT 3
energy of WSN nodes and thus, allow to carry off a massive amount of IoMT data.
Recently, cluster-based routing has attracted due to less communication overhead and
reduced routing energy consumption. The cluster members of a cluster can save their
routing energy expense because of the routing assistance of the cluster head. But,
since everything has its pros and cons, cluster-based protocol has its disadvantage.
In case of battery discharge or malfunctioning of the CH, all the cluster members
would fail to transmit their collected data to the destination, which would defeat the
smooth functioning of the IoMT application. Thus, the fault-tolerant approaches are
in use in cooperation with cluster-based routing protocol for IoMT applications. Here,
we illustrate the working principle of a fault-tolerant-based routing protocol used for
IoMT.
1.2.1 Fault tolerant routing protocol
The fault-tolerant routing protocol [29] uses cluster-based scenario for the processing
of IoMT application data. It uses the cluster join method for handling faulty cluster
heads (CHs). Whenever a fault is detected in a CH, its cluster members are the first
ones to catch it as they fail to receive an acknowledgment from their CH upon receiv-
ing transmitted data. So, they broadcast a help message to their neighboring CHs for
re-selecting their CH. Out of all the respondents of the help message, the closest one
is selected as the best backup CH. But, even this method has some leftover loopholes
described as follows:
• If there are many faulty CHs, they will introduce a new problem of help message
explosion.
• Each member of the faulty CH cluster may select a different CH as its backup for
data routing, and then there would be a problem in handling that cluster’s data.
The possible solutions to the above problems are:
• One of the solutions is the re-clustering of the entire network nodes. But, this is a
laborious and costly process.
• Another solution is to keep a CH entirely for backing up the failed CHs by not
initially assigning a cluster to it.
• The other solution includes identifying overlapping nodes (two nodes having sim-
ilar coverage area) and putting one of such nodes into sleep mode, then wake it up
only when the existing CH fails.
In view of these solutions to overcome the issues, a fault-tolerant routing protocol
is illustrated in [29]. The protocol performs the following operations:
1. The protocol initially detects the failure in CHs and record faulty and nonfaulty
CHs.
2. It analyzes the energy levels of nonfaulty CHs and attempts to construct the virtual
CH. The data is routed through this virtual CH to the destination node.](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-22-320.jpg)
![4 CHAPTER 1 A review on IoMT routing protocols and quality of service
3. The maximum fault-tolerant capability of nonfaulty CHs is calculated to accom-
modate the members of faulty CHs.
4. The protocol uses a flow-bipartite graph (FBG) to estimate the energy cost of
IoMT data routing. A flow-based pairs of faulty and fault-free CHs are formed to
create an FBG.
5. Using the FBG, a flow transmission pattern is identified between the faulty and
fault-free CHs. This indicates that which of the faulty CH is tolerated by which of
the fault-free CHs.
The virtual CH and FBG construction methods used in this routing protocol are
explained below:
Virtual CH
The steps involved for creating of a virtual CH and virtual super frame are:
1. The destination node organizes the available energy of all the failure-free CHs, as
shown in Fig. 1.1 [29]. There are three components in that total available energy:
the energy taken for receiving the IoMT data from all the failure-free members,
the energy is taken to aggregate the IoMT data obtained from the cluster members,
and the energy taken to route the aggregated data from the CH to the destination
node.
2. The remaining energy of a failure-free CH after using the above three components
is used for virtual CH construction. The failure-free CHs receive IoMT data from
failure-affected cluster members, aggregate the data and then transmit them to
the destination node, as shown in Fig. 1.2 [29]. But, which failure-free CH is
tolerated by which of the faulty node remains unknown. Thus, the concept of
average transmission energy consumption is used to analyze energy used in fault
tolerance.
FIGURE 1.1
Organization of the virtual CH with the available energy of all the fault-free CHs.](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-23-320.jpg)
![1.2 Routing protocols in IoMT 5
FIGURE 1.2
Data aggregation of faulty CHs by fault-free CHs.
3. A virtual superframe structure is constituted for some failure-free and some
failure-affected members. The number of such virtual super frames transmitted
by the virtual CH is defined as its transmission capability. Thus, the transmission
capability is derived by dividing the sum of the energies of all failure-free CHs by
sum of failure free and fault-tolerant energy consumption.
4. Finally, the verification of fault-tolerant capability begins. After confirmation of
one or more CH failures, the minimum data of faulty CHs required to be handled
by the fault-free CHs is calculated. If that data is greater than the availability
of data received in fault-free CHs, then the purpose of IoMT application is not
solved.
Flow-Bipartite Graph
After creating virtual CH, flow-based pairs of faulty and fault-free CHs are formed
to develop a flow-bipartite graph (FBG). An FBG, as shown in Fig. 1.3 [29], is a
graph whose vertices are divided into the source and destination sets such that every
edge connects a vertex in source to a vertex in destination and is associated with
a transmission cost and an energy cost. Each destination vertex is attached with a
capacity cap. To establish an FBG, the following steps must be followed –
1. All faulty CHs act as source, and failure-free CHs as destination node.
2. The amount of input flow from each source vertex is set to the demanded amount
of data (i.e., the data to be gathered at failure-free CHs due to failed CHs).
3. The capacity cap of the destination vertex is set to the available energy of the
failure-free CHs.
4. The distance between the farthest member of a faulty CH and a fault-free CH
is calculated and compared with a node’s communication range. If the former is
larger, an edge can be created between those source and destination nodes.](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-24-320.jpg)
![6 CHAPTER 1 A review on IoMT routing protocols and quality of service
FIGURE 1.3
An example illustration of flow-bipartite graph formed by faulty and fault-free CHs.
5. The transmission cost of an FBG is calculated. It is the sum of cost of transmitting
data from failure affected members to the chosen fault-free CH and for transmit-
ting from failure-free CH to the destination node.
Now, the fault-tolerant energy cost of FBG is calculated.
1. It is the sum of energy taken to assist the transmissions of failure-affected mem-
bers to the destination node and the energy taken to transmit the IoMT data of the
failure-free cluster members to the destination node.
2. With this, fault-tolerant load distribution can be achieved by making two or more
fault-free CHs tolerate a single faulty CH.
3. The minimum cost flow (MCF) problem is also solved using the FBG approach
such that the total minimum transmission cost can be obtained.
1.2.2 DDSV routing protocol
An optimizing Delay and Delivery Ratio for Multimedia Big Data Collection in Mo-
bile Sensing Vehicles (DDSV) routing protocol [30] uses optimized routing criteria
so that the delay involves in data collection and data delivery is minimum. It consid-
ers a road scenario that has multiple intersections on a fixed distance. The two types
of moving vehicles, Bus and taxi, are considered on the road to receive the IoMT
data generated from the source, i.e., traffic light. The vehicles deliver the received
IoMT data to the other vehicles or the Data Collection (DCs) centers located at the](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-25-320.jpg)
![1.2 Routing protocols in IoMT 7
FIGURE 1.4
DDSV routing scenario.
intersection of the road segment based on the optimized routing criteria. The DDSV
routing scenario is shown in Fig. 1.4 [30].
The route for IoMT data transfer consists of intersections and road segments. The
protocol takes data packet priority as well as vehicular priority for decision-making
purposes. The data packet priorities are assigned on the basis of location or area from
where the data is generated. The routing scenario is partitioned into many urban and
suburban sections.
1. At each intersection i of the road, the routing decisions are taken. The routing
decision is based on the minimum of expected data delay (Di) on the possible
routes.
2. The expected data delay (Di) is computed as the function of two components:
movement probability (Pε
i,j ) of vehicle ε at intersection i to the routing decision
θi and data forwarding probability (dε
i,j ).
3. There are multiple routes between intersection i and intersection j to transfer
the IoMT data. The movement probability (Pε
i,j ) of vehicle ε at intersection i is
computed as
pε
i,j = α × Pε
i,j + (1 − α) × (1 − Sε) (1.1)
where Pε
i,j is the priority of vehicle ε and Sε is the probability of the vehicle ε
to travel in the suburban area. The Sε is defined as the ratio of time spent by
a vehicle in suburban area to the time it taken in moving as per the historical
trajectory datasets.
4. A vehicle finds the following possibilities, as shown in Fig. 1.5 [30], at each in-
tersection i to route the IoMT data.](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-26-320.jpg)
![1.2 Routing protocols in IoMT 9
b. The taxi type vehicles have no fixed trajectories and thus, the probability
considers the transmission range, vehicle speed as well as the wireless trans-
mission delay for computation. It is calculated as:
dε
i,j =
1 − e−r.oi,j
×
li,j × c
r
+ e−r.oi,j ×
li,j
vε
i,j
(1.4)
Where r is the transmission range of vehicle’s Wi-Fi channel, c is the wireless
delivery delay, oi,j and vi,j is the density degree and vehicle’s speed on the
road segment between intersection i and intersection j, respectively.
6. For routing purpose, it is assumed that the priority of the vehicle moving in the
suburban area is high as compared to vehicles moving in urban area.
1.2.3 Optimal routing for multihop social-based D2D
communications
An optimal routing algorithm for multihop communication between Device-to-
Device (D2D) is given in [31]. The algorithm is well suited for efficient IoMT data
communication and 5G networks. It considers the social behavior of the devices in
communication to their neighbors. The algorithm computes the trust probability for
D2D communications based on the rank model. Both, random and fixed locations
of the base stations are considered for measuring the connection probability (CP)
between any pair of devices. The measurement of CP is taken with and without,
channel state information (CSI). The network model of the Social-aware multihop
D2D routing algorithm is shown in Fig. 1.6 [31]. The transmission of information
between D2D transmitter and D2D destination using the number of D2D relays is
taking place. Here, the interference due to the cellular communication equipment
(CUEs) and position of base stations (BSs) is considered in follow-up. The Nodes
are working in half-duplex mode. A single antenna and D2D transmitter are assumed
to be at the origin, and D2D destination is at a fixed distance away from the origin,
whereas using the Poison Point Process (PPP), BS and CUEs are modeled.
With the help of real data traces from online social networks, trust connectivity
of D2D based on rank-based trust model is calculated. The probability that Di+1 is
trusted by Di is given as
P
(t)
i,i+1 =
1
GR
β
i (i + 1)
(1.5)
where β is the parameter from the rank-based model and G =
N
n=1[ 1
nβ ] is a nor-
malizing factor.
A trusted connection will be established only when two nodes can trust and com-
municate with each other and thus, the trusted connectivity probability (T-CP) is
defined as
P(c)
i,i+1 = P(t)
i,i+1P(c)
i,i+1 (1.6)](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-28-320.jpg)
![1.2 Routing protocols in IoMT 11
algorithm which helps in selecting optimal path. The optimal path satisfies:
PT −CP
π∗
= max
π∗ Sπ
i Sπ
Pi,i+1 (1.7)
where Sπ denotes the set of all potential paths between the D2D transmitter and
receiver. The routing algorithm finds the maximum T-CP between source and desti-
nation in a weighted graph with the help of standard Dijkstra’s algorithm. Initially,
at each D2D node, distance between itself and other D2D devices and BSs are cal-
culated and stored in topology information, which contains the neighbor list. The
adjacency T-CP matrix is obtained by using updated topology information. There-
fore, the D2D transmitter is set up as permanent node and all other nodes as temporary
nodes. Here, we calculate the T-CP for all possible neighbors links and find the link
with maximum T-CP. The intermediate destination node in selected link with max-
imum T-CP is set as permanent node and now keeps on finding maximum T-CP
unless all the nodes are marked as permanent node. Thus optimal path and maximum
TCP of multihop D2D communication are obtained. The Computational complexity
of routing algorithm is as same as that of Dijkstra algorithm, i.e., O(N2).
1.2.4 Green-RPL routing protocol
The Green-RPL [26] is an energy-efficient green routing protocol for IoMT. It is an
enhanced version of the RPL (Routing Protocol for Low-Power and Lossy Networks)
protocol [25]. RPL is a routing protocol for resource-constrained devices that uses
Destination Oriented Directed Acyclic Graph (DODAG) to maintain network topol-
ogy. This DAG comprises multihop routes from leaf nodes to the root node. RPL
optimizes an objective function and chooses the best path by selecting the desired
predecessor nodes starting from leaf nodes. The previous RPL protocol implementa-
tions are not feasible for IoMT and do not consider the multimedia data.
In contrast, the Green-RPL protocol considers the data generated from multimedia
devices. The Green-RPL protocol reduces energy consumption and carbon footprint
emissions together with QoS requirements of applications. To guarantee QoS for a
particular multimedia application, it determines the delay bound for the application.
For example, in VoIP applications, the delay limitation usually is 120 msec. To ensure
energy efficiency, the protocol considers the features of all the intermediary links
between the leaf nodes and the root node and estimates the energy consumption by the
chosen immediate predecessor node to support traffic needs for one more immediate
successor node. An optimization model is given for this protocol based on various
requirements and constraints.
In Green-RPL routing protocol, the desired immediate predecessor node is se-
lected according to the objective function minimizing the emissions of cumulative
path carbon footprints on all the links from the immediate predecessor node to the
root node, and satisfying the constraints such as cumulative path link energy, cumula-
tive path delay, idle time, and battery status of the immediate predecessor node. When
more than one immediate predecessor node satisfies the given constraints, then the](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-30-320.jpg)
![12 CHAPTER 1 A review on IoMT routing protocols and quality of service
node that provides the most greener path is chosen as the desired immediate prede-
cessor node. In cumulative path link energy constraint, energy consumption by a node
is calculated based on the quality of links in the selected route. The cumulative path
delay constraint specifies that the predefined delay threshold should not be increased.
According to the idle time constraint, an immediate predecessor node should be cho-
sen as the desired node only if its idle time is sufficient to support another child node.
In the last constraint battery status, a child node should choose an immediate prede-
cessor node as the desired one if the battery level of its energy resources is greater
than the predefined threshold.
1.2.5 Context-aware and load balancing RPL (CLRPL)
A modified version of RPL, known as Context-aware and load balancing RPL under
heavy and highly dynamic load for IoT and IoMT applications is discussed in [27].
It addresses the power depletion and packet loss problems associated with RPL. In
CLRPL, one objective function called as Context-Aware Objective Function (COAF)
and one routing metric called as Context-Aware Routing metric (CARF) has been
given based on the contexts of the nodes. Here, context means any information related
to nodes in the IoT network that can be used to characterize the nodes.
Through COAF, CLRPL selects the immediate predecessor of nodes by finding
their rank, Expected Transmission Count (ETX), and residual power level. Normally,
ETX is used to give a realistic estimate of the channel status that is based on link
losses. To avoid the loss during the initial phases of congestion and monitor both
queue status and link status, ETX is used with other metrics that can give the sta-
tus of the node queue in the network. The objective functions used in the previous
implementation of RPL do not consider the latter states of nodes in the route while
finding the rank. The rank value of a node may suffer from queue overflow and power
drainage, which may further affect the rank of all downward nodes in the path. Thus,
COAF recursively computes the latter states of the sequence of nodes in the route to-
wards the root. COAF circumvents the thundering herd effect in the network, which
happened in RPL, by actively considering the network states and gradually shifting
toward the real rank value from a high-rank value.
The protocol uses the routing metric, CARF, which considers the network traf-
fic dynamicity index, node rank, and buffer queue utilization to calculate the route
throughout heavy traffic conditions. Here, the network traffic dynamicity index shows
that how much link has been utilized in the past. CARF reduces the effect of upstream
immediate predecessors as it grows a long way in the path toward the root. Instead of
just considering only the rank of nodes, this metric considers the other factors related
to nodes and networks to select the suitable immediate predecessors in a network
with heavy traffic. CLRPL avoids routing loops through the use of a countermeasure
while selecting the best immediate predecessors using CARF and other parameters.
This protocol also ensures load balancing by taking the number of nodes having an
immediate predecessor from the list of candidate immediate predecessors for selec-
tion of the immediate predecessor.](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-31-320.jpg)
![1.2 Routing protocols in IoMT 13
Another enhanced version of RPL protocol named free bandwidth (FreeBW)-
RPL is given in [28] which suggests a novel objective function called FreeBW that
takes the FreeBW computations in the network layer. This protocol calculates the
maximum FreeBW using the volume of the bandwidth to select the routing path to
provide better performance.
1.2.6 Energy-Harvesting-Aware (EHA) routing protocol
Energy harvesting (EH) is evolving as an essential technology for IoT and field de-
ployable WSNs applications. In these applications, the deployed nodes or devices
may recharge their batteries to perform actions by extracting energy from renewable
sources such as radio frequency (RF) and solar-based radiation through the use of
EH techniques. Energy-Harvesting-Aware (EHA) routing protocol proposed in [32]
addresses the issues of Quality of Service and energy efficiency for IoT applications
at the network layer. The protocol can efficiently perform according to the harvested-
energy at IoT devices, the residual energy, and the dynamic traffic capacity produced
by IoT applications. EHA routing protocol is based on the energy prediction process
and energy-backoff process that describes cost metrics to choose the best path. In
the energy prediction process, an enhanced estimate of existing energy levels from
different sources has been determined using the Kalman filter method that considers
current statistics and previous time steps. In the energy backoff process, network life-
time has been increased by keeping the nodes having the lowest energy in sleep mode
until their energy level is restored to perform operations. The protocol also considers
hybrid energy harvesting sources and uses three EH techniques such as solar-based
EH technique, moving vehicle-based EH technique, and RF-based EH technique. The
EHA routing protocol first selects the node having the maximum level of residual en-
ergy and then finds the link with the least cost using the shortest path algorithm given
by Dijkstra to send the data packets to the target node from the origin nodes. Further,
by using the EH techniques, the protocol aims to determine the sustainable paths to
route the data packets.
1.2.7 Optimized 3-D deployment with lifetime constraint (O3DwLC)
protocol
The deployment of sensing devices is a very challenging task to satisfy both robust
connectivity and the long lifetime of devices subject to cost minimization. The most
common solution for handling such a situation is to install the relay nodes in the
target environment. The function of relay nodes is to forward the sensed data up
to the maximum distance sink node and thus requires additional storage and more
power. The energy of sensing nodes can be saved for further collecting forwarding
data. But installing the relay nodes is more challenging because of the roughness
of the environment and the requirement of 3-D infrastructure. The roughness of the
environment is due to natural calamities and nonexpected visiting birds and animals
because they may destroy nodes. For maximizing connectivity with little cost and](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-32-320.jpg)
![14
CHAPTER
1
A
review
on
IoMT
routing
protocols
and
quality
of
service
Table 1.2 Comparative analysis of routing protocols in IoMT (Part a).
Paper Architec-
ture
Entities Communica-
tion
mode
Link
establishment
criteria
Suitability criteria Optimal path
finding
Noiseless
path
selection
Environ-
ment
adaption
Goal Methodology
[29] Cluster Sensor
nodes, Sink
nodes
Device-to-
device
Transmission
energy
consumption,
Distance
Transmission
Cost, energy cost
Integer Linear
Programming
model solved
using IBM ILOG
CPLEX Optimizer
x x Designed a join
based method
which can
perform
preverification of
fault tolerance,
fault tolerant load
distribution and
fault tolerant cost
minimization.
A cluster based
fault tolerant
routing which
minimize total
energy
consumption
during packet
transmission. The
proposed
approach uses
virtual clustering
and flow graph
modeling to
handle faulty CH.
[30] General IoT
multimedia
storage
devices,
vehicle (bus
and taxi
type), DC
Vehicle-to-
vehicle
Message
establishment,
time spent on
road segment,
density, speed,
transmission
range
Prioritize over
probability of
delivery at nearest
data centers
Designed a
threshold based
optimal route
policy algorithm
x x Goal is reduction
of packet drop of
low priority data
and task delivery
within deadline.
Proposed routing
policy based on
delivery delay and
delivery
prioritization of
data.
[31] General IoT device,
Base station
(BS)
Device-to-
device
Distance,
density
Rank based
Trusted
Connectivity
Probability (T-CP)
Dijkstra Algorithm x Carried out an
analysis for cases
at known and
unknown channel
state information.
Author proposed
an approach of
Social relationship
based on T-CP
for multihop
routing path
selection using
tight lower bound
and exact closed
bound with
decode and
forward
mechanism.
(continued on next page)](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-33-320.jpg)
![1.2
Routing
protocols
in
IoMT
15
Table 1.2 (continued)
Paper Architec-
ture
Entities Communica-
tion
mode
Link
establishment
criteria
Suitability
criteria
Optimal path
finding
Noiseless
path
selection
Environ-
ment
adaption
Goal Methodology
[26] Tree
topology
Multimedia
device
Cumulative
path carbon
footprints,
cumulative
path delay,
cumulative
path link
energy, battery
status, idle
time for parent
node
Delay constraints,
battery
consumption,
type of energy
resource
– An adaptive and
dynamic
optimization
based RPL which
can support
multimedia
communication.
Proposed a
Green-RPL
routing to
minimize carbon
footprint
emissions and
energy
consumption
using tree like
network
topology.
[27] Tree
topology
IoT devices Point-to-point,
multipoint to
point, point to
multipoint
Remaining
power status,
queue
utilization for
parents to root
chain
Rank based
Context aware
routing metric
(CARF)
– x x Loop free and
less packet loss
as well less
power depletion
for RPL under
heavy and
dynamic load,
eliminates
thundering hard
and equality
illusion problem
Context aware
load balance RPL
with DODAG
structure find
route using two
steps as: 1)
CARF, 2) parent
selection
mechanism.
[32] General Sensors,
router
Device-to
device
Distance,
battery
voltage,
packet length,
data rate,
Energy
consumption
during listening,
transmitting,
receiving and
sleeping
Dijkstra Algorithm x Aimed to
maximize life-time
in an energy
efficient manner.
Proposed a
Kalman Filtering
(KF) adaptive
energy harvesting
routing by
enhancing energy
back-off
parameter. If
harvested energy
is high then use
current energy
and store the
excess energy](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-34-320.jpg)
![16
CHAPTER
1
A
review
on
IoMT
routing
protocols
and
quality
of
service
Table 1.3 Comparative analysis of routing protocols in IoMT (Part b).
Paper Merits Demerits Evaluation
metrics
Simulation
details
Assumption Complexity Designed
approach
Comparison Tool
[29] Average backup
energy
consumption for
new CH without
aggregation is
0.1856 mJ while
proposed
approach is
around 0.0098
mJ
Average failure
affected as
energy
consumption is
reduced around
10% at 0.2 failure
rate while 4%
increasing at
failure rate up to
0.5
Average backup
energy
consumption,
Average
failure-affected
energy
consumption,
Network lifespan
after fault
tolerance
Sensor nodes
(1000 with 10%
powerful sensor
and 90% sink
node), Sensed
data (500 bytes),
failure threshold
(0.2–0.5),
Simulation run
(50)
Same data size
sensed by each
member
– Virtual CH, Flow
bipartite graph
techniques
New CH without
aggregation, New
CH with
aggregation,
distributed join
CH, distributed
join CH with
multiple factor
NS2
[30] Average delay
during data
re-collection is
reduced around
17.3% for general
and 41.8% for
suburban and
average data
delivery ratio
improved up to
16.9% with
respect to OVDF.
Delay decreased
percentage graph
turns towards
deteriorate
around 11.11%
at 140 with
respect to 120 in
its own scheme.
Average delay,
Delivery ratio,
Data packet
impartiality
Multidataset of
Beijing city of
“T-drive
trajectory”, Data
packet size (512
bytes), Data
storage size
(10M),
Communication
range (150
meters), vehicles
(30–150)
– – DDSV (delay and
delivery ratio for
sensing vehicles)
OVDF −
[31] Optimal path
route by
proposed
algorithm is same
as ES.
Optimal path
selection is
depends on BS
position
Connectivity
Probability
Nodes (20), noise
variance (1), Path
loss exponent (4)
Node equipped
with half duplex
mode single
antenna.
O((log n-2)!) Trusted
Connectivity
Routing
Algorithm
Exhaustive
search (ES)
Monte Carlo
simulation
(continued on next page)](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-35-320.jpg)
![1.2
Routing
protocols
in
IoMT
17
Table 1.3 (continued)
Paper Merits Demerits Evaluation
metrics
Simulation
details
Assumption Complexity Designed
approach
Comparison Tool
[26] Average packet
delay is 20.4
msec, while ETX
and OF0 requires
47 msec and
26.2 msec.
Energy
consumption
depends on path
link quality. At
poor link
successful
delivery using
single hop
probably reduced
and causes
retransmission.
Number of
packet delivered,
energy
consumed.
ICMPv6, packet
size (128 byte),
datarate
(25 pkt/sec)
– – Green-RPL ETX, OF0 Cooja
Simulator for
Contiki-OS
[27] RPL experience
32% packet loss
while proposed
approach packet
loss is 15% at
100 nodes.
Instability and
inconsistency
increase at new
node joining and
frequent
messaging.
Queue loss ratio,
packet loss ratio,
lifetime,
overhead,
Runtime
Number of nodes
(50), Traffic rate
(15–100 pkt/min),
Packet format
(IPv6),
Transmission
power (15 dBm),
FIFO queue size
(20 pkts), Range
(100),
– – CLRPL RPL Cooja
Simulator for
Contiki-OS
[32] Less energy
consumption wrt
R-MPRT-Mod
(12.54%) and
AODV-EHA
(52.3%).
Works for very
small packet size
(1000 bits)
Average
consumed
energy, lifetime,
Alive nodes,
Packet loss ratio,
end-to-end delay
Data-rate
(250 Kbps),
Simulation period
(12 hr),
End-to-end delay
(0.5%),
Throughput
(5–80 pkt/s)
– – EHARA R-MPRT-Mod,
AODV-EHA
MATLAB®](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-36-320.jpg)
![18 CHAPTER 1 A review on IoMT routing protocols and quality of service
FIGURE 1.7
A two layered hierarchical and 3-D grid based hierarchical cluster based arrangement of
sensor nodes.
long-life sensing devices, installing and placing 3-D relay nodes is the most optimal
solution to handle the challenges of the wireless network in a harsh environment [33].
For improving performance, a two layered hierarchical setup can be assumed in
large-scaled environment. Lower layer nodes (SN) gather data and send it cluster
heads (CH) in the upper layer. The upper layer contains relay nodes (RN) for better
transmission. Cluster nodes summarize gather data from the lower layer, and then
with the help of relay node, it sends to the base station (BS) in the upper layer. In a 3-D
cubic grid-based model, the target environment is assumed partitioned in a grid-based
cluster in hierarchical manners. Cluster heads (CH) are placed in the most appropriate
vertices surrounded by the largest number of sensor nodes, and the base station is
placed on the fixed position as per requirement. For efficient connectivity between
cluster heads and base station (BS), it is mandatory to seek proper placement of relay
nodes in grid subject to minimize cost and maximize connectivity. Fig. 1.7 which is
taken from [33], shows a two-layered hierarchical and 3-D grid-based hierarchical
cluster-based arrangement of sensor nodes, respectively.
The main agenda is to search the proper placement of RNs so that cost of deploy-
ment could be minimized by reducing the total counts of CHs/RNs in model with the
constraint of strong connectivity, low cost, and longer network lifetime. The overall
cost can be minimizing by reducing the costlier equipment for the network scenario.
Since the cost of equipment is dependent on the functionality (RNs in this case be-
cause of additional storage and forwarding capacity) and thus, cost can be minimized
by adjusting the number of RNs. The strong connectivity can be achieved by optimal
design and reducing obstacles between communicating devices, and therefore, the
grid model is more suitable for this purpose.
Algorithm description
Turjman et al. [33] have introduced an Optimized 3-D deployment with Lifetime
Constraint (O3DwLC) algorithm to achieve the objective of minimal cost, long net-
work lifetime, and high connectivity. This algorithm works in two phases. In the first
phase, they designed a connected backbone structure using First Phase Relay Nodes
(FPRN). These FPRNs are placed in an optimized manner so that the requirements](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-37-320.jpg)
![1.2 Routing protocols in IoMT 19
of FPRNs become minimum for connecting cluster heads to the base station. Thus, a
minimum cost spanning tree (MST) is used between vertices in the grid model. MST
can be constructed using the 3-D grid vertices pointing location of nodes. Further,
the algorithm seeks the two closest nodes in the set of nodes, CHs and BS. In case
the closet two nodes are not adjacent on the 3-D grid, then it introduces again a mini-
mum number of grid vertices on which the relays have to be positioned to establish a
path between these two nodes. After connecting the closest two nodes repeatedly, the
algorithm looks for the next closest nodes and so on. The extra relay nodes (SPRN)
can be introduced in optimal manner to maximize network backbone connectivity.
The connectivity can be computed by analyzing the consecutiveness of graph nodes
in the grid. The algorithm locates optimized SPRNs position in the second phase of
processing to maximize lifetime and minimize the backbone generated in the first
phase.
For routing multimedia data, the algorithm’s performance is measured based on
backbone connectivity level, the number of CHs, RNs, BS nodes used, and the num-
ber of turns. The number of turns measures the total turns for the deployed network
in which it can stay operational.
Discussion. There are many cluster-based [29], general (non-cluster-based) [30,31,
27] and tree-topology-based [25,26] routing protocols are discussed in Table 1.2 and
1.3. In [29], the proposed work avoids backup workload shifting to new CH by re-
joining new cluster according to energy suitability. Although in this research work
sensed data size were same for all sensor and no deadline, which is an irrelevant
assumption. To prevent the challenge of [29], a deadline prioritize delivery routing
protocol is designed by Li et al. [30] and a dynamic load balanced routing protocol
on rank is proposed by Taghizadeh et al. [27]. Moreover, both approaches retransmit
packet to reduce packet drop rate which has no threshold during frequent messag-
ing which may cause overhead retransmit broadcast. Here, majority of research work
minimize delay, packet drop while some approach gives attention energy consump-
tion by designing green routing [26] and Energy-harvesting-aware routing [27]. The
proposed work of [26,27] based on “energy-store and energy-use” protocol by uti-
lizing the current available energy consumption of node and “excess-energy” for
future use. Although, the suitability metric in [26] avoids battery level which re-
sults to small lifetime and paper [27] ignores network heterogeneity. To overcome
the network heterogeneity a social-behavioral multihop optimal routing proposed for
device-to-device communication for dense IoMT environment in [31]. The proposed
work finds path based on rank according to distance and density for fixed base sta-
tion, while ignoring routing at outage period. The routing algorithms discussed here
attempt to improve the efficiency of the IoMT network, energy-saving, fault-tolerant
capability, optimal connection or link probability, etc. The energy-harvesting-aware
(EHA) routing protocol discussed the quality issues of the IoMT services and the
routing procedure. The other routing protocols specially designed for QoS are illus-
trated in the next subsequent section.](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-38-320.jpg)
![20 CHAPTER 1 A review on IoMT routing protocols and quality of service
1.3 Quality of Service (QoS) routing in IoMT
In IoMT, there are two types of applications in general – video-based applications,
which can tolerate some amount of packet-loss but requires timely delivery of a
packet, and other applications that are sensitive to packet loss due to their constrained
nature. The other category of applications relies upon retransmission to guarantee
message delivery. Both application categories require QoS to provide a better experi-
ence. The QoS protocols are discussed below in detail.
1.3.1 Traffic-aware QoS routing protocol
The traffic-aware QoS routing protocol [34] is designed to solve the packet delay
and packet loss issues. The protocol fulfills the QoS requirement of applications by
introducing a greedy-based approach based on Yen’s K-shortest paths algorithm for
computation of the optimal path. The routing protocol calculates the QoS path for all
the incoming flows in an SDN controller by taking all the conditions into account of
the SDN rule capacity constraint. The protocol considers the particular type of traffic
and the associated QoS requirement like packet loss and packet delay or both. The
routing protocol takes a graph and set of flows with their associated QoS requirement
as input. The user must prioritize delay-sensitive (ds) and loss-sensitive (ls) to each
flow. The protocol provides an output set of paths on which ds-flows and ls-flows can
be routed. The protocol performs the following operation:
• The protocol takes the cost of each link in terms of delay or packet loss as inputs
and returns a feasible routing for all the ds and ls flows.
• The ds and ls paths are considered with respect to assign priority with each appli-
cation.
• The ds and ls path are routed according to the path calculated, with the help of the
K-Shortest path algorithm.
• The last GET-QoS-Path function checks all the delay, loss, bandwidth, and check
rules if all are satisfied, then returns TRUE, else FALSE.
1.3.2 QoS-aware and Heterogeneously Clustered Routing (QHCR)
protocol
The QoS-aware and Heterogeneously Clustered Routing (QHCR) protocol [35]
is an energy-efficient routing protocol that provides the dedicated routes for the
bandwidth-hungry, delay-sensitive, real-time, and QoS-aware applications and saves
energy in the network. The QHCR protocol works in the following phases: Informa-
tion gathering phase, cluster head (CH) election phase, node association phase, and
intracluster communication. In the information-gathering phase, every node sends
and receives the broadcast messages to and from their neighbors to collect the energy
levels and other information. After exchanging the information from their neighbors,
each node maintains a neighbor table and updates the base station. The neighbor table
consists of information such as the number of neighbor nodes, initial energy levels](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-39-320.jpg)
![1.3 Quality of Service (QoS) routing in IoMT 21
of neighbor nodes, the distance of neighbor nodes from the base station, and dis-
tance between one neighbor node to another. This protocol uses CSMA/CD method
to minimize the collision in the network when two nodes broadcast at the same time.
QHCR is a centralized protocol in which the base station collects all information
from the nodes in the network and uses that information to elect the cluster head.
In the CH election phase, the cluster head is elected based on the cost value at each
energy level. This cost value is computed using the information stored in the node’s
neighbor table. A node having the lowest cost value is elected as the cluster head at
every energy level. The initial energy of nodes can be divided into four energy levels.
The nodes can be clustered based on these energy levels such that the nodes having
the same energy levels belong to the same cluster. In the node association phase, the
elected cluster head starts sending the broadcast messages to other nonCHs nodes
and member nodes of the clusters in the network. When a nonCHs node does not
get any broadcast message from any cluster head, they elect themselves as cluster
heads. During intracluster communication, when the distance between nonCH nodes
and CH or BS is very long, the nonCH nodes use intermediate nodes for forwarding
the messages to them or other nodes. Thus, the main goal of the QHCR protocol is
to select the best route for transmitting the messages with minimum delay by using
the intermediate nodes lie on multiple paths between the sending node and the cluster
head or base station.
1.3.3 QoS in wireless multimedia sensor network based IoMT
The IoMT applications use sensor nodes to gather data for subsequent processing.
The sensor nodes formulate an interconnected network to provide better coordination
for input data sharing. This network uses wireless links to exchange multimedia data,
popularly known as Wireless Multimedia Sensor Network (WMSN). The WMSN
is an integral part of IoMT networks. The objective of WMSN is to provide the
real-time delivery and standard QoS of multimedia data within constrained energy
requirements. The key QoS parameters are minimum packet drop, reduced latency,
in-order, and prioritized multimedia data delivery. Hamid and Hussain [36] presented
a detailed survey on layered and cross-layered approaches used to achieve QoS in
WSMN based IoMT. Here, the QoS requirements to deliver multimedia content are
explicitly examined on application, transport, routing, MAC, and physical layer. They
illustrate the use of less complex multimedia data coding schemes at the application
layer; Unequal Error Protection (UEP) method to provide reliability, priority-based
rate control methods to manage congestions, and optimized packet scheduling to tol-
erate the processing delay at the transport layer; geographic routing protocol [37] to
handle delay constrained multimedia data and multipath routing [38] to solve the load
balancing issue on aggregator sensor node at the routing layer; dynamic adjustment
of contention window (CW) to handle heterogeneous multimedia services, Multiple
Queuing methods to satisfy the scheduling requirements of different of multimedia
data flows and variable contention period to better channel access at the MAC layer;
dynamic control of beacon interval and use of Ultra-Wideband (UWB) technology to](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-40-320.jpg)
![22 CHAPTER 1 A review on IoMT routing protocols and quality of service
achieve reliable listening and a high rate of data transmission at the physical layer.
A special investigation on cross-layer approaches, i.e., optimized cooperation of dif-
ferent layers, to satisfy the QoS requirements of multimedia data is presented. The
cross-layer protocols in WMSN successfully achieve the real-time delay constraint,
rate adaptation, reliability, and optimized energy requirements.
1.3.4 SAMS framework for WMSN-based IoMT
With the increase of the Internet of Things (IoT) and real-time adjustment develop-
ment, people’s quality of life is improving. Multimedia wireless sensor nodes and
devices that make the Internet of Multimedia Things (IoMT) are essential for IoT
applications, which are diverse in nature. These constrained nodes and devices de-
pend on standardized communication stacks and effective routing protocols that form
the Wireless Multimedia Sensor Network (WMSN). Quick response services, traffic
control, smart cities, smart hospitals, smart agriculture, smart buildings, criminal in-
spection, security systems, Internet of bodies (IoB), and Industrial IoT (IIoT) are
examples of real-world multimedia technologies. IoMT devices need high protec-
tion, higher bandwidth, memory, and faster computing resources to process data.
This study suggests a model for a cluster-based hierarchical WMSN called Seam-
less and Authorized Streaming (SAMS) [39]. There are two phases for building
secure clusters of a SAMS: set-up and steady-state. During the set-up phase, two-
level authentication (i.e., between BS and elected CHs; and formation of clusters) is
performed, and during the steady-state phase, seamless data transmission is achieved.
Only legal nodes may transmit captured data to their Cluster Heads (CHs) after these
clusters have been established. During mutual authentication, each multimedia node
in the cluster regularly senses the environment and stores any data received in a buffer
with a predefined limit.
Each multimedia node waits for its turn to transmit the capture data using its allo-
cated timeslot/channel that are prescribed by its respective cluster head. This waiting
can cause excessive packet loss and end-to-end delay in wireless channels. A novel
channel allocation technique is used to keep the WMSNs’ Quality of Service (QoS)
at an acceptable scale to resolve this problem.
A member node in one cluster moves to a nearby CH if the latter has an available
channel for allocation in the event of a buffer overflow. Fig. 1.8 which is taken from
[39], depicts the SAMS framework’s network architecture. The timely and accurate
distribution of data is a key characteristic of IoMT. As a result, the SAMS outper-
forms existing schemes in terms of different QoS metrics like average packet loss
and average end-to-end delay while also providing robust protection against various
attacks (such as tampering, sinkhole, insider, Sybil, and password guessing).
Discussion. The research work reviewed in this section are QoS oriented protocols,
which improve the seamless real-time servicing for IoMT. The comparative analysis
of QoS protocols for IoMT is summarized in Table 1.4 and 1.5. The quality of service
routing protocol in [34] focuses on throughput, jitter and reliability for two types of
applications such as delay and loss sensitive using greedy heuristic based k-shortest](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-41-320.jpg)
![1.3
Quality
of
Service
(QoS)
routing
in
IoMT
23
Table 1.4 Comparative analysis of QoS protocols for IoMT (Part a).
Paper Architec-
ture
Entities Communica-
tion
mode
Link
establishment
criteria
Suitability
criteria
Optimal path
finding
Noiseless
path
selection
Environ-
ment
adaption
Goal Methodology
[34] General IoT sensors,
traffic lights
Device-to-
device
Delay
duration,
packet-loss
probability,
current
bandwidth
Delay sensitive
cost function and
loss sensitive
cost function
K-shortest path x X Author motivation
to design a
network routing
protocol which
focuses QoS
(throughput, jitter,
reliability etc.) for
delay and loss
sensitive
application.
Proposed
approach utilize a
programmatic
SDN and routes
delay and loss
sensitive data
using greedy
heuristic based
k-shortest path
algorithm.
[35] Cluster WSN,
sensor node
Device to
device
Energy
consumption
at sensing,
processing
and radio
communica-
tion, number
of bits
transmitted,
distance
Initial and current
energy cost
function.
– x X The aim is to
design a
QoS-aware
routing protocol
to deliver data for
real-time and
nonrealtime
multimedia
service with
balanced load
and fault
tolerable.
Proposed a
cluster based
energy efficient
routing according
to delay
sensitivity,
bandwidth, and
energy
fluctuation. CH is
selected
according to path
metric (initial
energy).
[37] Dis-
tributed
Image
Sensor
nodes
Device-to-
device
Distance,
queue size,
angular
bearing
towards
destination
Relative location
based cost
function
– X Aims to propose
a best effort QoS
support, low
delay with
avoidable in-node
routing table and
information
related to
outdated states
of sensor.
Designed a
routing algorithm
for event driven
real-time
geographical
distributed
environment
based on greedy
approach.
(continued on next page)](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-42-320.jpg)
![24
CHAPTER
1
A
review
on
IoMT
routing
protocols
and
quality
of
service
Table 1.4 (continued)
Paper Architec-
ture
Entities Communica-
tion
mode
Link
establishment
criteria
Suitability
criteria
Optimal path
finding
Noiseless
path
selection
Environ-
ment
adaption
Goal Methodology
[38] General Sensors Device-to-
device
Size of real
time and
nonrealtime
traffic.
Probability of
successful data
delivery
K-path finding x x Aims to maximize
network lifetime
through energy
balancing across
node according
to acceptable
delay.
Propose a QoS
and energy aware
routing technique
which balance
fault tolerance
using XOR based
forward error
correction
method.
[39] Cluster BS, sensors X to Y where
X,Y ε {BS,
sensors}
Payload,
distance
buffer overflow – x x Goal is to
propose secure
intercluster
communication
framework to
reduce packet
loss due
excessive waiting
for its turn to
send captured
data.
Proposed a
lightweight
secure AES
based streaming
for seamless
delivery. After
successful
authentication
Cluster formation
initiated
according to time
slot. Here, CM
moves according
to buffer overflow.](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-43-320.jpg)
![1.3
Quality
of
Service
(QoS)
routing
in
IoMT
25
Table 1.5 Comparative analysis of QoS protocols for IoMT (Part b).
Paper Merits Demerits Evaluation metrics Simulation details Assumption Designed
approach
Comparison Tool
[34] The proposed approach
improves QoS Violated
flows around 15%, 14%,
and 13% (using AttMpls)
and 39%, 38%, and
37% (using Goodnet)
wrt MRC, SPD, and
LARP respectively.
Forwarding without fine
grained QoS
End-to-end delay,
runtime, QoS
violated flow,
Active Node
Topology (AttMpls,
Goodnet),
bandwidth
(0.20–0.40 kbps),
packet size
(94–699 bytes)
Assume constant
jitter
SWAY Minimum Rule
Capacity (MRC),
Shortest path
delay ( SPD),
LAgrangian
Relaxation
Aggregated
cost(LARAC)
−
[35] Lifetime ends for
proposed algorithm at
2750 rounds while
ECHERP is 1300,
PASCCC is 2000, and
CCWM is 2100 rounds.
Due to no energy
harvesting feature the
proposed work
conserve no energy
from renewable source.
Lifetime, stability
period,
throughput,
average energy
consumption,
End-to-end delay
Sensors (100),
buffer size (256
k-bytes), Transmit
power (20 mW),
receiving power
(15 mW)
Fixed sensing
nodes, same
packet size,
QHCR CCWM, PASCCC,
ECHERP
MATLAB
[37] The proposed scheme
increases the life-time
from 29% to 43%
compared to angle
based approach
In-accurate location
information leads to
retransmission and
packet loss.
Number of alive
node, delay
Radio range
(75 m),
Nodes(196),
payload size (1),
energy (200–400
units)
Payload size is
fixed
Hybrid
scheme of
angle and
distance
Distance based
scheme, angle
based scheme
−
[38] Due to use of forward
error correction
mechanism the
proposed routing
outperforms than
MCMP to retrieve
original message.
MCMP protocol
outperform better than
proposed model with
nonreal-data traffic
because of low
overhead by queuing.
Energy
consumption,
delivery ratio,
average delay
Sensor node
(300),
transmission
range (25 m),
buffer size (256
K-bytes), transmit
and receive power
(13, 12 mW)
Fully connected
environment,
sensors must be
failure free
MCMP NS2
[39] Secure against replay,
DoS, Insider, Sybil,
Eavesdrop
Every time at cluster
formation CH
authenticates with BS
Overhead, energy
cost, resilience,
cluster size,
end-to-end delay
Buffer size (60),
AES-128
– SAMS Das approach,
Amin-biswas
approach
−](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-44-320.jpg)
![26 CHAPTER 1 A review on IoMT routing protocols and quality of service
FIGURE 1.8
Wireless network model.
path algorithm to overcome the bandwidth consumption while ignoring storage of
outdated states of sensors which is resolved by Savidge et al. in [37]. Here, the authors
presented a purely location-based QoS routing which can become as a faulty network
due to sharing of in-accurate location information by attacker node. To overcome the
above issue, an AES-128 authentication based secure and cluster-based approach is
designed in [39]. Although, here is an issue of frequent authentication at the time
of fault-tolerance between CH to BS can leads to overhead at fault-tolerance. Some
other cluster based approach has been designed with fault tolerance by selecting CH
and CM according to available bandwidth, delay as well as energy fluctuation for
real-time and nonrealtime IoT application with fixed payload size [35]. Moreover, the
approach is limited due to no energy consumption from renewable resource. In [38],
Othman and Yahya gave an energy-aware protocol based on balanced fault tolerance
and forward error correction method which shows significant overhead in queuing
process.
1.4 Conclusion and future directions
The Internet of Multimedia Thing (IoMT) has shown its applications in every cor-
ner of today’s day-to-day life. The multimedia data sharing among the users through
the IoMT network improves the user experiences at maximum height. However, the
challenges in IoMT like real-time data delivery, throughput, device energy constraint,
cross-layer issues, etc., still need to be handled in efficient manner. This chapter dis-
cusses the routing and QoS protocols to tackle the challenge of the IoMT network.
The design of routing and QoS protocols primarily concentrated on the low energy](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-45-320.jpg)
![References 27
consumption of the devices, minimum communication cost over the links, harvesting
energy to reuse, and providing real-time experience to the user. The in-depth dis-
cussion on each of the protocols is presented lucidly. Further, in future, we need to
collectively focus on Quality of Experience (QoE) parameters like inter disruption
time between subsequent packet deliveries, packet buffering interval on intermediate
nodes, and user satisfaction feedback with discussed QoS parameters such as packet
loss, end-to-end delay, error rate to evaluate the performance of IoMT applications.
An improved cross-layer model can be designed in view of the QoE as mentioned
above and QoS parameters. The smart routing protocol based on machine learning
and optimization techniques to discover the energy-efficient route for bandwidth-
hungry IoMT application need to be further explored. The storage requirements for
Big Multimedia Data (BMD) open the door to reinvestigate the energy harvesting-
based routing and link layer protocol and strengthen the provision of green routing
protocol for IoMT networks. The Software Defined Network (SDN) architecture can
be incorporated to improve the routing capabilities by decoupling network control
from forwarding decisions in IoMT applications. Moreover, some use cases can be
taken for different types of IoMT applications to exemplify identified challenges.
References
[1] L. Atzori, A. Iera, G. Morabito, The internet of things: a survey, Computer Networks
54 (15) (2010) 2787–2805.
[2] S. Greengard, The Internet of Things, MIT Press, 2021.
[3] https://www.statista.com/statistics/1101442/iot-number-of-connected-devices-
worldwide/. (Accessed 3 October 2021).
[4] F. Samie, L. Bauer, J. Henkel, Hierarchical classification for constrained IoT devices:
a case study on human activity recognition, IEEE Internet of Things Journal 7 (9) (2020)
8287–8295.
[5] Y.B. Zikria, M.K. Afzal, F. Ishmanov, S.W. Kim, H. Yu, A survey on routing protocols
supported by the Contiki Internet of things operating system, Future Generation Com-
puter Systems 82 (2018) 200–219.
[6] F. Javed, M.K. Afzal, M. Sharif, B.S. Kim, Internet of Things (IoT) operating systems
support, networking technologies, applications, and challenges: a comparative review,
IEEE Communications Surveys and Tutorials 20 (3) (2018) 2062–2100.
[7] A.K. Maurya, D. Singh, IoT enabling platforms, in: Internet of Things, An Indian Adap-
tation: Concepts and Applications, Wiley, 2021, pp. 93–122.
[8] S. Li, L. Da Xu, S. Zhao, 5G Internet of Things: a survey, Journal of Industrial Informa-
tion Integration 10 (2018) 1–9.
[9] G.A. Akpakwu, B.J. Silva, G.P. Hancke, A.M. Abu-Mahfouz, A survey on 5G networks
for the Internet of Things: communication technologies and challenges, IEEE Access 6
(2017) 3619–3647.
[10] T. Vaiyapuri, V.S. Parvathy, V. Manikandan, N. Krishnaraj, D. Gupta, K. Shankar,
A novel hybrid optimization for cluster-based routing protocol in information-centric
wireless sensor networks for IoT based mobile edge computing, Wireless Personal Com-
munications (2021) 1–24.](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-46-320.jpg)
![28 CHAPTER 1 A review on IoMT routing protocols and quality of service
[11] A.K. Maurya, D. Singh, Median predictor based data compression algorithm for wireless
sensor network, International Journal of Computer Applications 975 (2011) 8887.
[12] W. Sun, Z. Wang, G. Zhang, A QoS-guaranteed intelligent routing mechanism in
software-defined networks, Computer Networks 185 (2021) 107709.1.
[13] V.K. Quy, V.H. Nam, D.M. Linh, N.T. Ban, N.D. Han, A survey of QoS-aware routing
protocols for the MANET-WSN convergence scenarios in IoT networks, Wireless Per-
sonal Communications (2021) 1–14.
[14] A.K. Maurya, D. Singh, A. Kumar, R. Maurya, Random waypoint mobility model based
performance estimation of On-Demand routing protocols in MANET for CBR applica-
tions, in: 2014 International Conference on Computing for Sustainable Global Develop-
ment (INDIACom), IEEE, 2014, pp. 835–839.
[15] D. Singh, A.K. Maurya, A.K. Sarje, Comparative performance analysis of LANMAR,
LAR1, DYMO and ZRP routing protocols in MANET using Random Waypoint Mobil-
ity Model, in: 2011 3rd International Conference on Electronics Computer Technology,
IEEE, 2011, pp. 62–66.
[16] A.K. Maurya, D. Singh, A. Kumar, Performance comparison of DSR, OLSR and FSR
routing protocols in MANET using random waypoint mobility model, International Jour-
nal of Information and Electronics Engineering 3 (5) (2013) 440–443.
[17] A. Katiyar, D. Singh, R.S. Yadav, State-of-the-art approach to clustering protocols in
vanet: a survey, Wireless Networks 26 (7) (2020) 5307–5336.
[18] G. Sun, Y. Zhang, H. Yu, X. Du, M. Guizani, Intersection fog-based distributed rout-
ing for V2V communication in urban vehicular ad hoc networks, IEEE Transactions on
Intelligent Transportation Systems 21 (6) (2019) 2409–2426.
[19] G. Premsankar, M. Di Francesco, T. Taleb, Edge computing for the Internet of Things:
a case study, IEEE Internet of Things Journal 5 (2) (2018) 1275–1284.
[20] D. Kumar, A.K. Maurya, G. Baranwal, IoT services in healthcare industry with fog/edge
and cloud computing, in: IoT-Based Data Analytics for the Healthcare Industry, Elsevier,
Amsterdam, The Netherlands, 2021, pp. 81–103.
[21] A.K. Maurya, A.K. Tripathi, Deadline-constrained algorithms for scheduling of bag-
of-tasks and workflows in cloud computing environments, in: Proceedings of the 2nd
International Conference on High Performance Compilation, Computing and Communi-
cations, 2018, pp. 6–10.
[22] Y.B. Zikria, M.K. Afzal, S.W. Kim, Internet of Multimedia Things (IoMT): opportunities,
challenges and solutions, Sensors (Basel) 20 (8) (2020) 2334.
[23] A. Nauman, Y.A. Qadri, M. Amjad, Y.B. Zikria, M.K. Afzal, S.W. Kim, Multimedia
Internet of Things: a comprehensive survey, IEEE Access 8 (2020) 8202–8250.
[24] Quality of Service Technical White Paper, Microsoft, Redmond, WA, USA, 1999.
[25] T. Winter, P. Thubert, A. Brandt, J.W. Hui, R. Kelsey, P. Levis, K. Pister, R. Struik, J.
Vasseur, R.K. Alexander, RPL: IPv6 Routing Protocol for Low-Power and Lossy Net-
works, rfc, 6550, 1–157, 2012.
[26] S.A. Alvi, G.A. Shah, W. Mahmood, Energy efficient green routing protocol for inter-
net of multimedia things, in: 2015 IEEE Tenth International Conference on Intelligent
Sensors, Sensor Networks and Information Processing (ISSNIP), IEEE, 2015, pp. 1–6.
[27] S. Taghizadeh, H. Bobarshad, H. Elbiaze, CLRPL: context-aware and load balancing
RPL for IoT networks under heavy and highly dynamic load, IEEE Access 6 (2018)
23277–23291.
[28] H. Bouzebiba, M. Lehsaini, FreeBW-RPL: a new RPL protocol objective function for in-
ternet of multimedia things, Wireless Personal Communications 112 (2020) 1003–1023.](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-47-320.jpg)
![References 29
[29] J.W. Lin, P.R. Chelliah, M.C. Hsu, J.X. Hou, Efficient fault-tolerant routing in IoT wire-
less sensor networks based on bipartite-flow graph modeling, IEEE Access 7 (2019)
14022–14034.
[30] T. Li, S. Tian, A. Liu, H. Liu, T. Pei, DDSV: optimizing delay and delivery ratio for mul-
timedia big data collection in mobile sensing vehicles, IEEE Internet of Things Journal
5 (5) (2018) 3474–3486.
[31] G. Chen, J. Tang, J.P. Coon, Optimal routing for multihop social-based D2D communica-
tions in the Internet of Things, IEEE Internet of Things Journal 5 (3) (2018) 1880–1889.
[32] T.D. Nguyen, J.Y. Khan, D.T. Ngo, A distributed energy-harvesting-aware routing al-
gorithm for heterogeneous IoT networks, IEEE Transactions on Green Communications
and Networking 2 (4) (2018) 1115–1127.
[33] F.M. Al-Turjman, H.S. Hassanein, M.A. Ibnkahla, Efficient deployment of wireless sen-
sor networks targeting environment monitoring applications, Computer Communications
36 (2) (2013) 135–148.
[34] N. Saha, S. Bera, S. Misra, Sway: traffic-aware QoS routing in software-defined IoT,
IEEE Transactions on Emerging Topics in Computing (2018).
[35] M. Amjad, M.K. Afzal, T. Umer, B.S. Kim, QoS-aware and heterogeneously clustered
routing protocol for wireless sensor networks, IEEE Access 5 (2017) 10250–10262.
[36] Z. Hamid, F.B. Hussain, QoS in wireless multimedia sensor networks: a layered and
cross-layered approach, Wireless Personal Communications 75 (1) (2014) 729–757.
[37] L. Savidge, H. Lee, H. Aghajan, A. Goldsmith, QoS-based geographic routing for event-
driven image sensor networks, in: 2nd International Conference on Broadband Networks,
IEEE, 2005, pp. 991–1000.
[38] J. Ben-Othman, B. Yahya, Energy efficient and QoS based routing protocol for wireless
sensor networks, Journal of Parallel and Distributed Computing 70 (8) (2010) 849–857.
[39] M.A. Jan, M. Usman, X. He, A.U. Rehman, SAMS: a seamless and authorized multime-
dia streaming framework for WMSN-based IoMT, IEEE Internet of Things Journal 6 (2)
(2018) 1576–1583.](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-48-320.jpg)
![2
CHAPTER
Energy efficient data
communication in
Internet of Multimedia
Things (IoMT)
Shailendra Shuklaa, Asheesh Kumar Mani Tripathib, and Amit Kumar Singhc
aComputer Science and Engineering Department, MNNIT Allahabad, Allahabad, India
bInformation and Cyber Security Services, HCL, Noida, India
cComputer Science and Engineering Department, NIT Patna, Patna, India
2.1 Introduction
In the last couple of years, the advancement of low power and low-cost small-scaled
devices such as Micro Electro Mechanical Systems (MEMS) [1] has given the explo-
sive growth in the number of IoT devices [2] and increased scalability. IoT is useful
for the application in home automation, security, automated applications to provide
useful contextual information frequently to help with their works, and decision mak-
ing. Internet-enabled sensors and actuators’ operations can be rapid, efficient, and
more economical. Multimedia data in IoT [3] is bulky and specially meant for the
real-time application requires real-time communication. Few examples are real-time
multimedia-based surveillance systems [4] for smart home, office, at critical in-
frastructure, telemedicine service in the smart hospital, remote patient monitoring,
transportation management system, remote multimedia based monitoring, etc. Mul-
timedia services require higher processing and memory resources, which raises the
need for different modus operandi for multimedia-based IoT.
The IoT devices are mainly equipped with a mic and camera. The major task of
equipped devices is to sense the multimedia content and send it to sink nodes. For
accurate information to be communicated, knowledge of the exact location of the
sensor nodes in WSN [5] is mandatory. However, this location knowledge is possible
with the integration of GPS devices to the sensor devices, but an increase in cost and
reduction in battery life of sensor nodes are some of the demerits of this approach.
So an alternate approach for this is to detect the most important nodes [6] in the
networks.
Critical node detection aka boundary node (B-N) [6] actually means to find the
nodes, which reside in those positions of the network, which if removed then causes
the partition in the networks. Hence, with the detection of boundary nodes, we can
achieve connectivity and coverage of the network at a minimal cost. Till now, there
Internet of Multimedia Things (IoMT). https://doi.org/10.1016/B978-0-32-385845-8.00007-1
Copyright © 2022 Elsevier Inc. All rights reserved.
31](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-49-320.jpg)
![32 CHAPTER 2 Energy efficient data communication in IoMT
are various methods have been proposed which focus on the detection of boundary
nodes in the sensor networks. But to the best of our knowledge, no method provides
the virtual coordinates [7] simultaneously to the detected boundary nodes. So in this
chapter, we propose an algorithm, which not only detects the boundary nodes but
also provides the virtual coordinates to them, which later contributes to providing
the virtual coordinates to all the remaining nonboundary nodes in the wireless sensor
networks. In IoMT the nodes are required to be capable of highly resourceful which
is not possible as IoT devices are low power and fewer storage devices. Hence the
detection of boundary nodes and virtual coordinate optimize the resource require-
ments.
The main contribution of this chapter is to develop a distributed algorithm to
detect the boundary nodes and then virtual coordinate in IoT networks using the cen-
tral node (ς). The proposed algorithm is free from any hardware dependability and
flooding-based communications, it simply uses the RSSI value and cosine rules of
trigonometry and. Finally, the proposed algorithm is deployed over the RPL proto-
col. Remaining chapter has been organized as follows: Section 2.2 covers the related
work. Section 2.3 deals with system model. Our proposed method is described in
section 2.4. Section 2.5 deals with the results and comparison analysis with other
proposed approach. And, finally we conclude our chapter in section 2.6.
2.2 Related work
This section of the chapter is divided into subparts in the first part presents the re-
ported work on IoMT content dissemination protocol and the second part presents
the various boundary detection and virtual coordinate algorithms.
2.2.1 Routing protocol for IoMT
Reported work on WSN in multimedia IoT (IoMT) shows various protocols which
support the low power consumption and heterogeneity in IoT [8–11]. The WSN
deployed for IoMT uses CoRE protocol rather than HTTP for the Internet and a
lightweight OMA-DM to support low power and low resource IoMT devices. RPL
(Routing Protocol for Low-Power and Lossy Networks) [12] is a routing protocol for
resource-constrained devices that uses Destination Oriented Directed Acyclic Graph
(DODAG) to maintain network topology. This DAG comprises multihop routes from
leaf nodes to the root node. RPL optimizes an objective function and chooses the best
path by selecting the desired predecessor nodes starting from leaf nodes. However,
they are not developed for large multimedia data transmission.
The Green-RPL [13] is an energy-efficient routing protocol for IoMT. It is an
enhanced version of the RPL protocol [12]. The previous RPL protocol implemen-
tations are not feasible for IoMT and do not consider multimedia data. In contrast,
the Green-RPL protocol considers the data generated from multimedia devices. The
Green-RPL protocol reduces energy consumption and carbon footprint emissions](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-50-320.jpg)
![2.2 Related work 33
together with QoS requirements of applications. To guarantee QoS for a particular
multimedia application, it determines the delay bound for the application. For exam-
ple, in VoIP applications, the delay limitation usually is 120 msec. To ensure energy
efficiency, the protocol considers the features of all the intermediary links between
the leaf nodes and the root node and estimates the energy consumption by the cho-
sen immediate predecessor node to support traffic needs for one more immediate
successor node. An optimization model is given for this protocol based on various
requirements and constraints. In Green-RPL routing protocol, the desired immedi-
ate predecessor node is selected according to the objective function minimizing the
emissions of cumulative path carbon footprints on all the links from the immediate
predecessor node to the root node, and satisfying the constraints such as cumulative
path link energy, cumulative path delay, idle time, and battery status of the imme-
diate predecessor node. When more than one immediate predecessor node satisfies
the given constraints, then the node that provides the most greener path is chosen as
the desired immediate predecessor node. In cumulative path link energy constraint,
energy consumption by a node is calculated based on the quality of links in the se-
lected route. The cumulative path delay constraint specifies that the predefined delay
threshold should not be increased. According to the idle time constraint, an imme-
diate predecessor node should be chosen as the desired node only if its idle time is
sufficient to support another child node. In the last constraint battery status, a child
node should choose an immediate predecessor node as the desired one if the bat-
tery level of its energy resources is greater than the predefined threshold. A modified
version of RPL, known as Context-aware and load balancing RPL under heavy and
highly dynamic load for IoT and IoMT applications is discussed in [14]. It addresses
the power depletion and packet loss problems associated with RPL. Paper [15] pro-
poses an efficient protocol CoUDP for real-time IoT multimedia communications.
Paper [16] proposes a four-layer approach physical devices layer, network layer, com-
bination layer, and context layer for IoMT data communications. Caching policy for
popularity-aware video caching in the topology-aware content-centric network is pro-
posed in the [17]. The paper [17] uses scalable video coding for fast video delivery
and cached each video layer. Paper [18] proposed an approach to achieve reliability
and reduce delay in IoMT networks. To enhance the QoS metrics, and efficient data
gathering scheme is designed in [19]. Cross layer protocols for IoMT are proposed in
[20] and [21]. The major limitation of reported work is most of them do not consider
the energy efficiency and fault-tolerant approach for critical IoMT applications.
2.2.2 Boundary detection and virtual coordinate algorithms
According to [22], Boundary detection algorithms can be classified into 3 categories,
geometrical approach, statistical approach, and topological approach. Paper [23] is
considered as a geometrical approach as it is based on the exact geometric location
identification using global positioning devices such as GPS. Using the location infor-
mation, sensor nodes identify the voids to create an alternative path for data traffic.
The author has used the term stuck node for the nodes which are closer to the desti-
nation than its one-hop neighbor. Paper [24] is based on some graph techniques and](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-51-320.jpg)
![34 CHAPTER 2 Energy efficient data communication in IoMT
computational geometry, to find out the closed polygonal boundaries of the voids. It
is also based on the assumption that every node must be able to obtain its location.
The main drawback with the geometrical approach is the requirement of an external
hardware device to find out the geographical position, which is not only costly but
also doesn’t work indoor and in bad weather conditions.
The statistical approach is based on some complex mathematical and statistical
functions and categorizes the nodes into the interior and boundary nodes. Paper [25]
uses the restricted stress centrality measure which is a subclass of centrality indices
and makes a comparison of it with a defined threshold value that is whether it is
above or below the threshold value, based on which it categorizes the nodes into the
above-mentioned classes. The main drawback of this approach is its impractical node
distribution and high-density requirement. Paper [26] utilized the fact that the degree
of nodes that are closer to the boundary of the network will be less in comparison to
the node interior nodes. No requirement of global positioning devices is one of the
major advantages of the statistical approaches, but these methods work only with the
denser regions. Moreover, these methods require a node degree of more than 100 or
higher.
The third category is the topological approaches which use only the topological
information available to sensor nodes in the network. Paper [27] describes a two-step
algorithm boundary recognition and topology extraction. A combinatorial structure
called flower and augmented cycles are searched in this approach which is the base of
this algorithm, but it is not always the case where one can find a flower-like structure
in a randomly deployed network which is the major drawback of this approach. Paper
[28] builds a shortest path tree and finds out cut nodes by flooding the network. Using
the cut nodes, it finds out the network boundary. Paper [29] and paper [30] use the
iso-contours to recognize the network boundary. Iso-contours are build using hop-
counts from a root node and the end nodes of these iso-contours are treated as the
boundary nodes. The drawback with this approach is its high-density requirement and
flooding of more than 3 different nodes. Paper [31] describes a 3 step approach for
the detection of boundary nodes. In the first step, which is the information collection
step, it builds an x-hop neighbor list. 2nd step is the path construction, in which the
link between the x-hop neighbors is identified by building the first set and it uses
a recursive procedure to construct the path and finally, the last set is made which
contains the node which cannot be extended further. 3rd step is the path checking
phase which categorizes the node as an interior or boundary node according to the
intersection of neighbors of the first set and last set. According to my understanding,
a lot of computation is done over a single node recursively for the path construction,
which is the only drawback of this approach.
2.3 System model
An IoT network is assumed where sensor nodes are homogeneously spread into a
region L in such a way that it forms a connected graph G(V, E). In a connected graph](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-52-320.jpg)
![2.4 Proposed approach 35
G(V, E), V represents the set of nodes and E represents the link between the nodes.
The Nbr(v) represents the neighbor set of a node v (v ∈ V ). Each node is equipped
with a small range camera and has a communication range ‘R’, also each node can
determine the distance to its direct one hope neighbors using ranging techniques like
RSSI, TOA, or TDOA [33] also they are not aware of their location. In consonance
with Internet-Connected multimedia, we envisage our network broadly consists of
the following components: Central Node (ς): A node ς ∈ V is said to be a cen-
tral node if its graph centrality is maximum [34]. Boundary Node (B-N): A node
B − N ∈ V is said to be a Boundary node if it lies on the critical position of the
network graph G. Network Boundary: Network boundary is the imaginary circle
formed by the Boundary nodes after the first phase of the algorithm. Interior Node
(I − N): A node I − N ∈ V is said to be an interior node if it lies inside the network
boundary and not a Boundary nodes (B-N).
2.3.1 Problem description
The problem can be formulated as follows: To find an IoT node from a set
Min(B − N) ∈ V such that it lies at the most critical places of the networks, detec-
tion of a central node (ς) in the given graph G(V, E) which is at a minimum distance
(min-dis) from all the nodes of networks. Our second problem is to find the nodes
which are at the maximum distance from the central node and provide them some
virtual coordinates so that a virtual overlay coordinate is a form for the data commu-
nications. The main contribution of this chapter is to develop a distributed algorithm
to detect the virtual coordinate which is based on graph centrality (ς). Developed an
algorithm to find the network boundary in IoT networks using the central node and
simple cosine rules of trigonometry and finally, deploying the algorithms over the
RPL protocol.
2.4 Proposed approach
2.4.1 Working of algorithm
The idea behind the proposed approach is to detect the boundary nodes and then to
assign virtual coordinates to these boundary nodes (B-N) using the graph central-
ity. To assign virtual coordinates it uses RSSI (Received signal strength indicator),
Pr(RI(Nv[i])), to detect the distance between target nodes (mobile nodes present
in the network). Since the boundary nodes are detected hence the anchor nodes [35]
are detected from boundary nodes. Finally, trilateration is used to detect the virtual
coordinate and critical nodes.
We have divided our proposed approach into 2 phases. In phase-I, the bound-
ary nodes detection algorithm also acts as anchor nodes. In phases II-a and II-b, the
distance estimation is used to detect the center node, and then a virtual coordinate
algorithm is proposed.](https://image.slidesharecdn.com/23055026-250513033626-de847318/85/Internet-Of-Multimedia-Things-Iomt-Techniques-And-Applications-Shailendra-Shukla-53-320.jpg)
Internet Of Multimedia Things Iomt Techniques And Applications Shailendra Shukla
- 1. Internet Of Multimedia Things Iomt Techniques And Applications Shailendra Shukla download https://ebookbell.com/product/internet-of-multimedia-things-iomt- techniques-and-applications-shailendra-shukla-46110052 Explore and download more ebooks at ebookbell.com
- 2. Here are some recommended products that we believe you will be interested in. You can click the link to download. Internet Of Multimedia Things Iomt Techniques And Applications Shailendra Shuklaamit Kumar Singhgautam Srivastava Amit Kumar Singh Gautam Srivastava Fatos Xhafa https://ebookbell.com/product/internet-of-multimedia-things-iomt- techniques-and-applications-shailendra-shuklaamit-kumar-singhgautam- srivastava-amit-kumar-singh-gautam-srivastava-fatos-xhafa-46704108 Multimedia Technologies In The Internet Of Things Environment Volume 3 Raghvendra Kumar https://ebookbell.com/product/multimedia-technologies-in-the-internet- of-things-environment-volume-3-raghvendra-kumar-47173296 Multimedia Technologies In The Internet Of Things Environment Volume 2 Raghvendra Kumar https://ebookbell.com/product/multimedia-technologies-in-the-internet- of-things-environment-volume-2-raghvendra-kumar-47173298 Multimedia Technologies In The Internet Of Things Environment 1st Ed Raghvendra Kumar https://ebookbell.com/product/multimedia-technologies-in-the-internet- of-things-environment-1st-ed-raghvendra-kumar-22473788
- 3. Management Of Multimedia On The Internet 5th Ifip Ieee International Conference On Management Of Multimedia Networks And Services Mmns 2002 Santa Barbara Ca Usa October 69 2002 Proceedings 1st Edition Kevin C Almeroth https://ebookbell.com/product/management-of-multimedia-on-the- internet-5th-ifip-ieee-international-conference-on-management-of- multimedia-networks-and-services-mmns-2002-santa-barbara-ca-usa- october-69-2002-proceedings-1st-edition-kevin-c-almeroth-2444442 Management Of Multimedia On The Internet 4th Ifipieee International Conference On Management Of Multimedia Networks And Services Mmns 2001 Chicago Il Usa October 29 November 1 2001 Proceedings 1st Edition Nagasuresh Seelam https://ebookbell.com/product/management-of-multimedia-on-the- internet-4th-ifipieee-international-conference-on-management-of- multimedia-networks-and-services-mmns-2001-chicago-il-usa- october-29-november-1-2001-proceedings-1st-edition-nagasuresh- seelam-4605232 Quality Of Service For Internet Multimedia Jitae Shin Daniel C Lee https://ebookbell.com/product/quality-of-service-for-internet- multimedia-jitae-shin-daniel-c-lee-982478 Designing And Evaluating Emanagement Decision Tools The Integration Of Decision And Negotiation Models Into Internetmultimedia Technologies 1st Edition Giampiero E G Beroggi Auth https://ebookbell.com/product/designing-and-evaluating-emanagement- decision-tools-the-integration-of-decision-and-negotiation-models- into-internetmultimedia-technologies-1st-edition-giampiero-e-g- beroggi-auth-4238936 Internet Of Things For Smart Environments Gonalo Marques Alfonso Gonzlezbriones https://ebookbell.com/product/internet-of-things-for-smart- environments-gonalo-marques-alfonso-gonzlezbriones-46180932
- 6. Internet of Multimedia Things (IoMT) Techniques and Applications
- 8. Internet of Multimedia Things (IoMT) Techniques and Applications Edited by Shailendra Shukla MNNIT Allahabad, Allahabad, India Amit Kumar Singh NIT Patna, Patna, India Gautam Srivastava Brandon University, Brandon, MB, Canada Series Editor Fatos Xhafa Universitat Politècnica de Catalunya, Spain
- 9. Academic Press is an imprint of Elsevier 125 London Wall, London EC2Y 5AS, United Kingdom 525 B Street, Suite 1650, San Diego, CA 92101, United States 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom Copyright © 2022 Elsevier Inc. All rights reserved. MATLAB® is a trademark of The MathWorks, Inc. and is used with permission. The MathWorks does not warrant the accuracy of the text or exercises in this book. This book’s use or discussion of MATLAB® software or related products does not constitute endorsement or sponsorship by The MathWorks of a particular pedagogical approach or particular use of the MATLAB® software. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. ISBN: 978-0-323-85845-8 For information on all Academic Press publications visit our website at https://www.elsevier.com/books-and-journals Publisher: Mara Conner Editorial Project Manager: Ivy Dawn Torre Production Project Manager: Sojan P. Pazhayattil Designer: Vicky Pearson Esser Typeset by VTeX
- 10. Contributors Chady Abou Jaoude Antonine University, Faculty of Engineering – TICKET Lab, Beirut, Lebanon Gaurav Bhatnagar Department of Mathematics, Indian Institute of Technology Jodhpur, India Zahi Al Chami Antonine University, Faculty of Engineering – TICKET Lab, Beirut, Lebanon Universite de Pau et des Pays de l’Adour, E2S UPPA, LIUPPA, Anglet, France Chiranjoy Chattopadhyay Department of Computer Science and Engineering, Indian Institute of Technology Jodhpur, India Richard Chbeir Universite de Pau et des Pays de l’Adour, E2S UPPA, LIUPPA, Anglet, France Rupesh Kumar Dewang Department of Computer Science & Engineering, Motilal Nehru National Institute of Technology Allahabad, Prayagraj, India Vivek K. Dwivedi Department of Electronics and Communication Engineering, Jaypee Institute of Information Technology, Noida, India Shreya Goyal Department of Computer Science and Engineering, Indian Institute of Technology Jodhpur, India Puneet Kumar Jain Department of Computer Science and Engineering, National Institute of Technology Rourkela, Odisha, India Bailey Janeczko Department of Math and Computer Science, Brandon University, Brandon, MB, Canada Niharika Keshari Department of Computer Science & Engineering, Motilal Nehru National Institute of Technology Allahabad, Prayagraj, India xiii
- 11. xiv Contributors Prateek Ishwar Khade Department of Computer Science & Information Systems, Birla Institute of Technology and Science, Pilani, Jhunjhunu, Rajasthan, India Alok Kumar Department of Electronics and Communication Engineering, Jaypee University of Information Technology, Solan, India Om Prakash Mahela Power System Planning Division, Rajasthan Rajya Vidyut Prasaran Nigam Ltd., Jaipur, India Ashish Kumar Maurya Department of Computer Science & Engineering, Motilal Nehru National Institute of Technology Allahabad, Prayagraj, India Shweta Pandit Department of Electronics and Communication Engineering, Jaypee University of Information Technology, Solan, India Amitesh Singh Rajput Department of Computer Science & Information Systems, Birla Institute of Technology and Science, Pilani, Jhunjhunu, Rajasthan, India Yasar Abbas Ur Rehman TCL Corporate Research Hong Kong, Hong Kong Shailendra Shukla Computer Science and Engineering Department, MNNIT Allahabad, Allahabad, India Amit Kumar Singh Computer Science and Engineering Department, NIT Patna, Patna, India Dinesh Singh Department of Computer Science & Engineering, Motilal Nehru National Institute of Technology Allahabad, Prayagraj, India Ghanshyam Singh Centre for Smart Information and Communication Systems, Department of Electrical and Electronics Engineering Science, University of Johannesburg, Johannesburg, South Africa Shyam Pratap Singh Department of Electronics and Communication Engineering, Galgotia College of Engineering and Technology, Greater Noida, Uttar Pradesh, India
- 12. Contributors xv Gautam Srivastava Department of Math and Computer Science, Brandon University, Brandon, MB, Canada Muhammad Tariq National University of Computer and Emerging Sciences (NUCES), Peshawar, Pakistan Prabhat Thakur Department of Electrical and Electronics Engineering Science, University of Johannesburg, Johannesburg, South Africa Asheesh Kumar Mani Tripathi Information and Cyber Security Services, HCL, Noida, India
- 13. Contents Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii CHAPTER 1 A review on Internet of Multimedia Things (IoMT) routing protocols and quality of service . . . . . . . . . . 1 Dinesh Singh, Ashish Kumar Maurya, Rupesh Kumar Dewang, and Niharika Keshari 1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Routing protocols in IoMT . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.2.1 Fault tolerant routing protocol . . . . . . . . . . . . . . . . . . . 3 1.2.2 DDSV routing protocol . . . . . . . . . . . . . . . . . . . . . . . . 6 1.2.3 Optimal routing for multihop social-based D2D communications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.2.4 Green-RPL routing protocol . . . . . . . . . . . . . . . . . . . . 11 1.2.5 Context-aware and load balancing RPL (CLRPL) . . . . 12 1.2.6 Energy-Harvesting-Aware (EHA) routing protocol . . . 13 1.2.7 Optimized 3-D deployment with lifetime constraint (O3DwLC) protocol . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1.3 Quality of Service (QoS) routing in IoMT . . . . . . . . . . . . . . . 20 1.3.1 Traffic-aware QoS routing protocol . . . . . . . . . . . . . . . 20 1.3.2 QoS-aware and Heterogeneously Clustered Routing (QHCR) protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 1.3.3 QoS in wireless multimedia sensor network based IoMT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 1.3.4 SAMS framework for WMSN-based IoMT . . . . . . . . . 22 1.4 Conclusion and future directions . . . . . . . . . . . . . . . . . . . . . . 26 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 CHAPTER 2 Energy efficient data communication in Internet of Multimedia Things (IoMT) . . . . . . . . . . . . . . . . . . . . . . . 31 Shailendra Shukla, Asheesh Kumar Mani Tripathi, and Amit Kumar Singh 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.2 Related work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.2.1 Routing protocol for IoMT . . . . . . . . . . . . . . . . . . . . . 32 2.2.2 Boundary detection and virtual coordinate algorithms . 33 2.3 System model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 2.3.1 Problem description . . . . . . . . . . . . . . . . . . . . . . . . . . 35 2.4 Proposed approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 v
- 14. vi Contents 2.4.1 Working of algorithm . . . . . . . . . . . . . . . . . . . . . . . . . 35 2.4.2 Phase I: boundary detection algorithms . . . . . . . . . . . . 36 2.4.3 Phase II: assignment of virtual coordinates to boundary nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 2.5 Implementation and results . . . . . . . . . . . . . . . . . . . . . . . . . . 40 2.5.1 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 2.5.2 Simulation setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 2.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 CHAPTER 3 Visual information processing and transmission in Wireless Multimedia Sensor Networks: a deep learning based practical approach . . . . . . . . . . . . . . . 47 Yasar Abbas Ur Rehman and Muhammad Tariq 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 3.2 Literature review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.3 Deep learning based practical approach for WMSN . . . . . . . . 51 3.3.1 Convolutional Neural Networks . . . . . . . . . . . . . . . . . . 52 3.3.2 Activation units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 3.3.3 Max pooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 3.3.4 Batch normalization . . . . . . . . . . . . . . . . . . . . . . . . . . 54 3.3.5 Regularization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 3.3.6 Transfer learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 3.3.7 Recurrent Neural Networks . . . . . . . . . . . . . . . . . . . . . 55 3.3.8 Auto-Encoders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 3.3.9 Generative Adversarial Networks . . . . . . . . . . . . . . . . 55 3.3.10 Mobile neural networks . . . . . . . . . . . . . . . . . . . . . . . . 55 3.4 Computer vision algorithms in WMSN . . . . . . . . . . . . . . . . . 56 3.4.1 Object detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 3.4.2 Semantic segmentation . . . . . . . . . . . . . . . . . . . . . . . . 58 3.4.3 Image restoration, superresolution, and semantic colorization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 3.5 Conclusion and future considerations . . . . . . . . . . . . . . . . . . . 59 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 CHAPTER 4 Cognitive radio-medium access control protocol for Internet of Multimedia Things (IoMT) . . . . . . . . . 67 Shweta Pandit, Prabhat Thakur, Alok Kumar, and Ghanshyam Singh 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 4.2 Difference in Internet of Things and Internet-of-Multimedia- Things . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 4.2.1 Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
- 15. Contents vii 4.2.2 Performance parameters . . . . . . . . . . . . . . . . . . . . . . . 71 4.2.3 Scalar data of IoT and big data of IoMT . . . . . . . . . . . 72 4.3 Internet-of-Multimedia-Things and cognitive radio . . . . . . . . 73 4.3.1 Feasibility of cognitive radio to multimedia traffic . . . . 73 4.3.2 Why cognitive radio is a potential candidate for Internet-of-Multimedia-Things? . . . . . . . . . . . . . . . . . 73 4.3.3 Cognitive radio in IoMT network . . . . . . . . . . . . . . . . 75 4.4 CR-MAC protocols for Internet-of-Multimedia-Things . . . . . 77 4.5 Challenges in cognitive radio based IoMT network . . . . . . . . 88 4.5.1 Spectrum sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 4.5.2 Spectrum sharing/management . . . . . . . . . . . . . . . . . . 90 4.5.3 Spectrum mobility . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 4.5.4 Miscellaneous challenges . . . . . . . . . . . . . . . . . . . . . . 93 4.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 CHAPTER 5 Multimedia nano communication for healthcare – noise analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Shyam Pratap Singh, Vivek K. Dwivedi, and Ghanshyam Singh 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 5.2 Noise in nano communication and statistical tools . . . . . . . . . 100 5.3 Fundamentals on various noises in MNC . . . . . . . . . . . . . . . . 102 5.3.1 Additive inverse Gaussian (IG) noise . . . . . . . . . . . . . . 104 5.3.2 Normal inverse Gaussian noise . . . . . . . . . . . . . . . . . . 105 5.3.3 Stable distribution noise . . . . . . . . . . . . . . . . . . . . . . . 105 5.3.4 Modified Nakagami distribution noise . . . . . . . . . . . . . 105 5.3.5 Radiation absorption noise . . . . . . . . . . . . . . . . . . . . . 106 5.3.6 Sampling and counting noise . . . . . . . . . . . . . . . . . . . . 106 5.3.7 Molecular displacement noise . . . . . . . . . . . . . . . . . . . 107 5.3.8 Drug delivery noise . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 5.3.9 Reactive obstacle noise . . . . . . . . . . . . . . . . . . . . . . . . 108 5.3.10 External noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 5.3.11 System noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 5.4 Fundamentals on various noises in EMNC . . . . . . . . . . . . . . . 110 5.4.1 Johnson–Nyquist noise/thermal noise . . . . . . . . . . . . . 110 5.4.2 Black-body noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 5.4.3 Doppler-shift-induced noise . . . . . . . . . . . . . . . . . . . . 111 5.4.4 Molecular absorption noise . . . . . . . . . . . . . . . . . . . . . 112 5.4.5 Body radiation noise . . . . . . . . . . . . . . . . . . . . . . . . . . 112 5.5 Physical and/or stochastic models for noise in MNC . . . . . . . 113 5.5.1 Additive inverse Gaussian noise (AIGN) . . . . . . . . . . . 113 5.5.2 Normal inverse Gaussian noise (NIGN) . . . . . . . . . . . . 113 5.5.3 Stable distribution noise . . . . . . . . . . . . . . . . . . . . . . . 114
- 16. viii Contents 5.5.4 Modified Nakagami distribution noise . . . . . . . . . . . . . 115 5.5.5 Radiation absorption noise . . . . . . . . . . . . . . . . . . . . . 116 5.5.6 Sampling and counting noise . . . . . . . . . . . . . . . . . . . . 116 5.5.7 Molecular displacement noise . . . . . . . . . . . . . . . . . . . 117 5.5.8 Drug delivery noise . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 5.5.9 Reactive obstacle noise . . . . . . . . . . . . . . . . . . . . . . . . 120 5.5.10 External noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 5.5.11 System noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 5.6 Physical and/or stochastic models for noise in EMNC . . . . . . 122 5.6.1 Johnson–Nyquist noise/thermal noise . . . . . . . . . . . . . 122 5.6.2 Black-body noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 5.6.3 Doppler-shift-induced noise . . . . . . . . . . . . . . . . . . . . 123 5.6.4 Molecular absorption noise . . . . . . . . . . . . . . . . . . . . . 123 5.6.5 Body radiation noise . . . . . . . . . . . . . . . . . . . . . . . . . . 124 5.7 Simulation results of different noises under nano communication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 5.8 Open research challenges on noises in nano communication . . 125 5.8.1 Challenges on the noises in MNC . . . . . . . . . . . . . . . . 126 5.8.2 Challenges on the noises in EMNC . . . . . . . . . . . . . . . 128 5.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 CHAPTER 6 The use of deep learning in image analysis for the study of oncology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 Bailey Janeczko and Gautam Srivastava 6.1 The difficulties in meeting demand . . . . . . . . . . . . . . . . . . . . . 133 6.1.1 The medical imaging equipment issue . . . . . . . . . . . . . 134 6.2 What is deep learning? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 6.2.1 Neuron’s and activation functions . . . . . . . . . . . . . . . . 136 6.2.2 Feedforward neural networks . . . . . . . . . . . . . . . . . . . 138 6.2.3 Recurrent neural networks . . . . . . . . . . . . . . . . . . . . . . 139 6.2.4 Long-short term memory model . . . . . . . . . . . . . . . . . 140 6.2.5 Convolutional neural networks . . . . . . . . . . . . . . . . . . 140 6.3 Deep learning techniques and processes . . . . . . . . . . . . . . . . . 142 6.3.1 Backpropagation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 6.3.2 Image segmentation . . . . . . . . . . . . . . . . . . . . . . . . . . 142 6.3.3 Object localization in the study of oncology . . . . . . . . 142 6.4 Data difficulties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 6.4.1 Stepping away from supervised-learning . . . . . . . . . . . 143 6.4.2 Dimension reduction . . . . . . . . . . . . . . . . . . . . . . . . . . 144 6.4.3 Synthetic data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 6.5 The current uses of deep learning image analysis . . . . . . . . . . 147 6.6 The future of deep learning image analysis in the study of oncology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
- 17. Contents ix 6.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 CHAPTER 7 Automatic analysis of the heart sound signal to build smart healthcare system . . . . . . . . . . . . . . . . . . . 151 Puneet Kumar Jain and Om Prakash Mahela 7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 7.1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 7.1.2 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 7.2 Literature survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 7.2.1 Denoising algorithms . . . . . . . . . . . . . . . . . . . . . . . . . 154 7.2.2 Segmentation algorithms . . . . . . . . . . . . . . . . . . . . . . . 155 7.2.3 Feature extraction and classification algorithms . . . . . . 156 7.3 Methods and materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158 7.3.1 Theoretical background about TQWT . . . . . . . . . . . . . 158 7.3.2 HSMM based heart sound signal segmentation . . . . . . 161 7.3.3 Machine learning based classification algorithms . . . . . 162 7.4 Proposed methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 7.4.1 Data preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 7.4.2 TQWT based denoising algorithm . . . . . . . . . . . . . . . . 166 7.4.3 Segmentation using Springer’s HSMM algorithm . . . . 169 7.4.4 Feature extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 7.4.5 Classification of heart sound signal . . . . . . . . . . . . . . . 172 7.5 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 7.5.1 Performance evaluation metrics . . . . . . . . . . . . . . . . . . 173 7.5.2 Results using the SVM method . . . . . . . . . . . . . . . . . . 174 7.5.3 Results using KNN method . . . . . . . . . . . . . . . . . . . . . 175 7.5.4 Results using ensemble method . . . . . . . . . . . . . . . . . . 177 7.5.5 Comparison of the proposed method with other methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 7.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 CHAPTER 8 Efficient single image haze removal using CLAHE and Dark Channel Prior for Internet of Multimedia Things . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 Prateek Ishwar Khade and Amitesh Singh Rajput 8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 8.1.1 Efficient multimedia processing for IoMT . . . . . . . . . . 190 8.2 Background and related work . . . . . . . . . . . . . . . . . . . . . . . . . 191 8.2.1 Dark Channel Prior (DCP) . . . . . . . . . . . . . . . . . . . . . 191 8.2.2 Contrast-Limited Adaptive Histogram Equalization (CLAHE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
- 18. x Contents 8.3 Analysis of adaptive contrast enhancement with DCP and its optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 8.3.1 Adaptive contrast enhancement with DCP . . . . . . . . . . 193 8.3.2 Optimization for computational advantage . . . . . . . . . . 193 8.4 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 8.4.1 Quality of haze removal . . . . . . . . . . . . . . . . . . . . . . . 196 8.4.2 Performance assessment . . . . . . . . . . . . . . . . . . . . . . . 199 8.4.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 8.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 CHAPTER 9 A supervised and unsupervised image quality assessment framework in real-time . . . . . . . . . . . . . . 203 Zahi Al Chami, Chady Abou Jaoude, and Richard Chbeir 9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 9.1.1 Motivating scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 9.2 Related work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 9.2.1 Full-reference image quality assessment . . . . . . . . . . . 206 9.2.2 No-reference image quality assessment . . . . . . . . . . . . 207 9.3 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208 9.4 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 9.4.1 Data model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 9.4.2 Data manipulation functions . . . . . . . . . . . . . . . . . . . . 210 9.5 Data quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 9.5.1 Neural network architecture for image quality assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 9.5.2 Face alignment anomaly detection . . . . . . . . . . . . . . . . 215 9.5.3 Image score . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 9.6 Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 9.6.1 Stream processing module . . . . . . . . . . . . . . . . . . . . . . 217 9.6.2 Back-end module . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218 9.7 Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 9.7.1 Experimental setup . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 9.7.2 Experimental protocol . . . . . . . . . . . . . . . . . . . . . . . . . 222 9.7.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 9.8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229 CHAPTER 10 A computational approach to understand building floor plan images using machine learning techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Shreya Goyal, Chiranjoy Chattopadhyay, and Gaurav Bhatnagar
- 19. Contents xi 10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 10.2 Motivation of the problem . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 10.3 Problem statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237 10.3.1 Brief description of the work done . . . . . . . . . . . . . . . . 237 10.4 Literature survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238 10.4.1 State of the art in graphic recognition . . . . . . . . . . . . . 238 10.4.2 State of the art in floor plan analysis . . . . . . . . . . . . . . 239 10.4.3 Publicly available floor plan datasets . . . . . . . . . . . . . . 240 10.4.4 Symbol spotting in document images . . . . . . . . . . . . . 240 10.4.5 Image description generation . . . . . . . . . . . . . . . . . . . . 241 10.4.6 Evaluation of text generation . . . . . . . . . . . . . . . . . . . . 242 10.5 Descriptive narration generation from floor plan images . . . . . 242 10.5.1 System overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 10.5.2 Room annotation learning model . . . . . . . . . . . . . . . . . 244 10.5.3 Bag of decor (BoD) . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 10.5.4 Semistructured description generation . . . . . . . . . . . . . 248 10.5.5 Experimental findings . . . . . . . . . . . . . . . . . . . . . . . . . 249 10.5.6 Results for description generation . . . . . . . . . . . . . . . . 251 10.6 Application to smart homes and buildings . . . . . . . . . . . . . . . 253 10.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
- 20. 1 CHAPTER A review on Internet of Multimedia Things (IoMT) routing protocols and quality of service Dinesh Singh, Ashish Kumar Maurya, Rupesh Kumar Dewang, and Niharika Keshari Department of Computer Science & Engineering, Motilal Nehru National Institute of Technology Allahabad, Prayagraj, India 1.1 Introduction During the last decade, Internet of Things (IoT) has grown very fast and connected numerous devices worldwide. IoT can provide anytime, anyplace connectivity for anyone and anything [1,2]. Presently, 23.8 billion units of devices are connected to the Internet, and this amount is expected to jump 41.2 billion units by 2025 [3]. The IoT devices have limited memory, size, energy, and computing capabilities [4]. Thus, these IoT connected devices highly depend on reliable communication network and efficient routing protocols [5]. Technical developments in IoT operating systems [6], IoT enabling platforms [7], 5G IoT [8,9], wireless sensor networks (WSNs) [10, 11], software defined networks (SDNs) [12], mobile ad hoc networks (MANETs) [13–16], vehicular ad hoc networks (VANETs) [17,18], fog/edge computing [19,20], cloud computing [21], are facilitating IoT to comprehend to connect anything and anywhere [22]. The massive growth in multimedia on-demand traffics such as images, audios, and videos, has evolved the concept of Internet of Multimedia Things (IoMT) from IoT [22]. IoMT devices produce massive amount of multimedia data and are more constrained than IoT devices. They need more processing power, massive memory storage, adequate bandwidth, and consumes more power to support the underlying application efficiently [22,23]. The comparison between key data characteristics of IoT and IoMT [22] has shown in Table 1.1. The IoMT has diverse applications like smart grid, smart cities, industrial IoT, smart homes, health IoT, smart farming and agriculture, traffic monitoring, soil health monitoring, and satellite control systems. Most of these applications are interactive applications that involve reliable and timely delivery of data. Hence, it demands efficient routing protocols, and enforces stringent quality of service (QoS) parameters [23]. Due to transmission of unstructured, bulky and multimedia data over the network, IoMT needs revision of existing routing pro- Internet of Multimedia Things (IoMT). https://doi.org/10.1016/B978-0-32-385845-8.00006-X Copyright © 2022 Elsevier Inc. All rights reserved. 1
- 21. 2 CHAPTER 1 A review on IoMT routing protocols and quality of service Table 1.1 Comparison between key data characteristics of IoT and IoMT. S. No. IoT data Multimedia data 1 Linear data Gigantic data 2 Limited bandwidth Adequate Bandwidth 3 Less memory storage Massive memory storage 4 Limited processing Excessive processing 5 Delay tolerant Delay sensitive 6 Less power consumption More power consumption tocols of IoT which focus on energy aware computing, efficient feature extraction, and optimizing routing criteria to minimize delays [23]. The quality of service per- formance in IoMT is determined by various metrics such as bandwidth, packet loss ratio, delay, throughput, resource management, and energy conservation [23]. The quality of service in technical white paper [24] is defined by Microsoft as: “Network QoS refers to the ability of the network to handle this traffic such that it meets the service needs of certain applications.” The frequent packet loss and redundant packet delivery degrade the quality of constrained multimedia applications in IoMT network. The massive amount of data produced in IoMT network from many heterogeneous sources creates the processing overburden on the routing devices, resulting in packet loss. The multimedia traffic is growing very fast which increases newer challenges for computing, sharing, storing, and transmitting the multimedia data. Computing data in IoMT needs novel methods for fog/edge and cloud computing devices. Similarly, for storing multimedia data, different compression/decompression methods are re- quired in IoMT [22]. For example, utilizing a standard IoT routing protocol called RPL (Routing protocol for low-power and lossy networks) [25] in IoMT deployment scenarios, requires more improvements in terms of fault tolerance, energy-awareness, delay-awareness, and load balancing [22]. The Green-RPL [26], Context-Aware and Load Balancing RPL (CLRPL) [27], and free bandwidth (FreeBW)-RPL [28] are some modified versions of RPL protocol for IoMT. In this chapter, we give a com- prehensive review on routing protocols and quality of service in IoMT. The remainder of the chapter is organized as follows: In section 1.2, we illustrate the working methodology of routing protocols used in the IoMT network. The QoS routing protocols are explained in section 1.3 of this chapter. The conclusion and future directions is presented in section 1.4. 1.2 Routing protocols in IoMT Wireless Sensor Network (WSN) performs a crucial role in assisting IoMT in diverse application spectrum. The energy constraint of nodes in WSN somehow restricts it to fulfill the expectations of IoMT applications. The efficient routing protocols save the
- 22. 1.2 Routing protocols in IoMT 3 energy of WSN nodes and thus, allow to carry off a massive amount of IoMT data. Recently, cluster-based routing has attracted due to less communication overhead and reduced routing energy consumption. The cluster members of a cluster can save their routing energy expense because of the routing assistance of the cluster head. But, since everything has its pros and cons, cluster-based protocol has its disadvantage. In case of battery discharge or malfunctioning of the CH, all the cluster members would fail to transmit their collected data to the destination, which would defeat the smooth functioning of the IoMT application. Thus, the fault-tolerant approaches are in use in cooperation with cluster-based routing protocol for IoMT applications. Here, we illustrate the working principle of a fault-tolerant-based routing protocol used for IoMT. 1.2.1 Fault tolerant routing protocol The fault-tolerant routing protocol [29] uses cluster-based scenario for the processing of IoMT application data. It uses the cluster join method for handling faulty cluster heads (CHs). Whenever a fault is detected in a CH, its cluster members are the first ones to catch it as they fail to receive an acknowledgment from their CH upon receiv- ing transmitted data. So, they broadcast a help message to their neighboring CHs for re-selecting their CH. Out of all the respondents of the help message, the closest one is selected as the best backup CH. But, even this method has some leftover loopholes described as follows: • If there are many faulty CHs, they will introduce a new problem of help message explosion. • Each member of the faulty CH cluster may select a different CH as its backup for data routing, and then there would be a problem in handling that cluster’s data. The possible solutions to the above problems are: • One of the solutions is the re-clustering of the entire network nodes. But, this is a laborious and costly process. • Another solution is to keep a CH entirely for backing up the failed CHs by not initially assigning a cluster to it. • The other solution includes identifying overlapping nodes (two nodes having sim- ilar coverage area) and putting one of such nodes into sleep mode, then wake it up only when the existing CH fails. In view of these solutions to overcome the issues, a fault-tolerant routing protocol is illustrated in [29]. The protocol performs the following operations: 1. The protocol initially detects the failure in CHs and record faulty and nonfaulty CHs. 2. It analyzes the energy levels of nonfaulty CHs and attempts to construct the virtual CH. The data is routed through this virtual CH to the destination node.
- 23. 4 CHAPTER 1 A review on IoMT routing protocols and quality of service 3. The maximum fault-tolerant capability of nonfaulty CHs is calculated to accom- modate the members of faulty CHs. 4. The protocol uses a flow-bipartite graph (FBG) to estimate the energy cost of IoMT data routing. A flow-based pairs of faulty and fault-free CHs are formed to create an FBG. 5. Using the FBG, a flow transmission pattern is identified between the faulty and fault-free CHs. This indicates that which of the faulty CH is tolerated by which of the fault-free CHs. The virtual CH and FBG construction methods used in this routing protocol are explained below: Virtual CH The steps involved for creating of a virtual CH and virtual super frame are: 1. The destination node organizes the available energy of all the failure-free CHs, as shown in Fig. 1.1 [29]. There are three components in that total available energy: the energy taken for receiving the IoMT data from all the failure-free members, the energy is taken to aggregate the IoMT data obtained from the cluster members, and the energy taken to route the aggregated data from the CH to the destination node. 2. The remaining energy of a failure-free CH after using the above three components is used for virtual CH construction. The failure-free CHs receive IoMT data from failure-affected cluster members, aggregate the data and then transmit them to the destination node, as shown in Fig. 1.2 [29]. But, which failure-free CH is tolerated by which of the faulty node remains unknown. Thus, the concept of average transmission energy consumption is used to analyze energy used in fault tolerance. FIGURE 1.1 Organization of the virtual CH with the available energy of all the fault-free CHs.
- 24. 1.2 Routing protocols in IoMT 5 FIGURE 1.2 Data aggregation of faulty CHs by fault-free CHs. 3. A virtual superframe structure is constituted for some failure-free and some failure-affected members. The number of such virtual super frames transmitted by the virtual CH is defined as its transmission capability. Thus, the transmission capability is derived by dividing the sum of the energies of all failure-free CHs by sum of failure free and fault-tolerant energy consumption. 4. Finally, the verification of fault-tolerant capability begins. After confirmation of one or more CH failures, the minimum data of faulty CHs required to be handled by the fault-free CHs is calculated. If that data is greater than the availability of data received in fault-free CHs, then the purpose of IoMT application is not solved. Flow-Bipartite Graph After creating virtual CH, flow-based pairs of faulty and fault-free CHs are formed to develop a flow-bipartite graph (FBG). An FBG, as shown in Fig. 1.3 [29], is a graph whose vertices are divided into the source and destination sets such that every edge connects a vertex in source to a vertex in destination and is associated with a transmission cost and an energy cost. Each destination vertex is attached with a capacity cap. To establish an FBG, the following steps must be followed – 1. All faulty CHs act as source, and failure-free CHs as destination node. 2. The amount of input flow from each source vertex is set to the demanded amount of data (i.e., the data to be gathered at failure-free CHs due to failed CHs). 3. The capacity cap of the destination vertex is set to the available energy of the failure-free CHs. 4. The distance between the farthest member of a faulty CH and a fault-free CH is calculated and compared with a node’s communication range. If the former is larger, an edge can be created between those source and destination nodes.
- 25. 6 CHAPTER 1 A review on IoMT routing protocols and quality of service FIGURE 1.3 An example illustration of flow-bipartite graph formed by faulty and fault-free CHs. 5. The transmission cost of an FBG is calculated. It is the sum of cost of transmitting data from failure affected members to the chosen fault-free CH and for transmit- ting from failure-free CH to the destination node. Now, the fault-tolerant energy cost of FBG is calculated. 1. It is the sum of energy taken to assist the transmissions of failure-affected mem- bers to the destination node and the energy taken to transmit the IoMT data of the failure-free cluster members to the destination node. 2. With this, fault-tolerant load distribution can be achieved by making two or more fault-free CHs tolerate a single faulty CH. 3. The minimum cost flow (MCF) problem is also solved using the FBG approach such that the total minimum transmission cost can be obtained. 1.2.2 DDSV routing protocol An optimizing Delay and Delivery Ratio for Multimedia Big Data Collection in Mo- bile Sensing Vehicles (DDSV) routing protocol [30] uses optimized routing criteria so that the delay involves in data collection and data delivery is minimum. It consid- ers a road scenario that has multiple intersections on a fixed distance. The two types of moving vehicles, Bus and taxi, are considered on the road to receive the IoMT data generated from the source, i.e., traffic light. The vehicles deliver the received IoMT data to the other vehicles or the Data Collection (DCs) centers located at the
- 26. 1.2 Routing protocols in IoMT 7 FIGURE 1.4 DDSV routing scenario. intersection of the road segment based on the optimized routing criteria. The DDSV routing scenario is shown in Fig. 1.4 [30]. The route for IoMT data transfer consists of intersections and road segments. The protocol takes data packet priority as well as vehicular priority for decision-making purposes. The data packet priorities are assigned on the basis of location or area from where the data is generated. The routing scenario is partitioned into many urban and suburban sections. 1. At each intersection i of the road, the routing decisions are taken. The routing decision is based on the minimum of expected data delay (Di) on the possible routes. 2. The expected data delay (Di) is computed as the function of two components: movement probability (Pε i,j ) of vehicle ε at intersection i to the routing decision θi and data forwarding probability (dε i,j ). 3. There are multiple routes between intersection i and intersection j to transfer the IoMT data. The movement probability (Pε i,j ) of vehicle ε at intersection i is computed as pε i,j = α × Pε i,j + (1 − α) × (1 − Sε) (1.1) where Pε i,j is the priority of vehicle ε and Sε is the probability of the vehicle ε to travel in the suburban area. The Sε is defined as the ratio of time spent by a vehicle in suburban area to the time it taken in moving as per the historical trajectory datasets. 4. A vehicle finds the following possibilities, as shown in Fig. 1.5 [30], at each in- tersection i to route the IoMT data.
- 27. 8 CHAPTER 1 A review on IoMT routing protocols and quality of service FIGURE 1.5 DDSV outing decision at each intersection. The movement probability (Pε i,j ) of vehicle ε at intersection i in view of the above routing possibilities can be computed as Pε i,j (θi) = ⎡ ⎣ edges 1 − ϕε1 i,k ⎤ ⎦ × ⎡ ⎣ϕε i,j ⎛ ⎝1 − edges ωε1 i,k ⎞ ⎠ + ωε i,j − ωε i,j ϕε i,j ⎤ ⎦ (1.2) Where ϕε i,j is the probability of the vehicle ε to move on the road segment between intersection i and intersection j and ωε i,j is the probability of the vehicle ε to deliver the data packet to another vehicle moving on the road segment between intersection i and intersection j. 5. The data forwarding probability (dε i,j ) for a vehicle ε depends on the vehicle type, i.e., Bus or taxi. a. The Bus vehicle has predefined trajectories and the probability depends on the length of the road segment and average speed of the vehicle only. For Bus type vehicles, it is calculated as: dε i,j = se,r ∈Si,j le,r vε e,r (1.3) where le,r is the length of one of the subset of the road segment between intersection i and intersection j and ve,r is the average speed of the Bus on the respective road segment.
- 28. 1.2 Routing protocols in IoMT 9 b. The taxi type vehicles have no fixed trajectories and thus, the probability considers the transmission range, vehicle speed as well as the wireless trans- mission delay for computation. It is calculated as: dε i,j = 1 − e−r.oi,j × li,j × c r + e−r.oi,j × li,j vε i,j (1.4) Where r is the transmission range of vehicle’s Wi-Fi channel, c is the wireless delivery delay, oi,j and vi,j is the density degree and vehicle’s speed on the road segment between intersection i and intersection j, respectively. 6. For routing purpose, it is assumed that the priority of the vehicle moving in the suburban area is high as compared to vehicles moving in urban area. 1.2.3 Optimal routing for multihop social-based D2D communications An optimal routing algorithm for multihop communication between Device-to- Device (D2D) is given in [31]. The algorithm is well suited for efficient IoMT data communication and 5G networks. It considers the social behavior of the devices in communication to their neighbors. The algorithm computes the trust probability for D2D communications based on the rank model. Both, random and fixed locations of the base stations are considered for measuring the connection probability (CP) between any pair of devices. The measurement of CP is taken with and without, channel state information (CSI). The network model of the Social-aware multihop D2D routing algorithm is shown in Fig. 1.6 [31]. The transmission of information between D2D transmitter and D2D destination using the number of D2D relays is taking place. Here, the interference due to the cellular communication equipment (CUEs) and position of base stations (BSs) is considered in follow-up. The Nodes are working in half-duplex mode. A single antenna and D2D transmitter are assumed to be at the origin, and D2D destination is at a fixed distance away from the origin, whereas using the Poison Point Process (PPP), BS and CUEs are modeled. With the help of real data traces from online social networks, trust connectivity of D2D based on rank-based trust model is calculated. The probability that Di+1 is trusted by Di is given as P (t) i,i+1 = 1 GR β i (i + 1) (1.5) where β is the parameter from the rank-based model and G = N n=1[ 1 nβ ] is a nor- malizing factor. A trusted connection will be established only when two nodes can trust and com- municate with each other and thus, the trusted connectivity probability (T-CP) is defined as P(c) i,i+1 = P(t) i,i+1P(c) i,i+1 (1.6)
- 29. 10 CHAPTER 1 A review on IoMT routing protocols and quality of service FIGURE 1.6 Network model of optimal routing for multihop social-based D2D Communications. where the connectivity probability between Di and Di+1 is denoted as P (c) i,i+1. In case of random BS and fixed BS scenario, both CSI aware and not CSI aware situations are actively taken for the computation of the CP. In CSI aware situation, the channel state information between BSs and D2D transmitter is used in CP com- putation. However, none of the channel state information between BSs and D2D transmitter is used in CP computation. In random BS scenario, the CP between any D2D devices for given Di,i+1 depends on the intensity of BS and CUE, path lose exponent, transmit power of each CUE, threshold Ith. However, in fixed BS scenario, rest other parameters are similar to one that we use in random scenario except the two. Here, in place of intensity, the location and number of BSs is used in CP computation between any D2D devices for given Di,i+1. The objective of the routing algorithm is to find the optimal path between the D2D transmitter and the receiver. The path selection is based on the maximum value of the T-CP at every intermediate link. The maximum T-CP between D2D transmitter and D2D destination in presence of multiple D2D relays can be obtained by routing
- 30. 1.2 Routing protocols in IoMT 11 algorithm which helps in selecting optimal path. The optimal path satisfies: PT −CP π∗ = max π∗ Sπ i Sπ Pi,i+1 (1.7) where Sπ denotes the set of all potential paths between the D2D transmitter and receiver. The routing algorithm finds the maximum T-CP between source and desti- nation in a weighted graph with the help of standard Dijkstra’s algorithm. Initially, at each D2D node, distance between itself and other D2D devices and BSs are cal- culated and stored in topology information, which contains the neighbor list. The adjacency T-CP matrix is obtained by using updated topology information. There- fore, the D2D transmitter is set up as permanent node and all other nodes as temporary nodes. Here, we calculate the T-CP for all possible neighbors links and find the link with maximum T-CP. The intermediate destination node in selected link with max- imum T-CP is set as permanent node and now keeps on finding maximum T-CP unless all the nodes are marked as permanent node. Thus optimal path and maximum TCP of multihop D2D communication are obtained. The Computational complexity of routing algorithm is as same as that of Dijkstra algorithm, i.e., O(N2). 1.2.4 Green-RPL routing protocol The Green-RPL [26] is an energy-efficient green routing protocol for IoMT. It is an enhanced version of the RPL (Routing Protocol for Low-Power and Lossy Networks) protocol [25]. RPL is a routing protocol for resource-constrained devices that uses Destination Oriented Directed Acyclic Graph (DODAG) to maintain network topol- ogy. This DAG comprises multihop routes from leaf nodes to the root node. RPL optimizes an objective function and chooses the best path by selecting the desired predecessor nodes starting from leaf nodes. The previous RPL protocol implementa- tions are not feasible for IoMT and do not consider the multimedia data. In contrast, the Green-RPL protocol considers the data generated from multimedia devices. The Green-RPL protocol reduces energy consumption and carbon footprint emissions together with QoS requirements of applications. To guarantee QoS for a particular multimedia application, it determines the delay bound for the application. For example, in VoIP applications, the delay limitation usually is 120 msec. To ensure energy efficiency, the protocol considers the features of all the intermediary links between the leaf nodes and the root node and estimates the energy consumption by the chosen immediate predecessor node to support traffic needs for one more immediate successor node. An optimization model is given for this protocol based on various requirements and constraints. In Green-RPL routing protocol, the desired immediate predecessor node is se- lected according to the objective function minimizing the emissions of cumulative path carbon footprints on all the links from the immediate predecessor node to the root node, and satisfying the constraints such as cumulative path link energy, cumula- tive path delay, idle time, and battery status of the immediate predecessor node. When more than one immediate predecessor node satisfies the given constraints, then the
- 31. 12 CHAPTER 1 A review on IoMT routing protocols and quality of service node that provides the most greener path is chosen as the desired immediate prede- cessor node. In cumulative path link energy constraint, energy consumption by a node is calculated based on the quality of links in the selected route. The cumulative path delay constraint specifies that the predefined delay threshold should not be increased. According to the idle time constraint, an immediate predecessor node should be cho- sen as the desired node only if its idle time is sufficient to support another child node. In the last constraint battery status, a child node should choose an immediate prede- cessor node as the desired one if the battery level of its energy resources is greater than the predefined threshold. 1.2.5 Context-aware and load balancing RPL (CLRPL) A modified version of RPL, known as Context-aware and load balancing RPL under heavy and highly dynamic load for IoT and IoMT applications is discussed in [27]. It addresses the power depletion and packet loss problems associated with RPL. In CLRPL, one objective function called as Context-Aware Objective Function (COAF) and one routing metric called as Context-Aware Routing metric (CARF) has been given based on the contexts of the nodes. Here, context means any information related to nodes in the IoT network that can be used to characterize the nodes. Through COAF, CLRPL selects the immediate predecessor of nodes by finding their rank, Expected Transmission Count (ETX), and residual power level. Normally, ETX is used to give a realistic estimate of the channel status that is based on link losses. To avoid the loss during the initial phases of congestion and monitor both queue status and link status, ETX is used with other metrics that can give the sta- tus of the node queue in the network. The objective functions used in the previous implementation of RPL do not consider the latter states of nodes in the route while finding the rank. The rank value of a node may suffer from queue overflow and power drainage, which may further affect the rank of all downward nodes in the path. Thus, COAF recursively computes the latter states of the sequence of nodes in the route to- wards the root. COAF circumvents the thundering herd effect in the network, which happened in RPL, by actively considering the network states and gradually shifting toward the real rank value from a high-rank value. The protocol uses the routing metric, CARF, which considers the network traf- fic dynamicity index, node rank, and buffer queue utilization to calculate the route throughout heavy traffic conditions. Here, the network traffic dynamicity index shows that how much link has been utilized in the past. CARF reduces the effect of upstream immediate predecessors as it grows a long way in the path toward the root. Instead of just considering only the rank of nodes, this metric considers the other factors related to nodes and networks to select the suitable immediate predecessors in a network with heavy traffic. CLRPL avoids routing loops through the use of a countermeasure while selecting the best immediate predecessors using CARF and other parameters. This protocol also ensures load balancing by taking the number of nodes having an immediate predecessor from the list of candidate immediate predecessors for selec- tion of the immediate predecessor.
- 32. 1.2 Routing protocols in IoMT 13 Another enhanced version of RPL protocol named free bandwidth (FreeBW)- RPL is given in [28] which suggests a novel objective function called FreeBW that takes the FreeBW computations in the network layer. This protocol calculates the maximum FreeBW using the volume of the bandwidth to select the routing path to provide better performance. 1.2.6 Energy-Harvesting-Aware (EHA) routing protocol Energy harvesting (EH) is evolving as an essential technology for IoT and field de- ployable WSNs applications. In these applications, the deployed nodes or devices may recharge their batteries to perform actions by extracting energy from renewable sources such as radio frequency (RF) and solar-based radiation through the use of EH techniques. Energy-Harvesting-Aware (EHA) routing protocol proposed in [32] addresses the issues of Quality of Service and energy efficiency for IoT applications at the network layer. The protocol can efficiently perform according to the harvested- energy at IoT devices, the residual energy, and the dynamic traffic capacity produced by IoT applications. EHA routing protocol is based on the energy prediction process and energy-backoff process that describes cost metrics to choose the best path. In the energy prediction process, an enhanced estimate of existing energy levels from different sources has been determined using the Kalman filter method that considers current statistics and previous time steps. In the energy backoff process, network life- time has been increased by keeping the nodes having the lowest energy in sleep mode until their energy level is restored to perform operations. The protocol also considers hybrid energy harvesting sources and uses three EH techniques such as solar-based EH technique, moving vehicle-based EH technique, and RF-based EH technique. The EHA routing protocol first selects the node having the maximum level of residual en- ergy and then finds the link with the least cost using the shortest path algorithm given by Dijkstra to send the data packets to the target node from the origin nodes. Further, by using the EH techniques, the protocol aims to determine the sustainable paths to route the data packets. 1.2.7 Optimized 3-D deployment with lifetime constraint (O3DwLC) protocol The deployment of sensing devices is a very challenging task to satisfy both robust connectivity and the long lifetime of devices subject to cost minimization. The most common solution for handling such a situation is to install the relay nodes in the target environment. The function of relay nodes is to forward the sensed data up to the maximum distance sink node and thus requires additional storage and more power. The energy of sensing nodes can be saved for further collecting forwarding data. But installing the relay nodes is more challenging because of the roughness of the environment and the requirement of 3-D infrastructure. The roughness of the environment is due to natural calamities and nonexpected visiting birds and animals because they may destroy nodes. For maximizing connectivity with little cost and
- 33. 14 CHAPTER 1 A review on IoMT routing protocols and quality of service Table 1.2 Comparative analysis of routing protocols in IoMT (Part a). Paper Architec- ture Entities Communica- tion mode Link establishment criteria Suitability criteria Optimal path finding Noiseless path selection Environ- ment adaption Goal Methodology [29] Cluster Sensor nodes, Sink nodes Device-to- device Transmission energy consumption, Distance Transmission Cost, energy cost Integer Linear Programming model solved using IBM ILOG CPLEX Optimizer x x Designed a join based method which can perform preverification of fault tolerance, fault tolerant load distribution and fault tolerant cost minimization. A cluster based fault tolerant routing which minimize total energy consumption during packet transmission. The proposed approach uses virtual clustering and flow graph modeling to handle faulty CH. [30] General IoT multimedia storage devices, vehicle (bus and taxi type), DC Vehicle-to- vehicle Message establishment, time spent on road segment, density, speed, transmission range Prioritize over probability of delivery at nearest data centers Designed a threshold based optimal route policy algorithm x x Goal is reduction of packet drop of low priority data and task delivery within deadline. Proposed routing policy based on delivery delay and delivery prioritization of data. [31] General IoT device, Base station (BS) Device-to- device Distance, density Rank based Trusted Connectivity Probability (T-CP) Dijkstra Algorithm x Carried out an analysis for cases at known and unknown channel state information. Author proposed an approach of Social relationship based on T-CP for multihop routing path selection using tight lower bound and exact closed bound with decode and forward mechanism. (continued on next page)
- 34. 1.2 Routing protocols in IoMT 15 Table 1.2 (continued) Paper Architec- ture Entities Communica- tion mode Link establishment criteria Suitability criteria Optimal path finding Noiseless path selection Environ- ment adaption Goal Methodology [26] Tree topology Multimedia device Cumulative path carbon footprints, cumulative path delay, cumulative path link energy, battery status, idle time for parent node Delay constraints, battery consumption, type of energy resource – An adaptive and dynamic optimization based RPL which can support multimedia communication. Proposed a Green-RPL routing to minimize carbon footprint emissions and energy consumption using tree like network topology. [27] Tree topology IoT devices Point-to-point, multipoint to point, point to multipoint Remaining power status, queue utilization for parents to root chain Rank based Context aware routing metric (CARF) – x x Loop free and less packet loss as well less power depletion for RPL under heavy and dynamic load, eliminates thundering hard and equality illusion problem Context aware load balance RPL with DODAG structure find route using two steps as: 1) CARF, 2) parent selection mechanism. [32] General Sensors, router Device-to device Distance, battery voltage, packet length, data rate, Energy consumption during listening, transmitting, receiving and sleeping Dijkstra Algorithm x Aimed to maximize life-time in an energy efficient manner. Proposed a Kalman Filtering (KF) adaptive energy harvesting routing by enhancing energy back-off parameter. If harvested energy is high then use current energy and store the excess energy
- 35. 16 CHAPTER 1 A review on IoMT routing protocols and quality of service Table 1.3 Comparative analysis of routing protocols in IoMT (Part b). Paper Merits Demerits Evaluation metrics Simulation details Assumption Complexity Designed approach Comparison Tool [29] Average backup energy consumption for new CH without aggregation is 0.1856 mJ while proposed approach is around 0.0098 mJ Average failure affected as energy consumption is reduced around 10% at 0.2 failure rate while 4% increasing at failure rate up to 0.5 Average backup energy consumption, Average failure-affected energy consumption, Network lifespan after fault tolerance Sensor nodes (1000 with 10% powerful sensor and 90% sink node), Sensed data (500 bytes), failure threshold (0.2–0.5), Simulation run (50) Same data size sensed by each member – Virtual CH, Flow bipartite graph techniques New CH without aggregation, New CH with aggregation, distributed join CH, distributed join CH with multiple factor NS2 [30] Average delay during data re-collection is reduced around 17.3% for general and 41.8% for suburban and average data delivery ratio improved up to 16.9% with respect to OVDF. Delay decreased percentage graph turns towards deteriorate around 11.11% at 140 with respect to 120 in its own scheme. Average delay, Delivery ratio, Data packet impartiality Multidataset of Beijing city of “T-drive trajectory”, Data packet size (512 bytes), Data storage size (10M), Communication range (150 meters), vehicles (30–150) – – DDSV (delay and delivery ratio for sensing vehicles) OVDF − [31] Optimal path route by proposed algorithm is same as ES. Optimal path selection is depends on BS position Connectivity Probability Nodes (20), noise variance (1), Path loss exponent (4) Node equipped with half duplex mode single antenna. O((log n-2)!) Trusted Connectivity Routing Algorithm Exhaustive search (ES) Monte Carlo simulation (continued on next page)
- 36. 1.2 Routing protocols in IoMT 17 Table 1.3 (continued) Paper Merits Demerits Evaluation metrics Simulation details Assumption Complexity Designed approach Comparison Tool [26] Average packet delay is 20.4 msec, while ETX and OF0 requires 47 msec and 26.2 msec. Energy consumption depends on path link quality. At poor link successful delivery using single hop probably reduced and causes retransmission. Number of packet delivered, energy consumed. ICMPv6, packet size (128 byte), datarate (25 pkt/sec) – – Green-RPL ETX, OF0 Cooja Simulator for Contiki-OS [27] RPL experience 32% packet loss while proposed approach packet loss is 15% at 100 nodes. Instability and inconsistency increase at new node joining and frequent messaging. Queue loss ratio, packet loss ratio, lifetime, overhead, Runtime Number of nodes (50), Traffic rate (15–100 pkt/min), Packet format (IPv6), Transmission power (15 dBm), FIFO queue size (20 pkts), Range (100), – – CLRPL RPL Cooja Simulator for Contiki-OS [32] Less energy consumption wrt R-MPRT-Mod (12.54%) and AODV-EHA (52.3%). Works for very small packet size (1000 bits) Average consumed energy, lifetime, Alive nodes, Packet loss ratio, end-to-end delay Data-rate (250 Kbps), Simulation period (12 hr), End-to-end delay (0.5%), Throughput (5–80 pkt/s) – – EHARA R-MPRT-Mod, AODV-EHA MATLAB®
- 37. 18 CHAPTER 1 A review on IoMT routing protocols and quality of service FIGURE 1.7 A two layered hierarchical and 3-D grid based hierarchical cluster based arrangement of sensor nodes. long-life sensing devices, installing and placing 3-D relay nodes is the most optimal solution to handle the challenges of the wireless network in a harsh environment [33]. For improving performance, a two layered hierarchical setup can be assumed in large-scaled environment. Lower layer nodes (SN) gather data and send it cluster heads (CH) in the upper layer. The upper layer contains relay nodes (RN) for better transmission. Cluster nodes summarize gather data from the lower layer, and then with the help of relay node, it sends to the base station (BS) in the upper layer. In a 3-D cubic grid-based model, the target environment is assumed partitioned in a grid-based cluster in hierarchical manners. Cluster heads (CH) are placed in the most appropriate vertices surrounded by the largest number of sensor nodes, and the base station is placed on the fixed position as per requirement. For efficient connectivity between cluster heads and base station (BS), it is mandatory to seek proper placement of relay nodes in grid subject to minimize cost and maximize connectivity. Fig. 1.7 which is taken from [33], shows a two-layered hierarchical and 3-D grid-based hierarchical cluster-based arrangement of sensor nodes, respectively. The main agenda is to search the proper placement of RNs so that cost of deploy- ment could be minimized by reducing the total counts of CHs/RNs in model with the constraint of strong connectivity, low cost, and longer network lifetime. The overall cost can be minimizing by reducing the costlier equipment for the network scenario. Since the cost of equipment is dependent on the functionality (RNs in this case be- cause of additional storage and forwarding capacity) and thus, cost can be minimized by adjusting the number of RNs. The strong connectivity can be achieved by optimal design and reducing obstacles between communicating devices, and therefore, the grid model is more suitable for this purpose. Algorithm description Turjman et al. [33] have introduced an Optimized 3-D deployment with Lifetime Constraint (O3DwLC) algorithm to achieve the objective of minimal cost, long net- work lifetime, and high connectivity. This algorithm works in two phases. In the first phase, they designed a connected backbone structure using First Phase Relay Nodes (FPRN). These FPRNs are placed in an optimized manner so that the requirements
- 38. 1.2 Routing protocols in IoMT 19 of FPRNs become minimum for connecting cluster heads to the base station. Thus, a minimum cost spanning tree (MST) is used between vertices in the grid model. MST can be constructed using the 3-D grid vertices pointing location of nodes. Further, the algorithm seeks the two closest nodes in the set of nodes, CHs and BS. In case the closet two nodes are not adjacent on the 3-D grid, then it introduces again a mini- mum number of grid vertices on which the relays have to be positioned to establish a path between these two nodes. After connecting the closest two nodes repeatedly, the algorithm looks for the next closest nodes and so on. The extra relay nodes (SPRN) can be introduced in optimal manner to maximize network backbone connectivity. The connectivity can be computed by analyzing the consecutiveness of graph nodes in the grid. The algorithm locates optimized SPRNs position in the second phase of processing to maximize lifetime and minimize the backbone generated in the first phase. For routing multimedia data, the algorithm’s performance is measured based on backbone connectivity level, the number of CHs, RNs, BS nodes used, and the num- ber of turns. The number of turns measures the total turns for the deployed network in which it can stay operational. Discussion. There are many cluster-based [29], general (non-cluster-based) [30,31, 27] and tree-topology-based [25,26] routing protocols are discussed in Table 1.2 and 1.3. In [29], the proposed work avoids backup workload shifting to new CH by re- joining new cluster according to energy suitability. Although in this research work sensed data size were same for all sensor and no deadline, which is an irrelevant assumption. To prevent the challenge of [29], a deadline prioritize delivery routing protocol is designed by Li et al. [30] and a dynamic load balanced routing protocol on rank is proposed by Taghizadeh et al. [27]. Moreover, both approaches retransmit packet to reduce packet drop rate which has no threshold during frequent messag- ing which may cause overhead retransmit broadcast. Here, majority of research work minimize delay, packet drop while some approach gives attention energy consump- tion by designing green routing [26] and Energy-harvesting-aware routing [27]. The proposed work of [26,27] based on “energy-store and energy-use” protocol by uti- lizing the current available energy consumption of node and “excess-energy” for future use. Although, the suitability metric in [26] avoids battery level which re- sults to small lifetime and paper [27] ignores network heterogeneity. To overcome the network heterogeneity a social-behavioral multihop optimal routing proposed for device-to-device communication for dense IoMT environment in [31]. The proposed work finds path based on rank according to distance and density for fixed base sta- tion, while ignoring routing at outage period. The routing algorithms discussed here attempt to improve the efficiency of the IoMT network, energy-saving, fault-tolerant capability, optimal connection or link probability, etc. The energy-harvesting-aware (EHA) routing protocol discussed the quality issues of the IoMT services and the routing procedure. The other routing protocols specially designed for QoS are illus- trated in the next subsequent section.
- 39. 20 CHAPTER 1 A review on IoMT routing protocols and quality of service 1.3 Quality of Service (QoS) routing in IoMT In IoMT, there are two types of applications in general – video-based applications, which can tolerate some amount of packet-loss but requires timely delivery of a packet, and other applications that are sensitive to packet loss due to their constrained nature. The other category of applications relies upon retransmission to guarantee message delivery. Both application categories require QoS to provide a better experi- ence. The QoS protocols are discussed below in detail. 1.3.1 Traffic-aware QoS routing protocol The traffic-aware QoS routing protocol [34] is designed to solve the packet delay and packet loss issues. The protocol fulfills the QoS requirement of applications by introducing a greedy-based approach based on Yen’s K-shortest paths algorithm for computation of the optimal path. The routing protocol calculates the QoS path for all the incoming flows in an SDN controller by taking all the conditions into account of the SDN rule capacity constraint. The protocol considers the particular type of traffic and the associated QoS requirement like packet loss and packet delay or both. The routing protocol takes a graph and set of flows with their associated QoS requirement as input. The user must prioritize delay-sensitive (ds) and loss-sensitive (ls) to each flow. The protocol provides an output set of paths on which ds-flows and ls-flows can be routed. The protocol performs the following operation: • The protocol takes the cost of each link in terms of delay or packet loss as inputs and returns a feasible routing for all the ds and ls flows. • The ds and ls paths are considered with respect to assign priority with each appli- cation. • The ds and ls path are routed according to the path calculated, with the help of the K-Shortest path algorithm. • The last GET-QoS-Path function checks all the delay, loss, bandwidth, and check rules if all are satisfied, then returns TRUE, else FALSE. 1.3.2 QoS-aware and Heterogeneously Clustered Routing (QHCR) protocol The QoS-aware and Heterogeneously Clustered Routing (QHCR) protocol [35] is an energy-efficient routing protocol that provides the dedicated routes for the bandwidth-hungry, delay-sensitive, real-time, and QoS-aware applications and saves energy in the network. The QHCR protocol works in the following phases: Informa- tion gathering phase, cluster head (CH) election phase, node association phase, and intracluster communication. In the information-gathering phase, every node sends and receives the broadcast messages to and from their neighbors to collect the energy levels and other information. After exchanging the information from their neighbors, each node maintains a neighbor table and updates the base station. The neighbor table consists of information such as the number of neighbor nodes, initial energy levels
- 40. 1.3 Quality of Service (QoS) routing in IoMT 21 of neighbor nodes, the distance of neighbor nodes from the base station, and dis- tance between one neighbor node to another. This protocol uses CSMA/CD method to minimize the collision in the network when two nodes broadcast at the same time. QHCR is a centralized protocol in which the base station collects all information from the nodes in the network and uses that information to elect the cluster head. In the CH election phase, the cluster head is elected based on the cost value at each energy level. This cost value is computed using the information stored in the node’s neighbor table. A node having the lowest cost value is elected as the cluster head at every energy level. The initial energy of nodes can be divided into four energy levels. The nodes can be clustered based on these energy levels such that the nodes having the same energy levels belong to the same cluster. In the node association phase, the elected cluster head starts sending the broadcast messages to other nonCHs nodes and member nodes of the clusters in the network. When a nonCHs node does not get any broadcast message from any cluster head, they elect themselves as cluster heads. During intracluster communication, when the distance between nonCH nodes and CH or BS is very long, the nonCH nodes use intermediate nodes for forwarding the messages to them or other nodes. Thus, the main goal of the QHCR protocol is to select the best route for transmitting the messages with minimum delay by using the intermediate nodes lie on multiple paths between the sending node and the cluster head or base station. 1.3.3 QoS in wireless multimedia sensor network based IoMT The IoMT applications use sensor nodes to gather data for subsequent processing. The sensor nodes formulate an interconnected network to provide better coordination for input data sharing. This network uses wireless links to exchange multimedia data, popularly known as Wireless Multimedia Sensor Network (WMSN). The WMSN is an integral part of IoMT networks. The objective of WMSN is to provide the real-time delivery and standard QoS of multimedia data within constrained energy requirements. The key QoS parameters are minimum packet drop, reduced latency, in-order, and prioritized multimedia data delivery. Hamid and Hussain [36] presented a detailed survey on layered and cross-layered approaches used to achieve QoS in WSMN based IoMT. Here, the QoS requirements to deliver multimedia content are explicitly examined on application, transport, routing, MAC, and physical layer. They illustrate the use of less complex multimedia data coding schemes at the application layer; Unequal Error Protection (UEP) method to provide reliability, priority-based rate control methods to manage congestions, and optimized packet scheduling to tol- erate the processing delay at the transport layer; geographic routing protocol [37] to handle delay constrained multimedia data and multipath routing [38] to solve the load balancing issue on aggregator sensor node at the routing layer; dynamic adjustment of contention window (CW) to handle heterogeneous multimedia services, Multiple Queuing methods to satisfy the scheduling requirements of different of multimedia data flows and variable contention period to better channel access at the MAC layer; dynamic control of beacon interval and use of Ultra-Wideband (UWB) technology to
- 41. 22 CHAPTER 1 A review on IoMT routing protocols and quality of service achieve reliable listening and a high rate of data transmission at the physical layer. A special investigation on cross-layer approaches, i.e., optimized cooperation of dif- ferent layers, to satisfy the QoS requirements of multimedia data is presented. The cross-layer protocols in WMSN successfully achieve the real-time delay constraint, rate adaptation, reliability, and optimized energy requirements. 1.3.4 SAMS framework for WMSN-based IoMT With the increase of the Internet of Things (IoT) and real-time adjustment develop- ment, people’s quality of life is improving. Multimedia wireless sensor nodes and devices that make the Internet of Multimedia Things (IoMT) are essential for IoT applications, which are diverse in nature. These constrained nodes and devices de- pend on standardized communication stacks and effective routing protocols that form the Wireless Multimedia Sensor Network (WMSN). Quick response services, traffic control, smart cities, smart hospitals, smart agriculture, smart buildings, criminal in- spection, security systems, Internet of bodies (IoB), and Industrial IoT (IIoT) are examples of real-world multimedia technologies. IoMT devices need high protec- tion, higher bandwidth, memory, and faster computing resources to process data. This study suggests a model for a cluster-based hierarchical WMSN called Seam- less and Authorized Streaming (SAMS) [39]. There are two phases for building secure clusters of a SAMS: set-up and steady-state. During the set-up phase, two- level authentication (i.e., between BS and elected CHs; and formation of clusters) is performed, and during the steady-state phase, seamless data transmission is achieved. Only legal nodes may transmit captured data to their Cluster Heads (CHs) after these clusters have been established. During mutual authentication, each multimedia node in the cluster regularly senses the environment and stores any data received in a buffer with a predefined limit. Each multimedia node waits for its turn to transmit the capture data using its allo- cated timeslot/channel that are prescribed by its respective cluster head. This waiting can cause excessive packet loss and end-to-end delay in wireless channels. A novel channel allocation technique is used to keep the WMSNs’ Quality of Service (QoS) at an acceptable scale to resolve this problem. A member node in one cluster moves to a nearby CH if the latter has an available channel for allocation in the event of a buffer overflow. Fig. 1.8 which is taken from [39], depicts the SAMS framework’s network architecture. The timely and accurate distribution of data is a key characteristic of IoMT. As a result, the SAMS outper- forms existing schemes in terms of different QoS metrics like average packet loss and average end-to-end delay while also providing robust protection against various attacks (such as tampering, sinkhole, insider, Sybil, and password guessing). Discussion. The research work reviewed in this section are QoS oriented protocols, which improve the seamless real-time servicing for IoMT. The comparative analysis of QoS protocols for IoMT is summarized in Table 1.4 and 1.5. The quality of service routing protocol in [34] focuses on throughput, jitter and reliability for two types of applications such as delay and loss sensitive using greedy heuristic based k-shortest
- 42. 1.3 Quality of Service (QoS) routing in IoMT 23 Table 1.4 Comparative analysis of QoS protocols for IoMT (Part a). Paper Architec- ture Entities Communica- tion mode Link establishment criteria Suitability criteria Optimal path finding Noiseless path selection Environ- ment adaption Goal Methodology [34] General IoT sensors, traffic lights Device-to- device Delay duration, packet-loss probability, current bandwidth Delay sensitive cost function and loss sensitive cost function K-shortest path x X Author motivation to design a network routing protocol which focuses QoS (throughput, jitter, reliability etc.) for delay and loss sensitive application. Proposed approach utilize a programmatic SDN and routes delay and loss sensitive data using greedy heuristic based k-shortest path algorithm. [35] Cluster WSN, sensor node Device to device Energy consumption at sensing, processing and radio communica- tion, number of bits transmitted, distance Initial and current energy cost function. – x X The aim is to design a QoS-aware routing protocol to deliver data for real-time and nonrealtime multimedia service with balanced load and fault tolerable. Proposed a cluster based energy efficient routing according to delay sensitivity, bandwidth, and energy fluctuation. CH is selected according to path metric (initial energy). [37] Dis- tributed Image Sensor nodes Device-to- device Distance, queue size, angular bearing towards destination Relative location based cost function – X Aims to propose a best effort QoS support, low delay with avoidable in-node routing table and information related to outdated states of sensor. Designed a routing algorithm for event driven real-time geographical distributed environment based on greedy approach. (continued on next page)
- 43. 24 CHAPTER 1 A review on IoMT routing protocols and quality of service Table 1.4 (continued) Paper Architec- ture Entities Communica- tion mode Link establishment criteria Suitability criteria Optimal path finding Noiseless path selection Environ- ment adaption Goal Methodology [38] General Sensors Device-to- device Size of real time and nonrealtime traffic. Probability of successful data delivery K-path finding x x Aims to maximize network lifetime through energy balancing across node according to acceptable delay. Propose a QoS and energy aware routing technique which balance fault tolerance using XOR based forward error correction method. [39] Cluster BS, sensors X to Y where X,Y ε {BS, sensors} Payload, distance buffer overflow – x x Goal is to propose secure intercluster communication framework to reduce packet loss due excessive waiting for its turn to send captured data. Proposed a lightweight secure AES based streaming for seamless delivery. After successful authentication Cluster formation initiated according to time slot. Here, CM moves according to buffer overflow.
- 44. 1.3 Quality of Service (QoS) routing in IoMT 25 Table 1.5 Comparative analysis of QoS protocols for IoMT (Part b). Paper Merits Demerits Evaluation metrics Simulation details Assumption Designed approach Comparison Tool [34] The proposed approach improves QoS Violated flows around 15%, 14%, and 13% (using AttMpls) and 39%, 38%, and 37% (using Goodnet) wrt MRC, SPD, and LARP respectively. Forwarding without fine grained QoS End-to-end delay, runtime, QoS violated flow, Active Node Topology (AttMpls, Goodnet), bandwidth (0.20–0.40 kbps), packet size (94–699 bytes) Assume constant jitter SWAY Minimum Rule Capacity (MRC), Shortest path delay ( SPD), LAgrangian Relaxation Aggregated cost(LARAC) − [35] Lifetime ends for proposed algorithm at 2750 rounds while ECHERP is 1300, PASCCC is 2000, and CCWM is 2100 rounds. Due to no energy harvesting feature the proposed work conserve no energy from renewable source. Lifetime, stability period, throughput, average energy consumption, End-to-end delay Sensors (100), buffer size (256 k-bytes), Transmit power (20 mW), receiving power (15 mW) Fixed sensing nodes, same packet size, QHCR CCWM, PASCCC, ECHERP MATLAB [37] The proposed scheme increases the life-time from 29% to 43% compared to angle based approach In-accurate location information leads to retransmission and packet loss. Number of alive node, delay Radio range (75 m), Nodes(196), payload size (1), energy (200–400 units) Payload size is fixed Hybrid scheme of angle and distance Distance based scheme, angle based scheme − [38] Due to use of forward error correction mechanism the proposed routing outperforms than MCMP to retrieve original message. MCMP protocol outperform better than proposed model with nonreal-data traffic because of low overhead by queuing. Energy consumption, delivery ratio, average delay Sensor node (300), transmission range (25 m), buffer size (256 K-bytes), transmit and receive power (13, 12 mW) Fully connected environment, sensors must be failure free MCMP NS2 [39] Secure against replay, DoS, Insider, Sybil, Eavesdrop Every time at cluster formation CH authenticates with BS Overhead, energy cost, resilience, cluster size, end-to-end delay Buffer size (60), AES-128 – SAMS Das approach, Amin-biswas approach −
- 45. 26 CHAPTER 1 A review on IoMT routing protocols and quality of service FIGURE 1.8 Wireless network model. path algorithm to overcome the bandwidth consumption while ignoring storage of outdated states of sensors which is resolved by Savidge et al. in [37]. Here, the authors presented a purely location-based QoS routing which can become as a faulty network due to sharing of in-accurate location information by attacker node. To overcome the above issue, an AES-128 authentication based secure and cluster-based approach is designed in [39]. Although, here is an issue of frequent authentication at the time of fault-tolerance between CH to BS can leads to overhead at fault-tolerance. Some other cluster based approach has been designed with fault tolerance by selecting CH and CM according to available bandwidth, delay as well as energy fluctuation for real-time and nonrealtime IoT application with fixed payload size [35]. Moreover, the approach is limited due to no energy consumption from renewable resource. In [38], Othman and Yahya gave an energy-aware protocol based on balanced fault tolerance and forward error correction method which shows significant overhead in queuing process. 1.4 Conclusion and future directions The Internet of Multimedia Thing (IoMT) has shown its applications in every cor- ner of today’s day-to-day life. The multimedia data sharing among the users through the IoMT network improves the user experiences at maximum height. However, the challenges in IoMT like real-time data delivery, throughput, device energy constraint, cross-layer issues, etc., still need to be handled in efficient manner. This chapter dis- cusses the routing and QoS protocols to tackle the challenge of the IoMT network. The design of routing and QoS protocols primarily concentrated on the low energy
- 46. References 27 consumption of the devices, minimum communication cost over the links, harvesting energy to reuse, and providing real-time experience to the user. The in-depth dis- cussion on each of the protocols is presented lucidly. Further, in future, we need to collectively focus on Quality of Experience (QoE) parameters like inter disruption time between subsequent packet deliveries, packet buffering interval on intermediate nodes, and user satisfaction feedback with discussed QoS parameters such as packet loss, end-to-end delay, error rate to evaluate the performance of IoMT applications. An improved cross-layer model can be designed in view of the QoE as mentioned above and QoS parameters. The smart routing protocol based on machine learning and optimization techniques to discover the energy-efficient route for bandwidth- hungry IoMT application need to be further explored. The storage requirements for Big Multimedia Data (BMD) open the door to reinvestigate the energy harvesting- based routing and link layer protocol and strengthen the provision of green routing protocol for IoMT networks. The Software Defined Network (SDN) architecture can be incorporated to improve the routing capabilities by decoupling network control from forwarding decisions in IoMT applications. Moreover, some use cases can be taken for different types of IoMT applications to exemplify identified challenges. References [1] L. Atzori, A. Iera, G. Morabito, The internet of things: a survey, Computer Networks 54 (15) (2010) 2787–2805. [2] S. Greengard, The Internet of Things, MIT Press, 2021. [3] https://www.statista.com/statistics/1101442/iot-number-of-connected-devices- worldwide/. (Accessed 3 October 2021). [4] F. Samie, L. Bauer, J. Henkel, Hierarchical classification for constrained IoT devices: a case study on human activity recognition, IEEE Internet of Things Journal 7 (9) (2020) 8287–8295. [5] Y.B. Zikria, M.K. Afzal, F. Ishmanov, S.W. Kim, H. Yu, A survey on routing protocols supported by the Contiki Internet of things operating system, Future Generation Com- puter Systems 82 (2018) 200–219. [6] F. Javed, M.K. Afzal, M. Sharif, B.S. Kim, Internet of Things (IoT) operating systems support, networking technologies, applications, and challenges: a comparative review, IEEE Communications Surveys and Tutorials 20 (3) (2018) 2062–2100. [7] A.K. Maurya, D. Singh, IoT enabling platforms, in: Internet of Things, An Indian Adap- tation: Concepts and Applications, Wiley, 2021, pp. 93–122. [8] S. Li, L. Da Xu, S. Zhao, 5G Internet of Things: a survey, Journal of Industrial Informa- tion Integration 10 (2018) 1–9. [9] G.A. Akpakwu, B.J. Silva, G.P. Hancke, A.M. Abu-Mahfouz, A survey on 5G networks for the Internet of Things: communication technologies and challenges, IEEE Access 6 (2017) 3619–3647. [10] T. Vaiyapuri, V.S. Parvathy, V. Manikandan, N. Krishnaraj, D. Gupta, K. Shankar, A novel hybrid optimization for cluster-based routing protocol in information-centric wireless sensor networks for IoT based mobile edge computing, Wireless Personal Com- munications (2021) 1–24.
- 47. 28 CHAPTER 1 A review on IoMT routing protocols and quality of service [11] A.K. Maurya, D. Singh, Median predictor based data compression algorithm for wireless sensor network, International Journal of Computer Applications 975 (2011) 8887. [12] W. Sun, Z. Wang, G. Zhang, A QoS-guaranteed intelligent routing mechanism in software-defined networks, Computer Networks 185 (2021) 107709.1. [13] V.K. Quy, V.H. Nam, D.M. Linh, N.T. Ban, N.D. Han, A survey of QoS-aware routing protocols for the MANET-WSN convergence scenarios in IoT networks, Wireless Per- sonal Communications (2021) 1–14. [14] A.K. Maurya, D. Singh, A. Kumar, R. Maurya, Random waypoint mobility model based performance estimation of On-Demand routing protocols in MANET for CBR applica- tions, in: 2014 International Conference on Computing for Sustainable Global Develop- ment (INDIACom), IEEE, 2014, pp. 835–839. [15] D. Singh, A.K. Maurya, A.K. Sarje, Comparative performance analysis of LANMAR, LAR1, DYMO and ZRP routing protocols in MANET using Random Waypoint Mobil- ity Model, in: 2011 3rd International Conference on Electronics Computer Technology, IEEE, 2011, pp. 62–66. [16] A.K. Maurya, D. Singh, A. Kumar, Performance comparison of DSR, OLSR and FSR routing protocols in MANET using random waypoint mobility model, International Jour- nal of Information and Electronics Engineering 3 (5) (2013) 440–443. [17] A. Katiyar, D. Singh, R.S. Yadav, State-of-the-art approach to clustering protocols in vanet: a survey, Wireless Networks 26 (7) (2020) 5307–5336. [18] G. Sun, Y. Zhang, H. Yu, X. Du, M. Guizani, Intersection fog-based distributed rout- ing for V2V communication in urban vehicular ad hoc networks, IEEE Transactions on Intelligent Transportation Systems 21 (6) (2019) 2409–2426. [19] G. Premsankar, M. Di Francesco, T. Taleb, Edge computing for the Internet of Things: a case study, IEEE Internet of Things Journal 5 (2) (2018) 1275–1284. [20] D. Kumar, A.K. Maurya, G. Baranwal, IoT services in healthcare industry with fog/edge and cloud computing, in: IoT-Based Data Analytics for the Healthcare Industry, Elsevier, Amsterdam, The Netherlands, 2021, pp. 81–103. [21] A.K. Maurya, A.K. Tripathi, Deadline-constrained algorithms for scheduling of bag- of-tasks and workflows in cloud computing environments, in: Proceedings of the 2nd International Conference on High Performance Compilation, Computing and Communi- cations, 2018, pp. 6–10. [22] Y.B. Zikria, M.K. Afzal, S.W. Kim, Internet of Multimedia Things (IoMT): opportunities, challenges and solutions, Sensors (Basel) 20 (8) (2020) 2334. [23] A. Nauman, Y.A. Qadri, M. Amjad, Y.B. Zikria, M.K. Afzal, S.W. Kim, Multimedia Internet of Things: a comprehensive survey, IEEE Access 8 (2020) 8202–8250. [24] Quality of Service Technical White Paper, Microsoft, Redmond, WA, USA, 1999. [25] T. Winter, P. Thubert, A. Brandt, J.W. Hui, R. Kelsey, P. Levis, K. Pister, R. Struik, J. Vasseur, R.K. Alexander, RPL: IPv6 Routing Protocol for Low-Power and Lossy Net- works, rfc, 6550, 1–157, 2012. [26] S.A. Alvi, G.A. Shah, W. Mahmood, Energy efficient green routing protocol for inter- net of multimedia things, in: 2015 IEEE Tenth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP), IEEE, 2015, pp. 1–6. [27] S. Taghizadeh, H. Bobarshad, H. Elbiaze, CLRPL: context-aware and load balancing RPL for IoT networks under heavy and highly dynamic load, IEEE Access 6 (2018) 23277–23291. [28] H. Bouzebiba, M. Lehsaini, FreeBW-RPL: a new RPL protocol objective function for in- ternet of multimedia things, Wireless Personal Communications 112 (2020) 1003–1023.
- 48. References 29 [29] J.W. Lin, P.R. Chelliah, M.C. Hsu, J.X. Hou, Efficient fault-tolerant routing in IoT wire- less sensor networks based on bipartite-flow graph modeling, IEEE Access 7 (2019) 14022–14034. [30] T. Li, S. Tian, A. Liu, H. Liu, T. Pei, DDSV: optimizing delay and delivery ratio for mul- timedia big data collection in mobile sensing vehicles, IEEE Internet of Things Journal 5 (5) (2018) 3474–3486. [31] G. Chen, J. Tang, J.P. Coon, Optimal routing for multihop social-based D2D communica- tions in the Internet of Things, IEEE Internet of Things Journal 5 (3) (2018) 1880–1889. [32] T.D. Nguyen, J.Y. Khan, D.T. Ngo, A distributed energy-harvesting-aware routing al- gorithm for heterogeneous IoT networks, IEEE Transactions on Green Communications and Networking 2 (4) (2018) 1115–1127. [33] F.M. Al-Turjman, H.S. Hassanein, M.A. Ibnkahla, Efficient deployment of wireless sen- sor networks targeting environment monitoring applications, Computer Communications 36 (2) (2013) 135–148. [34] N. Saha, S. Bera, S. Misra, Sway: traffic-aware QoS routing in software-defined IoT, IEEE Transactions on Emerging Topics in Computing (2018). [35] M. Amjad, M.K. Afzal, T. Umer, B.S. Kim, QoS-aware and heterogeneously clustered routing protocol for wireless sensor networks, IEEE Access 5 (2017) 10250–10262. [36] Z. Hamid, F.B. Hussain, QoS in wireless multimedia sensor networks: a layered and cross-layered approach, Wireless Personal Communications 75 (1) (2014) 729–757. [37] L. Savidge, H. Lee, H. Aghajan, A. Goldsmith, QoS-based geographic routing for event- driven image sensor networks, in: 2nd International Conference on Broadband Networks, IEEE, 2005, pp. 991–1000. [38] J. Ben-Othman, B. Yahya, Energy efficient and QoS based routing protocol for wireless sensor networks, Journal of Parallel and Distributed Computing 70 (8) (2010) 849–857. [39] M.A. Jan, M. Usman, X. He, A.U. Rehman, SAMS: a seamless and authorized multime- dia streaming framework for WMSN-based IoMT, IEEE Internet of Things Journal 6 (2) (2018) 1576–1583.
- 49. 2 CHAPTER Energy efficient data communication in Internet of Multimedia Things (IoMT) Shailendra Shuklaa, Asheesh Kumar Mani Tripathib, and Amit Kumar Singhc aComputer Science and Engineering Department, MNNIT Allahabad, Allahabad, India bInformation and Cyber Security Services, HCL, Noida, India cComputer Science and Engineering Department, NIT Patna, Patna, India 2.1 Introduction In the last couple of years, the advancement of low power and low-cost small-scaled devices such as Micro Electro Mechanical Systems (MEMS) [1] has given the explo- sive growth in the number of IoT devices [2] and increased scalability. IoT is useful for the application in home automation, security, automated applications to provide useful contextual information frequently to help with their works, and decision mak- ing. Internet-enabled sensors and actuators’ operations can be rapid, efficient, and more economical. Multimedia data in IoT [3] is bulky and specially meant for the real-time application requires real-time communication. Few examples are real-time multimedia-based surveillance systems [4] for smart home, office, at critical in- frastructure, telemedicine service in the smart hospital, remote patient monitoring, transportation management system, remote multimedia based monitoring, etc. Mul- timedia services require higher processing and memory resources, which raises the need for different modus operandi for multimedia-based IoT. The IoT devices are mainly equipped with a mic and camera. The major task of equipped devices is to sense the multimedia content and send it to sink nodes. For accurate information to be communicated, knowledge of the exact location of the sensor nodes in WSN [5] is mandatory. However, this location knowledge is possible with the integration of GPS devices to the sensor devices, but an increase in cost and reduction in battery life of sensor nodes are some of the demerits of this approach. So an alternate approach for this is to detect the most important nodes [6] in the networks. Critical node detection aka boundary node (B-N) [6] actually means to find the nodes, which reside in those positions of the network, which if removed then causes the partition in the networks. Hence, with the detection of boundary nodes, we can achieve connectivity and coverage of the network at a minimal cost. Till now, there Internet of Multimedia Things (IoMT). https://doi.org/10.1016/B978-0-32-385845-8.00007-1 Copyright © 2022 Elsevier Inc. All rights reserved. 31
- 50. 32 CHAPTER 2 Energy efficient data communication in IoMT are various methods have been proposed which focus on the detection of boundary nodes in the sensor networks. But to the best of our knowledge, no method provides the virtual coordinates [7] simultaneously to the detected boundary nodes. So in this chapter, we propose an algorithm, which not only detects the boundary nodes but also provides the virtual coordinates to them, which later contributes to providing the virtual coordinates to all the remaining nonboundary nodes in the wireless sensor networks. In IoMT the nodes are required to be capable of highly resourceful which is not possible as IoT devices are low power and fewer storage devices. Hence the detection of boundary nodes and virtual coordinate optimize the resource require- ments. The main contribution of this chapter is to develop a distributed algorithm to detect the boundary nodes and then virtual coordinate in IoT networks using the cen- tral node (ς). The proposed algorithm is free from any hardware dependability and flooding-based communications, it simply uses the RSSI value and cosine rules of trigonometry and. Finally, the proposed algorithm is deployed over the RPL proto- col. Remaining chapter has been organized as follows: Section 2.2 covers the related work. Section 2.3 deals with system model. Our proposed method is described in section 2.4. Section 2.5 deals with the results and comparison analysis with other proposed approach. And, finally we conclude our chapter in section 2.6. 2.2 Related work This section of the chapter is divided into subparts in the first part presents the re- ported work on IoMT content dissemination protocol and the second part presents the various boundary detection and virtual coordinate algorithms. 2.2.1 Routing protocol for IoMT Reported work on WSN in multimedia IoT (IoMT) shows various protocols which support the low power consumption and heterogeneity in IoT [8–11]. The WSN deployed for IoMT uses CoRE protocol rather than HTTP for the Internet and a lightweight OMA-DM to support low power and low resource IoMT devices. RPL (Routing Protocol for Low-Power and Lossy Networks) [12] is a routing protocol for resource-constrained devices that uses Destination Oriented Directed Acyclic Graph (DODAG) to maintain network topology. This DAG comprises multihop routes from leaf nodes to the root node. RPL optimizes an objective function and chooses the best path by selecting the desired predecessor nodes starting from leaf nodes. However, they are not developed for large multimedia data transmission. The Green-RPL [13] is an energy-efficient routing protocol for IoMT. It is an enhanced version of the RPL protocol [12]. The previous RPL protocol implemen- tations are not feasible for IoMT and do not consider multimedia data. In contrast, the Green-RPL protocol considers the data generated from multimedia devices. The Green-RPL protocol reduces energy consumption and carbon footprint emissions
- 51. 2.2 Related work 33 together with QoS requirements of applications. To guarantee QoS for a particular multimedia application, it determines the delay bound for the application. For exam- ple, in VoIP applications, the delay limitation usually is 120 msec. To ensure energy efficiency, the protocol considers the features of all the intermediary links between the leaf nodes and the root node and estimates the energy consumption by the cho- sen immediate predecessor node to support traffic needs for one more immediate successor node. An optimization model is given for this protocol based on various requirements and constraints. In Green-RPL routing protocol, the desired immedi- ate predecessor node is selected according to the objective function minimizing the emissions of cumulative path carbon footprints on all the links from the immediate predecessor node to the root node, and satisfying the constraints such as cumulative path link energy, cumulative path delay, idle time, and battery status of the imme- diate predecessor node. When more than one immediate predecessor node satisfies the given constraints, then the node that provides the most greener path is chosen as the desired immediate predecessor node. In cumulative path link energy constraint, energy consumption by a node is calculated based on the quality of links in the se- lected route. The cumulative path delay constraint specifies that the predefined delay threshold should not be increased. According to the idle time constraint, an imme- diate predecessor node should be chosen as the desired node only if its idle time is sufficient to support another child node. In the last constraint battery status, a child node should choose an immediate predecessor node as the desired one if the bat- tery level of its energy resources is greater than the predefined threshold. A modified version of RPL, known as Context-aware and load balancing RPL under heavy and highly dynamic load for IoT and IoMT applications is discussed in [14]. It addresses the power depletion and packet loss problems associated with RPL. Paper [15] pro- poses an efficient protocol CoUDP for real-time IoT multimedia communications. Paper [16] proposes a four-layer approach physical devices layer, network layer, com- bination layer, and context layer for IoMT data communications. Caching policy for popularity-aware video caching in the topology-aware content-centric network is pro- posed in the [17]. The paper [17] uses scalable video coding for fast video delivery and cached each video layer. Paper [18] proposed an approach to achieve reliability and reduce delay in IoMT networks. To enhance the QoS metrics, and efficient data gathering scheme is designed in [19]. Cross layer protocols for IoMT are proposed in [20] and [21]. The major limitation of reported work is most of them do not consider the energy efficiency and fault-tolerant approach for critical IoMT applications. 2.2.2 Boundary detection and virtual coordinate algorithms According to [22], Boundary detection algorithms can be classified into 3 categories, geometrical approach, statistical approach, and topological approach. Paper [23] is considered as a geometrical approach as it is based on the exact geometric location identification using global positioning devices such as GPS. Using the location infor- mation, sensor nodes identify the voids to create an alternative path for data traffic. The author has used the term stuck node for the nodes which are closer to the desti- nation than its one-hop neighbor. Paper [24] is based on some graph techniques and
- 52. 34 CHAPTER 2 Energy efficient data communication in IoMT computational geometry, to find out the closed polygonal boundaries of the voids. It is also based on the assumption that every node must be able to obtain its location. The main drawback with the geometrical approach is the requirement of an external hardware device to find out the geographical position, which is not only costly but also doesn’t work indoor and in bad weather conditions. The statistical approach is based on some complex mathematical and statistical functions and categorizes the nodes into the interior and boundary nodes. Paper [25] uses the restricted stress centrality measure which is a subclass of centrality indices and makes a comparison of it with a defined threshold value that is whether it is above or below the threshold value, based on which it categorizes the nodes into the above-mentioned classes. The main drawback of this approach is its impractical node distribution and high-density requirement. Paper [26] utilized the fact that the degree of nodes that are closer to the boundary of the network will be less in comparison to the node interior nodes. No requirement of global positioning devices is one of the major advantages of the statistical approaches, but these methods work only with the denser regions. Moreover, these methods require a node degree of more than 100 or higher. The third category is the topological approaches which use only the topological information available to sensor nodes in the network. Paper [27] describes a two-step algorithm boundary recognition and topology extraction. A combinatorial structure called flower and augmented cycles are searched in this approach which is the base of this algorithm, but it is not always the case where one can find a flower-like structure in a randomly deployed network which is the major drawback of this approach. Paper [28] builds a shortest path tree and finds out cut nodes by flooding the network. Using the cut nodes, it finds out the network boundary. Paper [29] and paper [30] use the iso-contours to recognize the network boundary. Iso-contours are build using hop- counts from a root node and the end nodes of these iso-contours are treated as the boundary nodes. The drawback with this approach is its high-density requirement and flooding of more than 3 different nodes. Paper [31] describes a 3 step approach for the detection of boundary nodes. In the first step, which is the information collection step, it builds an x-hop neighbor list. 2nd step is the path construction, in which the link between the x-hop neighbors is identified by building the first set and it uses a recursive procedure to construct the path and finally, the last set is made which contains the node which cannot be extended further. 3rd step is the path checking phase which categorizes the node as an interior or boundary node according to the intersection of neighbors of the first set and last set. According to my understanding, a lot of computation is done over a single node recursively for the path construction, which is the only drawback of this approach. 2.3 System model An IoT network is assumed where sensor nodes are homogeneously spread into a region L in such a way that it forms a connected graph G(V, E). In a connected graph
- 53. 2.4 Proposed approach 35 G(V, E), V represents the set of nodes and E represents the link between the nodes. The Nbr(v) represents the neighbor set of a node v (v ∈ V ). Each node is equipped with a small range camera and has a communication range ‘R’, also each node can determine the distance to its direct one hope neighbors using ranging techniques like RSSI, TOA, or TDOA [33] also they are not aware of their location. In consonance with Internet-Connected multimedia, we envisage our network broadly consists of the following components: Central Node (ς): A node ς ∈ V is said to be a cen- tral node if its graph centrality is maximum [34]. Boundary Node (B-N): A node B − N ∈ V is said to be a Boundary node if it lies on the critical position of the network graph G. Network Boundary: Network boundary is the imaginary circle formed by the Boundary nodes after the first phase of the algorithm. Interior Node (I − N): A node I − N ∈ V is said to be an interior node if it lies inside the network boundary and not a Boundary nodes (B-N). 2.3.1 Problem description The problem can be formulated as follows: To find an IoT node from a set Min(B − N) ∈ V such that it lies at the most critical places of the networks, detec- tion of a central node (ς) in the given graph G(V, E) which is at a minimum distance (min-dis) from all the nodes of networks. Our second problem is to find the nodes which are at the maximum distance from the central node and provide them some virtual coordinates so that a virtual overlay coordinate is a form for the data commu- nications. The main contribution of this chapter is to develop a distributed algorithm to detect the virtual coordinate which is based on graph centrality (ς). Developed an algorithm to find the network boundary in IoT networks using the central node and simple cosine rules of trigonometry and finally, deploying the algorithms over the RPL protocol. 2.4 Proposed approach 2.4.1 Working of algorithm The idea behind the proposed approach is to detect the boundary nodes and then to assign virtual coordinates to these boundary nodes (B-N) using the graph central- ity. To assign virtual coordinates it uses RSSI (Received signal strength indicator), Pr(RI(Nv[i])), to detect the distance between target nodes (mobile nodes present in the network). Since the boundary nodes are detected hence the anchor nodes [35] are detected from boundary nodes. Finally, trilateration is used to detect the virtual coordinate and critical nodes. We have divided our proposed approach into 2 phases. In phase-I, the bound- ary nodes detection algorithm also acts as anchor nodes. In phases II-a and II-b, the distance estimation is used to detect the center node, and then a virtual coordinate algorithm is proposed.
- 54. 36 CHAPTER 2 Energy efficient data communication in IoMT FIGURE 2.1 Node E as interior node (figure a) and boundary node (figure b). 2.4.2 Phase I: boundary detection algorithms Assume a smaller sub graph of four nodes A,B,C,E. Suppose node E wants to test whether it is a boundary or interior node. Node E tests the coverness information i.e. it is covered by at least three nodes. Fig. 2.1 a and b shows the placement of node E as interior node and boundary node. For any set of four noncollinear nodes with complete connectivity, as illustrated in Fig. 2.1. If node E lies inside the triangle then it is interior node otherwise it is a boundary node. The distance between the nodes (E,C), (E,B), and (E,A) is represented as x3, x2, and x1. We have used Lemma 1 to derive the relation to find the interior and boundary nodes and used HCR’s inverse Cosine Formula along with the property of inscribing a center point in a triangle to prove it. Lemma 1. A given node is interior node if it satisfies three relationships in terms of angle. 1. ∠ EAC ∠ CAB cos−1 b2 1 + c2 − x2 3 2x1c cos−1 c2 + b2 − a2 2cb (2.1) 2. ∠ EBC ∠ ABC cos−1 a2 + x2 2 − x32 2ax2 cos−1 a2 + c2 − b2 2ac (2.2) 3. ∠ ECA ∠ ACB cos−1 x2 3 + b2 − x2 1 2bx3 cos−1 a2 + b2 − c2 2ab (2.3) Proof. Angle divided into two parts will always be greater than the divided individual parts. For the boundary node this does not hold as illustrated in Fig. 2.1. Using Angle
- 55. Another Random Document on Scribd Without Any Related Topics
- 56. n'entendait pas qu'elle lui restât sur les bras. Mes administrés,—mes enfants!—n'auraient-ils pas éternellement le droit de me demander compte de leurs espérances déçues, de leur souscription gaspillée? Ils attendaient; ils avaient soif; et, en attendant, l'ancienne fontaine étant bouleversée par les maçons, la nouvelle n'existant pas encore, c'était chez moi que bêtes et gens venaient s'abreuver. Il y avait là de quoi faire prendre la campagne en horreur! Les faunes et les sylvains, la paix et la rêverie, s'enfuyaient au bruit de cette incessante cohue qui piétinait, criait, jurait, obstruait mes allées, brisait mes arbustes, salissait mon lavoir, écrasait mes fleurs, regardait derrière mes vitres et changeait mon jardin en place publique. Tout n'était-il pas préférable à ce provisoire? Ne valait-il pas mieux se jeter, comme Décius, dans l'abîme béant? Je me remémorais les noms de tous les grands bienfaiteurs de l'humanité, et je rougissais de honte en songeant au prix de quels sacrifices— souvent de quels martyres—ils avaient acheté ce titre glorieux. Je me reprochai mes hésitations comme un reste d'égoïsme littéraire ou mondain, et je me déterminai à passer outre. Je pus croire que mon héroïsme allait avoir sa récompense. Tout finit en ce monde, même les ouvrages interminables. Au bout d'un an la roue était placée, les tuyaux posés, les constructions achevées, la fontaine bâtie, le bassin creusé; le robinet, flambant neuf, ne demandait plus qu'à tourner pour nous verser ses trésors. L'ingénieur vint d'un air triomphant me prévenir que je n'avais qu'à fixer le jour de l'inauguration. Il fut décidé que ce serait le jour anniversaire de mon avénement à la mairie. Souvenir radieux, double fête, qui mêlerait toutes les ivresses du passé à toutes les joies de l'avenir! Une fois résigné sur la question d'argent, j'avais résolu de faire grandement les choses, et voici comment je réglai le programme de la journée: un bal champêtre aurait lieu sur la place; je danserais le premier quadrille avec la fille du percepteur des contributions, et, à un signal donné par le chef d'orchestre, la fontaine se mettrait à couler pendant que nous exécuterions, ma danseuse et moi, une
- 57. brillante pastourelle. Je ne prétendais pas copier les magnificences du troisième acte de la Juive et changer en vin le premier tribut de la source de Gigondas; mais du moins j'aurais soin que les bons villageois eussent constamment, pendant ce jour mémorable, du vin à mettre dans leur eau. Puis, après les premiers ébats, nous descendrions chez moi avec les notables du pays et l'élite de mes invités: un bon dîner nous attendrait, suivi, si nous étions en nombre, d'une sauterie au piano dans mon salon tapissé de toutes les fleurs de l'automne, comme un reposoir de procession. Ces riantes perspectives avaient achevé de me rasséréner. Les plaies d'argent se cicatrisaient à vue d'œil; je ne songeais plus qu'à ma gloire et au bonheur de mon peuple. Un seul nuage passait parfois sur ma félicité: que dis-je? ce qui m'inquiétait, au contraire, c'était l'absence de tout nuage, un ciel obstinément bleu depuis le commencement de l'été, une sécheresse implacable qui tarissait les rivières, épuisait les torrents, supprimait les sources, et m'inspirait sur le volume d'eau de ma fontaine des doutes invraisemblables, mais poignants. Quoique bien appauvri par mes profusions municipales, j'aurais donné dix écus d'une averse et dix louis d'une trombe. Vœux inutiles! Les jours succédaient aux jours, l'azur à l'azur, les vingt-cinq degrés Réaumur aux trente degrés centigrade. Je voyais bien une roue, des pistons, des tuyaux; mais tout cela ne fonctionnait pas encore; rien ne me prouvait que la chute d'eau fût assez forte pour que les pistons jouassent, pour que la roue tournât, pour que les tuyaux se remplissent; une ou deux fois je questionnai M. Jules: mais pouvais-je en obtenir une réponse catégorique? Il pressait la publication des bans et achetait la corbeille. «Alea jacta est!» avait dit un grand poëte en se préparant à noyer son pays. «Alea jacta est!» disais-je en m'apprêtant à désaltérer le mien.
- 58. XIX Je sus bientôt que l'inauguration de ma fontaine prenait dans le pays les proportions d'un événement. La province n'est pas difficile en fait de distractions et de commérages, et, depuis un an, il était clair que je préoccupais l'attention publique. Déjà ma nomination avait fort diverti les beaux esprits et les belles dames, curieux de savoir comment je concilierais le culte des Muses avec mes fonctions municipales. Un journaliste du chef-lieu n'avait pas peu contribué à ces flatteuses rumeurs en publiant sur mon installation triomphale un article fulgurant, où il peignait entre autres les vieillards de Gigondas éperdus d'émotion, ivres de joie, enflammés de vin de Tavel, embrassant, faute de mieux, le tronc de mes marronniers, que leurs grands-pères avaient plantés. Cette accolade donnée au règne végétal par le règne animal avait fait fortune, et d'écho en écho était arrivée jusqu'à mes confrères parisiens, qui en avaient ri aux larmes. Cette fois, ce même journaliste, ami et camarade de Jules Mayran, notre jeune ingénieur, tailla de nouveau sa plume des dimanches et écrivit l'article suivant: «Sursum! sursum! le grand œuvre de la décentralisation littéraire et artistique, scientifique et industrielle, fait chaque jour de nouveaux progrès. Déjà nous avons failli avoir cet hiver un opéra en deux actes, dont les paroles, la musique et les décors sont dus, comme on sait, à trois de nos compatriotes. Si cette solennité dramatique et musicale a été retardée, c'est que notre Laruette, engagé pour les
- 59. secondes basses-tailles, a cru devoir résilier son engagement, et que la chanteuse à roulades, idole de notre intelligent parterre, n'a pas voulu s'abaisser à chanter un rôle de Dugazon. Mais tout nous fait croire que ces légères difficultés seront levées pour la saison prochaine, et ce jour-là nos dilettanti n'auront plus rien à envier à la moderne Babylone. Espérons-le, grand Dieu! espérons-le! Nous avons vu paraître, ce printemps, chez notre libraire à la mode, un roman, la Bergère du Ventoux, écrit par un membre de notre Académie, et qui laisse bien loin derrière lui les productions indigestes des Balzac, des George Sand, des Dumas, aussi affligeantes pour la morale que pour le goût. Enfin nous savons tous qu'une des plus modestes communes de notre département, la commune de Gigondas, a, depuis un an, pour maire un écrivain distingué, M. Georges de Vernay, qui, chargé des palmes parisiennes, est venu en apporter le tribut à son pays natal. Il signe aujourd'hui les actes administratifs de cette même plume qui a signé tant de fines critiques et d'intéressantes nouvelles. Que dis-je? il prépare en ce moment à sa chère commune un bienfait qui doit attirer éternellement sur son nom les bénédictions de ses administrés. Secondé par un ingénieur habile de notre ville, M. Jules Mayran, il a fait construire une machine qui élèvera jusque sur le plateau du village une eau que, de temps immémorial, les malheureux habitants étaient obligés de venir chercher au bas de leur montagne. Ce magnifique travail est maintenant terminé. C'est dimanche prochain, 15 octobre, qu'aura lieu l'inauguration de cette belle œuvre de décentralisation aquatique. Une fête champêtre sera offerte à cette occasion par M. le maire, dont l'imagination poétique ménagera, nous en sommes sûrs, de charmantes surprises à ses visiteurs. Utile dulci! Nous présumons assez bien de nos lecteurs et de nos lectrices pour être certains que l'élite de notre fashion, les dames les plus haut placées, notre brillante jeunesse, nos plus éminents fonctionnaires, nos savants et nos artistes, se feront une fête de prendre leur part de cette splendide journée. Oui, nous répondrons tous à cet appel du talent descendu de ses sphères idéales pour devenir le bienfaiteur de l'humanité. Sursum! sursum!»
- 60. On le voit, si les grands acteurs de mélodrame font précéder leur entrée par un tremolo de violoncelles et de violons, l'entrée en fonctions de ma fontaine était aussi annoncée par une assez belle ritournelle. Le grand jour arrivé, je me levai avant l'aurore: la persistance du beau temps avait redoublé mes inquiétudes. Non-seulement il n'était pas tombé une goutte d'eau depuis six mois, mais le soleil d'août, attardé en plein octobre, donnait à la campagne un faux air d'Arabie Pétrée. Pas un nuage, pas un souffle d'air; le ciel était d'un bleu de turquoise, et le thermomètre marquait dix-huit degrés à sept heures du matin. Nous devions faire avec le mécanicien et ses ouvriers une répétition générale, afin d'être sûrs que notre prima donna—l'eau— ne manquerait pas sa réplique. En ce moment le fils Chapuzot,—c'est le nom du mécanicien,—jeune garçon de quatorze à quinze ans, accourut tout essoufflé, et, après m'avoir tiré par la manche de mon habit, il me dit à demi-voix en me prenant à part: —Nous n'avons que deux litres par seconde: il n'y a pas de quoi faire tourner la roue!... Avez-vous vu au théâtre, dans certaines pièces modernes, un caissier venir annoncer à son maître que sa maison est en faillite, au moment où s'allument les lustres du bal et où l'on entend le roulement des premières voitures? Ma situation était tout aussi tragique, et je sentis un horrible frisson courir de la racine de mes cheveux à la plante de mes pieds. Comment faire? Il était sept heures; mes invités devaient arriver à onze, et la fête commencer à midi. —Il faut que la roue tourne! m'écriai-je avec cette énergie du désespoir qui ne calcule pas ses paroles. —Mais, monsieur le maire, c'est impossible.
- 61. —Impossible, petit malheureux! Tu veux donc me déshonorer?... Écoute... qu'il y ait de l'eau jusqu'à ce soir, et puis... la sécheresse, la soif, le néant, la tombe. Demain n'existe pas pour les désespérés! Il n'y a pas assez d'eau, dis-tu, pour que la roue tourne toute seule?... eh bien! fais-la tourner... recrute tous les gamins du village; qu'ils s'y attellent à tour de rôle; je serai grand et généreux... promets-leur de l'argent, beaucoup d'argent... De l'eau à tout prix! sauve-moi du ridicule et de la honte: songe que j'attends dans quelques heures le préfet, le général et les plus belles dames de la ville... va... va!... Ah! s'il ne s'agissait que de livrer ma tête! Chapuzot s'inclina avec un sourire narquois et courut exécuter mes ordres. J'étais pâle; une sueur froide mouillait mes tempes; et cependant je fus beau de dissimulation stoïque; je me retournai vers mon adjoint et mes conseillers, et, couvrant mes douleurs d'un masque marmoréen, je leur dis: —Ce n'est rien, messieurs; tout va bien. Pendant les trois heures qui suivirent, ma fermeté ne se démentit pas un instant; mais j'enviai les jeunes Lacédémoniens, qui n'avaient à cacher qu'un renard dans leur poitrine. Nous assistâmes à une grand'messe en musique, qui mit tout le monde d'accord—excepté les chantres—pour remercier Dieu des bienfaits de cette journée. A la sortie, j'interrogeai du regard mon ami Chapuzot: il me fit signe que mes ordres s'exécutaient et que nos pompes vivantes s'étaient mises à l'ouvrage. Bientôt nous vîmes poindre les premières voitures, et, si j'avais pu, dans ce moment de crise, être accessible aux fumées de l'amour-propre, j'aurais eu lieu d'être satisfait. Évidemment Gigondas, sa fontaine et son maire avaient ce jour-là un succès de vogue. C'était en diminutif le tout Paris des premières représentations. Autorités, notabilités, beautés, élégances, tout affluait. Les plus jolies femmes du pays donnaient le bras à ses dignitaires les plus huppés. Elles furent d'une grâce charmante pour le critique changé en maire, que la plus lettrée de ces dames appela le loup devenu berger. Elles voulurent—notez ce
- 62. fait important—descendre, en se promenant, jusqu'à mon château, faire connaissance avec le salon, la salle à manger et la bibliothèque, situées au rez-de-chaussée. La table était dressée d'avance, et elles daignèrent approuver les nappes damassées, d'une éclatante blancheur, les fleurs et les fruits artistement groupés dans des vases de Chine, le vin de l'Hermitage dans des buires de Bohême. Puis elles se passèrent en minaudant mes livres de main en main, et admirèrent les reliures de Durut et de Bauzonnet, avec force compliments pour le propriétaire. Elles entrèrent ensuite au salon: l'une d'elles essaya le piano de Pleyel, qu'elle déclara excellent; et comme la chaleur allait croissant, mes belles visiteuses se débarrassèrent de leurs châles, de leurs écharpes, de leurs fourrures, de leurs mantelets, qu'elles déposèrent sur les divans. C'étaient des gazouillements joyeux, de frais sourires, d'aimables propos, auxquels, malgré tous mes efforts, je répondais avec une préoccupation visible qu'elles eurent la bonté d'attribuer aux fatigues administratives ou aux distractions poétiques. Midi approchait; nous remontâmes sur la place, qu'avait envahie une foule compacte. Les musiciens préludaient sur leurs instruments: la salle de bal, recouverte d'une tente, décorée de lauriers et de buis, attendait les danseurs. L'adjoint, le garde champêtre, le doyen de la fabrique, se tenaient près de la fontaine, où il ne manquait plus que de l'eau. C'était à ma danseuse que j'avais réservé l'honneur de tourner le robinet. Je voulus prouver que ma gloire ne m'avait pas fait oublier mon premier engagement, et je présentai galamment ma main gantée de blanc à mademoiselle Eugénie Blanchard, fille du percepteur des contributions. Le général et la préfète voulurent bien nous faire vis-à-vis. J'avais l'œil fixé sur l'horloge de la mairie, dont l'aiguille marquait midi moins deux minutes. Mon cœur palpitait; ma danseuse rougissait comme une pivoine. C'était un de ces instants solennels qui sont à la vie ordinaire ce que l'Himalaya est à nos collines. L'orchestre joua la chaîne des dames. Au moment où je battais un triomphant six-quatre devant la préfète, midi sonna. Je m'arrêtai
- 63. net; un long frémissement parcourut la foule: l'émotion, l'attente, le désir, l'enthousiasme étaient à leur zénith. Mademoiselle Eugénie, passée de l'écarlate au ponceau, s'approcha de la fontaine et tourna le robinet.... L'orchestre jouait déjà les premières mesures de l'air: Où peut-on être mieux qu'au sein de sa famille?... Rien ne coula. Rien! RIEN! RIEN! En ce moment, il me sembla que Shakspeare s'était trompé, et que Banquo s'appelait Desmousseaux de Givré. Un même cri, à grand'peine étouffé, vibra et mourut dans toutes ces poitrines. Mes courtisans se hâtèrent d'affirmer que l'eau n'avait pas eu le temps de monter et que nous allions la voir jaillir. L'adjoint se pencha sur le tuyau, et, y collant son oreille, il nous assura qu'il entendait distinctement le bouillonnement de l'eau qui montait. Je me penchai à mon tour, et j'entendis en effet quelque chose comme un bruit souterrain, pareil à celui que produit la pioche d'un mineur. Nous vécûmes encore cinq minutes sur ce bruit et sur cette espérance. Ces cinq minutes envolées, les visages s'allongèrent d'une façon effrayante. Il fallut bien convenir que ce bruit consolateur, au lieu de se rapprocher, s'éloignait. Dix autres minutes effleurèrent mon front brûlant de leurs ailes de plomb et blanchirent plusieurs mèches de mes cheveux. Je n'osais plus regarder autour de moi; ma main serrait convulsivement la main de ma danseuse, qui ne soufflait mot; je croyais lire ma honte inscrite sur toutes les figures. Un silence de glace avait succédé au joyeux murmure de la fête. L'orchestre se taisait; mes administrés étaient au désespoir, et mes invités réprimaient une forte envie de rire. Atterré, hébété, stupide, j'appelais tout bas une catastrophe, une révolution, une attaque d'apoplexie, un coup d'épée, un coup de tonnerre qui vînt rompre, fût-ce en m'écrasant, cette situation intolérable. Je fus exaucé: le coup de tonnerre demandé se personnifia dans ma servante, qui se précipita haletante sur la place, en criant: —Monsieur! Monsieur! il y a une fontaine dans votre salon!
- 64. A ces mots magiques, l'espèce d'enchantement qui nous tenait immobiles comme Bartholo dans le finale du Barbier de Séville cessa subitement. Nous descendîmes, nous roulâmes comme une avalanche au bas de la côte. Un poignant spectacle nous y attendait. Voici ce qui était arrivé. L'eau, aussi capricieuse que les nymphes et les naïades, ses mythologiques patronnes, avait déjoué traîtreusement les efforts de la science. Délogée du bassin où elle coulait depuis des siècles, violentée par une force motrice insuffisante, qui l'avait contrariée sans la dompter, elle s'était ouvert une issue, pendant que nous ajustions les tuyaux neufs destinés à la recevoir, et cette issue souterraine l'avait peu à peu conduite jusqu'au mur de mon rez-de- chaussée. Ce mur était vieux comme tout le reste de la maison: cependant l'irruption n'aurait pas été si soudaine, si les gamins du village, excités depuis le matin par mes ordres et par mes promesses, n'avaient tourné la roue avec une vigueur et un entrain dignes d'un meilleur sort. Cédant à cette impulsion énergique, mais s'obstinant à ne pas monter, l'eau avait suivi sa pente naturelle, et, élargissant une voie déjà frayée, elle était venue battre de sa masse poussée par le jeu des machines un mur lézardé. Quelques heures lui avaient suffi pour y faire sa trouée, et, par un redoublement d'ironie, à l'instant même où, d'après mon programme, elle devait jaillir dans la fontaine officielle, elle me donnait, à domicile, une représentation extraordinaire. La trouée s'était faite, à cinq pieds au- dessus du parquet, à travers une tapisserie des batailles d'Alexandre. Deux gravures, l'Entrée d'Henri IV à Paris et Atala, violemment décrochées, nageaient pêle-mêle avec les femmes de Darius. Le piano, les tables à jeu, renversés sens dessus dessous, ressemblaient à des noyés dont on n'aperçoit plus que les jambes. Les albums, les cahiers de musiques, les keepsakes, les tapis, les potiches, les cadres, les tentures, se confondaient dans un inexprimable chaos. De cette première station l'eau était arrivée dans la salle à manger et dans la bibliothèque, y exerçant des ravages plus cruels encore. Là où l'on avait salué, le matin, l'ordre,
- 65. l'arrangement et l'élégance, on ne voyait plus qu'une confusion inouïe, de tristes épaves flottant au gré de l'onde. Adieu mon beau linge, si religieusement soigné par ma pauvre Ursule! Adieu les fruits et les fleurs! Adieu les vases et les buires! Mon bon vin, échappé de ses bouteilles brisées, se mêlait à cette eau inhospitalière; mes dressoirs faisaient l'effet d'îles battues par la vague. Les jambons, les galantines, les volailles, le gibier, les soufflés, les compotes, les crèmes, prenaient un bain, côte à côte avec mes beaux livres et mes belles reliures. Mais, hélas! tout cela n'était rien encore, et j'aurais eu à me féliciter d'en être quitte à si bon marché. Les divans du salon avaient été renversés comme les autres meubles, et vous n'avez pas oublié que mes élégantes visiteuses y avaient déposé une partie de leur toilette, afin d'être plus lestes et plus champêtres. J'entendis de petits cris de douleur et de colère auprès desquels une condamnation capitale doit ressembler à un madrigal. «Grand Dieu! le mantelet de madame la préfète!—Ciel! le cachemire de madame la baronne!—Bonté divine! l'écharpe en dentelle de madame la marquise!—Maman, mon boa!—Maman, mon chapeau de paille d'Italie!»—Toutes ces merveilles d'élégance féminine nageaient ou se noyaient dans cette miniature du Déluge. Je n'ai plus gardé qu'un vague souvenir des moments qui suivirent. Je ne pensais plus, je ne sentais plus, je ne voyais plus. Ursule offrait une image de la statue du désespoir habillée de soie puce. J'avais de l'eau jusqu'à mi-jambe, et je ne m'en apercevais pas. Il me sembla que j'entendais des exclamations, des éclats de rire, puis mes invités demandant d'une voix brève leurs voitures, puis le bruit de ces voitures qui s'éloignaient. Il y avait là un médecin qui eut pitié de moi. Il me prit la main, me tâta le pouls, déclara que j'avais un violent accès de fièvre, donna ordre que l'on me hissât dans ma chambre, que l'on me fît mettre immédiatement au lit, que l'on me servît une potion calmante et qu'on fermât hermétiquement mes fenêtres. Ses ordres furent exécutés comme sur une machine inerte. Toutefois, comme le sens littéraire résiste chez moi aux plus terribles catastrophes, j'eus le temps, avant d'être emporté, d'ouïr les deux
- 66. mots suivants, qui furent comme l'oraison funèbre de mon programme: —On ne peut pas dire que M. le maire de Gigondas nous ait reçus sèchement, murmura le préfet. —C'est tout à fait une hospitalité d'homme de lettres, dit la Philaminte: chez lui la fontaine ne pouvait être qu'une fable.
- 67. XX COMME QUOI IL N'EST PAS NÉCESSAIRE POUR FAIRE UN FOUR, D'ÊTRE AUTEUR DRAMATIQUE Il me fallut, après cette catastrophe qui fit du bruit, quatre ou cinq mois pour me remettre le moral en équilibre. Quant aux avaries matérielles, elles ne sont pas encore réparées. Tout compte fait, et sans même compter l'immense déception administrative, il se trouva que le désastre absorbait au moins deux années de mon revenu. Nous nous promîmes, Ursule et moi, de redoubler d'économie. Le voyage en Italie fut ajourné jusqu'à la fusion définitive de l'élément piémontais et de l'élément napolitain, et le voyage en terre sainte jusqu'à la réconciliation radicale des Églises grecque et latine. Nous avions de la marge, et je commençais à me rasséréner, lorsque l'on vint m'annoncer que le four de la commune allait être vacant. Ce n'est pas une affaire sans importance que la direction du four communal. Il concentre, deux fois par semaine, la vie politique, intellectuelle et mondaine du village tout entier: il s'y débite, comme de juste, beaucoup de fagots; les commérages s'échauffent à cette température, et souvent des réputations de rosières ont été démolies entre deux fournées. Le boulanger ou fournier est un personnage considérable, presque un fonctionnaire: il dépend des
- 68. caprices de sa montre ou de son humeur de réveiller en sursaut, avant le chant du coq, la femme de l'adjoint, ou de brûler le gâteau à l'huile de la fille du marguillier. Il s'agissait donc de faire un bon choix qui réunît l'utile à l'agréable, et obtînt l'assentiment populaire; car je ne pouvais me dissimuler que, soit par suite de la mobilité proverbiale des masses ignorantes (en cela bien différentes des esprits cultivés), soit plutôt à cause de mes dernières mésaventures, ma popularité avait prodigieusement baissé. Or la voix publique me désignait unanimement, comme le plus digne, un jeune mitron de vingt à vingt et un ans, de la plus belle espérance, natif de Gigondas, mais ayant étudié à Avignon les secrets les plus délicats de la boulangerie. Ses parents étaient au nombre de mes administrés les plus pauvres: mais, justement fiers de leur fils qui ne devait pas manquer de donner du pain à sa famille, ils chuchotaient des paroles mystérieuses dont je n'ai compris le sens que plus tard. On me présenta le jeune homme qui s'appelait Hippolyte (familièrement Polyte), et que je n'avais pas vu depuis sa plus tendre enfance. C'était un beau garçon joufflu, haut en couleur, large d'épaules, ayant l'air heureux d'être au monde et enchanté de sa robuste personne; le type complet d'un Rodrigue de village pour qui tout Gigondas aurait eu les yeux de Chimène. Il me montra complaisamment ses bras musculeux, qui, sans doute, enfournaient son pain avec autant de grâce que Pourceaugnac en mettait à manger le sien. Fasciné par la superbe encolure et les façons victorieuses du beau Polyte, qui s'était fait escorter de toutes les commères de l'endroit, je lui annonçai que je le nommais fournier de la commune; il reçut cette faveur en homme à qui un refus ne semblait pas possible. «Voilà donc enfin, me disais-je, une affaire réglée sans encombre!» Bientôt, pourtant, je m'aperçus qu'Ursule était soucieuse. Elle avait avec le curé et avec la mère de Polyte de fréquentes conférences où paraissaient s'agiter de graves intérêts. Un jour que le curé dînait avec nous, je le vis faire un signe d'intelligence à ma sœur: puis il me prit à part, et me dit que le retour et le séjour de Polyte dans la paroisse l'inquiétait fort pour la partie la plus aimable, mais la plus
- 69. fragile de ses ouailles. Déjà il était moins content de sa congrégation; la veille, un dimanche à l'issue des vêpres, il avait vu trois ou quatre de ses plus vertueuses choristes rire et folâtrer avec le superbe mitron, qui les criblait de coups de poing dans le dos; ce qui est, comme on sait, la plus haute expression de la galanterie villageoise. Ce jeune homme était trop beau, trop déluré, trop séduisant: il rapportait au bercail quelque chose des civilisations dangereuses de la ville; bref, on redoutait un malheur, et si ce malheur arrivait, quel désespoir pour le curé! quel chagrin pour le maire! —Eh bien! dis-je gaiement, puisqu'il y a péril en la demeure, puisque Polyte est si redoutable, nous avons un moyen de neutraliser ce Lovelace: le voilà avec un état, un four et une petite maison que je lui loue pour rien: trouvons-lui une femme! Marions Polyte! —C'est ce que nous allions vous demander, mademoiselle votre sœur et moi, répliqua le curé un peu tranquillisé. Il était donc décidé que nous marierions Polyte. Avec qui? ce détail ne m'inquiétait guère: j'avais lieu de croire que le gaillard n'aurait que l'embarras du choix. Je lui en touchai quelques mots auxquels il répondit vaguement, mais d'un petit air guilleret et sournois qui me donnait beaucoup à penser. Pour le moment, l'essentiel, d'après Ursule et le curé, était de le piquer d'honneur, de le mettre au pied du mur matrimonial, en préparant d'avance le logement des deux époux; ce qui, en y ajoutant mes bontés, le four et les avantages personnels de Polyte, suffirait à faire de lui un des meilleurs partis du village. Ursule, en cette circonstance, se relâcha de sa parcimonie habituelle: on acheta du linge, une commode, un lit, une crédence; on fit recrépir au lait de chaux la chambre de l'escalier; le tout sur la cassette particulière du maire, qui, depuis longtemps, hélas! n'avait plus de cassette. Enfin, quand tout fut prêt, les draps pliés, les chemises marquées, les serviettes ourlées, les cloisons blanchies,
- 70. quand je croyais n'avoir plus qu'à jouir de mon ouvrage et à calculer intérieurement le nombre de blanches colombes arrachées aux pattes de ce ramier, une idée foudroyante me traversa de part en part: Polyte n'avait pas tiré à la conscription!... Je le fis venir, et lui dis avec une sévérité tout administrative: —Mais, malheureux! vous nous avez laissés faire des préparatifs qui me coûtent les yeux de la tête, et vous n'avez pas encore tiré au sort!... —C'est vrai, monsieur le maire, répondit-il en se dandinant; mais je suis bien tranquille: j'ai toujours eu du bonheur; je suis sûr de tirer le meilleur numéro de la classe.... D'ailleurs, ajouta-t-il finement, quand même je tirerais mauvais, tout le monde sait... qu'il dépend de monsieur le maire... de me faire exempter. Ici Polyte, malgré son aplomb, s'arrêta terrifié par l'expression de fureur qui se peignit tout à coup sur mon visage. Il faut savoir que les paysans du Midi, et probablement de toute la France, ont une superstition dont rien ne peut les guérir: c'est qu'il suffit d'avoir une certaine position sociale, d'occuper des fonctions quelconques, fût-ce les plus modestes, pour disposer arbitrairement de toutes les consciences administratives, chirurgicales et militaires, de qui dépend le sort des conscrits. J'ai beau me fâcher, m'emporter, sauter au plafond, rien n'y fait: les solliciteurs s'en vont bien convaincus que mon pouvoir est sans bornes, et que si je refuse de leur donner un petit coup de main, c'est faute de bonne volonté. Or, j'aimerais mieux, s'il le fallait absolument, commettre un vol à main armée ou croire au génie de M. de Pongerville, que tenter de faire réformer un conscrit aux dépens d'un autre, lequel pourrait avoir du malheur à la guerre ou à l'hôpital et laisser sa famille dans le désespoir ou la misère. Cette idée seule me fait frémir; aussi, toutes les fois qu'un de mes incorrigibles remet la question sur le tapis, je suis plus furieux que si l'on me lisait une tragédie. Je réussis pourtant à me contenir, pour ne pas trop compromettre ma dignité magistrale devant mon inférieur, et je dis froidement à Polyte:
- 71. —Vous avez donc des cas d'exemption? —Oui, monsieur le maire: un rhumatisme à la jambe gauche, un commencement d'anévrisme au cœur et la poitrine attaquée.... Notez que, dans son empressement, il était accouru en costume de four, et qu'à travers sa chemise entr'ouverte j'admirais un torse d'Hercule Farnèse. —Allez, mon ami, lui dis-je avec un calme très-mal joué, allez enfourner votre pain; quand le moment viendra, nous nous occuperons de vos infirmités. Le jour du tirage, Polyte se présenta devant l'urne, les épaules effacées et la bouche en cœur, comme un ténor qui va chanter son air. Hélas! son étoile lui fit faillite: il amena triomphalement le numéro deux. La consternation à Gigondas fut générale. Ce diable de Polyte était de ces gens qui ont, comme Létorières, la clef des cœurs: toutes les filles fondaient en larmes, comme si toutes avaient eu l'espoir de l'épouser. Leur douleur était aussi touchante que bavarde. Les parents du conscrit malheureux rôdaient sans cesse autour de moi, et recommençaient à l'envi ce duo mystérieux qui m'avait déjà si fort intrigué. On affectait de parler de mon crédit auprès du préfet, de mon ami le général, que je n'avais jamais vu. Les insinuations, les sollicitations, les prières, muettes ou formulées, m'arrivaient de toutes parts et sous toutes les formes. Il était clair que si je ne faisais rien pour tirer Polyte de ce mauvais pas, ma popularité, déjà fort en baisse, tomberait au-dessous de zéro. Pourtant je tenais bon, me bornant à répéter gravement que le drame se dénouerait le jour de la séance du conseil de révision. Ce jour fatal arriva, et le dénoûment fut tel que je l'avais prévu. Quand Polyte parut en costume de mitron du paradis terrestre, et que le conseil procéda à la révision de sa constitution, il y eut parmi ses juges un long murmure d'enthousiasme; je crus un moment que le général—un vieux de la vieille—allait se jeter sur lui comme un
- 72. ogre affamé de chair fraîche. Ce gracieux embonpoint, uni à cette riche musculature, plongea le chirurgien-major en extase. Aussi, lorsque Polyte essaya d'alléguer ses infirmités, l'admiration se changea en une explosion d'hilarité. Le rictus du lieutenant de gendarmerie s'ouvrit comme celui d'un crocodile, et le conseiller de préfecture fit un calembour. Le trop superbe numéro deux fut déclaré d'une voix unanime bon à partir. Mais il eut une compensation: on le proclama le plus bel homme de son canton, et le général lui affirma qu'avec un peu de protection il pourrait entrer dans les cent-gardes.
- 73. XXI Le lendemain de cette journée mémorable, Polyte entra chez moi de bon matin; il était cette fois en grande tenue, et sa figure exprimait une foule de sentiments complexes: —Monsieur le maire, me dit-il, si je suis obligé de partir, je manque ma fortune... —Votre fortune! répliquai-je, pas précisément... mais enfin nous aurions fait de notre mieux pour vous assurer les moyens de vivre honnêtement dans votre état. —Il s'agit bien de mon état! reprit-il avec un dédain magnifique; je veux parler de Lise Trinquier. —Lise Trinquier!... qu'est-ce que c'est que Lise Trinquier? —Lise Trinquier! vous ne connaissez pas Lise Trinquier? Mais c'est la fille du plus riche vétérinaire d'Avignon, proche voisin du boulanger chez qui j'étais apprenti... Lise a perdu sa mère, qui lui a laissé trente mille francs, déposés chez M. Girard, notaire, rue Banasterie. Son père vient de se remarier avec une femme de quarante-cinq ans, qui n'aura pas d'enfant; sa fille aura encore mieux de vingt-cinq mille francs de ce côté-là. Enfin, monsieur le maire, Lise a une tante... une vieille tante qui est sa marraine, qui l'aime comme sa fille, et dont elle sera l'unique héritière.... Cette tante, madame Cuminal, est immensément riche: elle possède une maison à
- 74. Montheux, un moulin, trois olivettes, un pré, un clos, un jardin potager; elle récolte, bon an, mal an, douze salmées de blé et quarante quintaux de garance... elle a une vigne, monsieur, et quelle vigne!... une vigne de deux hectares! —J'aimerais mieux que ce fût d'un hectare (du nectar), dis-je étourdiment, oubliant qu'un maire ne doit pas se permettre de paillettes. Polyte ne comprit pas: il était plongé jusqu'aux oreilles dans le Pactole de la tante Cuminal. —Enfin, poursuivit-il, sa fortune est évaluée à quatre-vingt mille francs; et tout cela sera pour sa nièce, pour Lise Trinquier! —Et Lise Trinquier est... —Folle de moi, fit Polyte en donnant à ces trois mots la valeur d'un long poëme. —Et on vous la donne, comme cela, tout uniment, sans que vous ayez à apporter autre chose que votre bonnet de coton? —Ah! pardon... on exige avant tout que je sois réformé ou... exonéré. Ceci méritait considération: on a vu des rois épouser des bergères; le roman nous a montré des filles de ducs et de marquis amoureuses de simples artisans. Pourquoi Polyte, me disais-je, ne serait-il pas adoré par Lise Trinquier? Évidemment les distances étaient moindres. D'une autre part, ce on me semblait un peu vague. Qu'était-ce, en réalité, que ce on? le père, la fille ou la tante? Séparément ou tous les trois ensemble? —Mon ami, dis-je à Polyte, je prendrai des renseignements, et s'ils me prouvent que vous m'avez dit la vérité... eh bien! nous verrons, nous aviserons.... Réformé, il n'y faut plus songer... exonéré, c'est un peu cher: deux mille cinq cents francs... et vous n'avez guère d'autres répondants que vos deux bras. Mais enfin, si réellement Lise
- 75. Trinquier vous aime, et si la tante Cuminal ne vous voit pas de trop mauvais œil, nous tâcherons d'arranger tout cela... Je n'ai certainement pas le cœur assez sec pour laisser un de mes conscrits manquer, faute d'un peu d'aide, ce parti californien. Cet adjectif si neuf (pour Gigondas) dépaysa un peu Polyte, qui ne s'en répandit pas moins en effusions de reconnaissance. Je me mis immédiatement en campagne, et averti par de pénibles expériences, je déployai cette fois tout le machiavélisme dont je me croyais pourvu. Mon vieux cheval tomba malade juste à point; je l'envoyai en pension chez Trinquier, le vétérinaire, afin d'avoir des intelligences dans la place; mes émissaires firent jaser les ouvriers et les voisins, et bientôt je sus, à n'en pas douter, que les renseignements fournis par Polyte étaient parfaitement exacts. Trinquier était riche; il avait eu de sa première femme une fille unique, qui s'appelait bien Lise, et à laquelle sa mère avait laissé, disait-on, une trentaine de mille francs. Je m'arrangeai pour voir moi-même Lise Trinquier au sortir de la messe: c'était une fille fort laide, très-brune et même passablement noire, dont les yeux, le teint, les sourcils abondants et la bouche ornée d'un commencement de moustache dénotaient le caractère inflammable. Mis en goût par ces premiers résultats, j'allai de ma personne à Montheux, le bourg habité par la tante Cuminal. Le percepteur des contributions me confirma tous les détails que Polyte m'avait donnés touchant les immeubles possédés par cette tante, qui passait à Montheux pour une marquise de Carabas. J'appris que Lise était en effet sa filleule et serait très-probablement son héritière. Enfin, je me transportai chez maître Girard, le notaire, que je connaissais de vieille date: il me répéta que les trente mille francs légués par la mère Trinquier et placés au cinq pour cent sur première hypothèque, seraient intégralement comptés à Lise le jour de son mariage. On le voit, tout s'ajustait admirablement au récit de Polyte. Cependant je ne fus pas satisfait: je voulais tout prévoir, tout calculer, n'avoir pas à me repentir plus tard de trop de précipitation et de confiance; je dis à Polyte:
- 76. —Mon garçon, tout cela est bel et bien: Lise existe, les chiffres sont exacts, la tante Cuminal a la physionomie de l'emploi; mais qui me garantit la nature du sentiment que vous avez inspiré à cette jeune fille? Est-ce une amourette, un caprice, une passion? Est-ce son cœur qui a parlé? est-ce seulement sa tête! Nous autres romanciers psychologistes, nous tenons grand compte de ces différences!... Polyte écarquilla de gros yeux, se demandant sans doute si je parlais turc ou iroquois. Puis sa face vermeille reprit son expression de contentement et de fatuité villageoise. Évidemment mes doutes l'humiliaient, non pas pour lui, mais pour moi et pour ma commune. Il gémissait d'avoir un maire aussi peu certain des moyens de séduction de ses administrés. —Monsieur, me dit-il enfin, c'est dimanche prochain le bal du Corps- Saint (quartier populaire d'Avignon). J'y serai, Lise y sera; vous pourrez la questionner vous-même: elle vous connaît (qui ne connaît pas M. le maire de Gigondas?); elle vous aime déjà comme mon bienfaiteur, et elle aura confiance en vous. Ces paroles, assez adroitement tournées, furent dites d'un ton de sécurité qui devait achever de me convaincre. Le dimanche, je ne manquai pas d'aller à ce bal, où dansaient gaiement toutes les grisettes et toutes les petites bourgeoises du quartier: Lise, en grande toilette, y figurait au premier rang; les galants affluaient; Polyte les dépassait de toute la tête, et les joues de sa danseuse, quand il battait devant elle un victorieux entrechat, offraient un heureux assemblage de coquelicot et de noir de fumée. Il me ménagea, entre deux quadrilles, une courte conversation avec elle; mais j'avais compté sans la pudeur et la timidité virginales. A toutes mes questions, insidieuses ou directes, Lise répondit par des monosyllabes dont un juge d'instruction aurait eu grand'peine à tirer parti. Aussi bien, pouvait-elle me répondre autrement? Ses yeux, tendrement fixés sur le beau Polyte, ne parlaient-ils pas pour elle? Lui demander davantage, n'était-ce pas méconnaître les susceptibilités féminines, attenter à une sensitive, porter une main brutale sur ces ailes de papillon qu'on appelle les rêves de jeune
- 77. fille, manquer en un mot à toutes les traditions de cette littérature des délicats, à laquelle j'avais eu un moment la prétention d'appartenir? Je me condamnai, pour ma pénitence, à venir en aide à Polyte. Mes renseignements n'étaient-ils pas complets? N'avais-je pas épuisé et même dépassé tout ce que pouvait exiger la plus minutieuse prudence? Je m'exécutai donc de bonne grâce. Trois jours après, j'empruntai, à l'insu de ma sœur, les deux mille cinq cents francs et je les comptai à Polyte, qui me fit un billet bien en règle sur un papier dont je payai le timbre. Je lui adressai, sur les conséquences formidables qu'aurait pour lui son insolvabilité, un speech qu'il écouta avec une scrupuleuse attention. Il m'appela son sauveur, emporta les rouleaux et s'en alla en sifflotant l'air de Fernand dans la Favorite. Quinze jours s'écoulèrent, puis six semaines, puis deux mois. Polyte continuait d'enfourner son pain à la satisfaction générale. Je profitai de notre première rencontre pour lui demander où en étaient ses préparatifs de mariage. —Ah! voilà... me dit-il d'un air un peu embarrassé; si la chose dépendait de Lise, ce serait déjà fait!... elle m'aime tant! ajouta-t-il en levant les yeux au ciel. Mais le père et la tante Cuminal ne veulent pas en entendre parler: ce sont des ambitieux, des orgueilleux, des vaniteux, qui me méprisent parce que je n'ai rien, et qui ont rêvé pour Lise un grand mariage: ils espèrent lui faire épouser le greffier Malingray... —Mais enfin le père Trinquier est remarié; sa fille a le bien de sa mère; elle est maîtresse de sa personne, et si elle vous aime véritablement... —Ah! c'est qu'elle est mineure, reprit Polyte en se grattant l'oreille, et... —Mineure, juste ciel! mais elle a de la barbe!... Je lui donnais vingt- trois ou vingt-quatre ans.
- 78. —Monsieur le maire, elle aura dix-huit ans aux prunes... —Aux prunes, grand Dieu!... Allons, j'ai fait une sottise; ce ne sera ni la première ni la dernière. Mais vous, petit malheureux, vous avez singulièrement abusé de ma confiance! Je ne voulus pas me tenir pour battu. La pureté de mes intentions, le désir de rattraper mes deux mille cinq cents francs, un certain goût de romanesque que j'avais gardé de ma vocation primitive, me donnèrent une hardiesse que je n'aurais jamais eue pour moi-même. Je demandai un rendez-vous à Lise Trinquier, et je l'obtins. J'interrogeai l'intéressante mineure avec un mélange d'autorité paternelle, de gravité municipale et de paradoxe sentimental. Ses réponses trahirent un défaut absolu d'énergie et d'initiative, et même, hélas! un certain penchant à sacrifier au Veau d'or, aux vanités de ce monde, à ce luxe effréné qui est la plaie de notre époque... Elle aimait bien Polyte, mais le greffier Malingray avait un joli pavillon à un demi-kilomètre de la ville, et il promettait de l'y conduire en voiture! Au reste, je n'eus pas le temps de m'abandonner aux réflexions mélancoliques que me suggérait cette nouvelle preuve de l'appauvrissement de l'esprit romanesque en France. A peine étions- nous ensemble, Lise et moi, depuis dix minutes, que la porte s'ouvrit avec fracas, et le père Trinquier parut, une énorme trique à la main... Rassurez-vous, mesdames, je dois ajouter bien vite que cette trique ne m'était point destinée. —Ah! monsieur le maire, me dit-il d'un ton où le respect et la colère se combinaient à des doses très-inégales, il est heureux pour vous que je ne sois pas aveugle; car je vous aurais tapé comme un sourd... Je croyais ma fille enfermée avec ce gueux de Polyte... Quant à vous, je vous respecte, parce qu'au fond vous n'êtes pas un méchant homme, et que, de père en fils, j'ai toujours ferré votre famille... mais vous faites-là un vilain métier. Vous qui avez mis le nez dans tous les livres, vous avez lu sans doute le Code pénal; vous savez, en cas de détournement de mineure, à quoi s'exposent les
- 79. complices... Je ne vous dis que ça.—Et toi, malheureuse, poursuivit-il en se tournant vers sa fille avec un geste de mélodrame, si tu ne veux pas que ce bâton te brise comme verre, tu vas me jurer devant Dieu et devant monsieur le maire de ne plus revoir ton infâme Polyte! —Oui, papa, oui, papa!... se hâta de répondre Lise en sanglotant. —Et d'épouser mon excellent ami, M. Simonin Malingray... Nouveaux sanglots. —Oui, papa, oui, papa... dit-elle enfin moins distinctement. Je compris que toute espérance était perdue, et je ne songeai plus qu'à sauver ma sortie. J'abaissai sur le père Trinquier un regard olympien; puis je dis à sa fille: —Mademoiselle, la poésie est morte, le roman se meurt; vivent les greffiers, et soyez heureuse!... Mais si jamais votre imagination avide d'idéal se débat, captive et meurtrie, dans les étreintes de la réalité; si jamais votre regard, un moment tourné vers les perspectives radieuses de l'infini, se reporte avec douleur sur l'étroit horizon d'un ménage vulgaire; si votre front, desséché par cette lourde atmosphère, appelle en vain des brises plus fraîches et plus douces; si votre cœur, rivé à sa chaîne, regrette les ardeurs et les délicatesses du véritable amour, souvenez-vous que vous avez fermé vous-même, à dix-huit ans, de vos mains fébriles, le livre à peine entr'ouvert du sentiment, de la rêverie, de l'enthousiasme et de la jeunesse! Souvenez-vous, mademoiselle, que vous aviez le goût du bonheur et que vous n'en avez pas eu le courage!!... Et je sortis majestueusement, laissant Lise et son père occupés à méditer le sens de mes paroles. Très-peu de temps après, Polyte s'arrachait les cheveux en apprenant le mariage de Lise avec M. Malingray, qui fit
- 80. magnifiquement les choses. La corbeille arriva tout droit de Paris, et le dîner de noces fut un des chefs-d'œuvre de Campé, ce cuisinier merveilleux qui a décentralisé la gastronomie. Cinq mois plus tard, je vis entrer dans mon salon le curé par une porte et Ursule par une autre; tous deux étaient pâles, mornes, effarés, suffoqués. Une horrible catastrophe se lisait d'avance dans leur attitude. —Ah! monsieur le maire, je vous l'avais bien dit, s'écria le digne homme, il faut marier Polyte, il le faut! Ce n'est plus seulement nécessaire, c'est urgent, très-urgent... —Très-urgent, répéta Ursule, les yeux baissés. —Marier Polyte? et avec qui? demandai-je. —Avec Madeleine Tournut, une de mes congréganistes, bredouilla le pauvre abbé en rougissant jusqu'aux oreilles. Madeleine Tournut était une assez jolie fille, mais pauvre comme le fut Job avant d'être duc. —Il le faut? —Il le faut. —Il le fallait, bégaya Ursule, qui, par cette variante, acheva d'éclaircir la situation. —Absolument? —Absolument. —Et promptement. Ces deux adverbes joints ne suffisaient pas pour servir de dot à Madeleine. Le jeune couple, riche d'amour, mais ne possédant pas d'autre richesse, fut marié gratis. Ursule, qui se reprochait sans
- 81. doute de ne pas avoir fait assez bonne garde, se punit aux dépens de sa bourse et de la mienne. Nous payâmes tout. Moyennant une indemnité annuelle dont je me reconnus débiteur envers la commune, j'assurai à Polyte pour dix ans la propriété de son four.—Quant à moi, mon four était complet.
- 82. XXII Ces trois épisodes peuvent vous donner une juste idée de mes succès administratifs et de mes économies municipales. Je pourrais encore vous en raconter huit ou dix du même genre; mais à quoi bon? Le cadre est trop étroit pour que les tableaux soient bien variés, et vous finiriez, mesdames, par me trouver très-ennuyeux si vous n'avez commencé par là: l'essentiel est de constater, en guise de moralité, que l'écharpe de maire ne m'a pas mieux réussi que la férule de critique: c'est que là-bas comme ici, à Paris comme au village, l'homme est toujours le même. Pour se gouverner à travers ses passions et ses vanités, il faut une habileté que je n'ai pas. Je m'étais brisé sur les récifs du boulevard Montmartre; j'ai échoué sur les écueils de ma pauvre commune de Gigondas. —Puissamment raisonné! dit M. Toupinel qui, malgré son tempérament sanguin, avait écouté ce long récit sans donner trop de marques d'impatience: mais, monsieur le maire ou monsieur le critique, il ne suffit pas d'être modeste; tout homme de lettres le serait autant que vous,—c'est une des qualités inhérentes à la profession,—il faut encore être clair et honnête; clair pour nous, pauvres Athéniens de Thèbes-la-Gaillarde, sur qui vos pseudonymes, à la la Bruyère ou par à-peu-près, produisent exactement l'effet de la lanterne magique du singe de Florian; honnête pour messieurs les Parisiens, qui, si vous publiez jamais vos Mémoires, ne manqueraient pas de vous accuser de ne pas avoir mis d'étiquette à vos
- 83. transparents. Entre nous qui ne comprenons pas assez et ceux qui comprendraient trop, vous n'avez qu'un moyen de tout concilier: c'est de nous donner, dès ce soir, le trousseau de clefs que vous avez sans doute dans votre poche... —Rien de plus juste, répliqua George de Vernay; ces diables de noms propres sont si terribles à manier, que je les ai momentanément ajustés à ma commodité particulière; mais, à présent, je suis à vos ordres; établissons, si vous le voulez, un dialogue par demandes et par réponses, comme dans le catéchisme: ce sera une sorte de table des matières... —Eh bien, attention! je commence:—Qui entendez-vous par Eutidème? —M. Jules Sandeau. —Et Théodecte? —M. Louis Veuillot. —Et Euphoriste? —M. Ernest Legouvé. —Et Iphicrate? —M. de Falloux. —Et Théonas? —Lacretelle. —Et Argyre? —M. Edmond About. —Et Colbach? —M. Louis Ulbach. —Et Porus Duclinquant? —M. Taxile Delord. —Et Clistorin? —Le docteur Véron. —Et Molossard? —M. Barbey d'Aurevilly.
- 84. —Et Schaunard? —Henry Mürger. —Et Caméléo? —M. Paulin Limayrac. —Et Marphise? —Madame Émile de Girardin, née Delphine Gay. —Et Lélia? —George Sand. (Alcade, saluez!) —Et Caritidès? —M. Sainte-Beuve. —Et Polycrate? —Gustave Planche. —Et Polychrome? —M. Théophile Gautier. —Et Bernier de Faux-Bissac? —M. Granier de Cassagnac. —Et Poisonnier? —M. Vivier. —Et Massimo? —M. Maxime du Camp. —Et Lorenzo? —M. Laurent Pichat. —Et Falconey? —Alfred de Musset. —Et Olympio? —M. Victor Hugo. —Et Julio? —M. Jules Janin. —Et Raphaël? —M. de Lamartine. —Et Bourimald? —M. Méry. —Et Hermagoras? —M. de Balzac.
- 85. —A la bonne heure! maintenant vous avez mon estime: reste à savoir si votre récit a ému la sensibilité de ces dames... On entoura, on applaudit, on plaignit George de Vernay; mais tout à coup, au milieu de cette ovation de province, une voix solennelle s'éleva pour protester: c'était celle de M. Margaret, vieux magistrat en retraite, qui passait pour le Nestor de la contrée: —Jeune homme! dit-il (George a cinquante ans), j'ai été intimement lié avec votre excellent père; ma vieille amitié vous a suivi, à votre insu, à travers toutes vos mésaventures parisiennes; et si j'ai, grâce à mon âge, mon franc parler avec tout le monde, ce n'est pas une raison pour que je vous épargne vos vérités. Rien, absolument rien, dans votre histoire, ne mérite l'intérêt qu'on vous témoigne. Tous vos malheurs viennent d'un défaut absolu de réflexion et de prévoyance, d'un manque d'équilibre intellectuel que je résume en ces termes: Vous aviez trop d'imagination pour un critique, pas assez pour un romancier: c'est pourquoi vous avez perpétuellement flotté entre vos impressions mobiles qui ôtaient à vos jugements littéraires toute solidité et toute fermeté, et vos lubies aristocratiques qui gâtaient à plaisir les créations de votre cerveau. Vous avez fait de la critique avec vos passions et du roman avec vos systèmes. Il en est résulté que vos appréciations des œuvres et des hommes ont sans cesse dépassé la mesure en bien ou en mal, et que vos fictions romanesques ont péri dans le faux et dans l'ennui. Vous, un critique! oh! que non pas! Il faut au critique de la gravité, et vous êtes léger; de la profondeur, et vous êtes superficiel; du savoir, et vous êtes ignorant; de l'Antiquité, et vous ne savez pas le latin!... —Oh! s'écria George avec un soubresaut, comme si on avait marché sur ses cors... —Non, vous ne le savez pas, reprit M. Margaret avec plus de force: Voyons! scandez-moi seulement ces trois mots: Urit fulgore suo!... —Urit, deux longues, bredouilla le patient, semblable à un aspirant au baccalauréat que son examinateur embarrasse; fulgo, deux
- 86. longues; re su, deux brèves; o, une longue; cet hémistiche ne peut entrer dans un hexamètre... —Et vous l'y avez mis, ignare que vous êtes! vous avez oublié, enim: Urit enim fulgore suo, ignorantus! —Ignoranta, ignorantum; dignus est intrare; cabricias arci thurum, Catalamus singulariter, exclama George pour se rattraper. —Oui, vous savez le latin de Molière; mais vous ne savez pas celui de Cicéron et de Virgile; voilà qui est dit!... —Mais j'ai eu, au concours général, un prix de vers latins, un prix de narration latine, un prix de discours latin et un prix de dissertation latine! —C'est possible; mais cela date de si loin! Moi aussi, j'ai dansé la gavotte, en 1807, comme Trénis; et aujourd'hui je ne saurais pas mettre un pied devant l'autre. Non, mon cher, vous n'êtes pas un critique; vous seriez tout au plus un causeur, si vous aviez su mener côte à côte vos défauts et vos qualités. Hélas! monsieur tranche du grand; monsieur a voulu se lancer dans le morceau d'apparat: ah! mon pauvre ami, qu'alliez-vous faire dans cette galère? Tenez, il y a dans vos volumes,—non pas, comme on l'a dit, en tête du premier, mais du quatrième—une grosse tartine philosophique et déclamatoire que je n'ai jamais pu digérer: cela s'appelle, je crois: la Littérature et les Honnêtes gens. Vilain titre, jeune homme, vilain titre! J'en ai vu un à peu près pareil, il y a quarante-trois ans, dans le Conservateur, qui n'a rien conservé du tout. Les Honnêtes gens! mais c'est donner à entendre qu'il y a des gens qui ne le sont pas; c'est médire de la société actuelle, qui du reste est au-dessus de semblables médisances. Vous avez, messieurs, de ces manières exclusives qui établissent des classes, des catégories, des camps, là où il ne devrait y avoir que de bons Français, appréciateurs éclairés des bonnes et belles choses. Ainsi vous dites encore: Nous autres catholiques. Quelle arrogance! mais tout le monde est catholique, excepté les protestants, les juifs et les Turcs; seulement, il y a ceux
- 87. qui vont à la messe, et ceux qui n'y vont pas; et ceux-là ont peut- être droit à plus d'égards que les autres: leur religion est en dedans, et vous n'êtes pas sans savoir que les sentiments contenus sont les plus vivaces. Votre titre était donc détestable, et vous en avez été cruellement puni. Grand Dieu! quel amphigouri! quel jargon métaphorique! «Telles sont les questions que je veux effleurer ici, comme on plante un jalon à l'entrée d'une route.»—Effleurer et planter dans la même phrase! Vraiment, vous méritez que je vous effleure la joue et que je vous plante là dès les premières lignes: ceci n'est rien. Voici qui enlève la paille: «Cette philosophie à la fois si destructive et si stérile, cette révolution si radicale et si impuissante, avaient montré l'homme réduit à lui-même dans un état de misère, de crime et de nudité: il ramenait sur sa poitrine les lambeaux de ses croyances, déchirées à tous les angles du chemin qui l'avait conduit des bosquets du paganisme-Pompadour aux marches de l'échafaud.» Ouf! ouf! ô Cathos! ô Madelon! ô Gali! ô Thomas! George baissait la tête, et j'ai su, depuis, qu'il était, sur ce malheureux morceau, si horriblement rempli de cartilages, tout à fait de l'avis de son critique: M. Toupinel vint à son secours: —Permettez, monsieur! dit-il au formidable octogénaire: est-il bien juste de prendre dans un ensemble de sept volumes le chapitre le plus mal réussi, et, dans ce chapitre, huit ou dix lignes qui, séparées du reste, n'en paraissent que plus boursouflées et plus grotesques? Quel ouvrage serait de force à résister à ce procédé? Voulez-vous un exemple? Je me souviens qu'en 1840 M. de Balzac se livra, vis-à-vis du premier volume de Port-Royal, de M. Sainte-Beuve, à un échenillage du même genre, et il fit rire tout Paris aux dépens de l'auteur et de l'œuvre. Et cependant l'œuvre a survécu, parce qu'elle est charmante, et aujourd'hui les mêmes gens de goût admirent à la fois Sainte-Beuve et Balzac: grande leçon, soit dit en passant, contre les querelles littéraires!... —Dont les gens de lettres ne profiteront pas, grommela entre ses dents M. Verbelin.
- 88. Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com