Iaetsd use of mobile communication in data-intensive

Use of Mobile Communication in Data-Intensive
Wireless Networks
Nuthanapati Jyothsna1
Reddy Sagar A C 2
Ravi Kumar G3
1
PG Student, Siddartha Institute of Science and Technology, Puttur chittoor(d) A.P, jyothsna_oct30@yahoo.com
2
Assistant Professor, Siddartha Institute of Science and Technology, Puttur chittoor(d) A.P, sagar.akkaru@gmail.com
3
Assistant Professor, Siddartha Institute of Science and Technology, Puttur chittoor(d) A.P, ravikumar.gunduru@gmail.com
Abstract: Wireless Networks are increasingly
used in different types of data-intensive
applications scenarios such as micro-weather
monitoring, meticulousness agriculture, and
audio/video observation. The sensor nodes are
minute and limited in power. Sensor types vary
according to the application of WN. Whatever be
the application the key challenge is to broadcast
all the data generated within an application’s life
span to the base station in the face of the fact that
sensor nodes have limited power supplies. The
concept of mobile communication is that the
mobile nodes change their locations so as to
minimize the total energy consumed by both
wireless transmission and locomotion. The
predictable methods, however, do not take into
account the energy level, and as a result they do
not always extend the network lifetime.
Keywords: Data-intensive; Energy;
communication; routing tree; WN
1. Introduction
Sensors have the capabilities of doing sensing,
data processing, and wirelessly transmitting
collected data back to base stations by way of
multiple-hop relay. Sensor itself supplies
necessary operation with limited battery energy.
Those operations that consume energy are
transmitting and receiving data, running
applications, measuring power, and even staying
in standby mode. Among others, data transmission
consumes the most energy. A wireless sensor
network (WSN) consists of spatially distributed
autonomous sensors to monitor physical or
environmental conditions,
such as temperature, sound , vibration , pressure
humidity, motion or pollutants and to
cooperatively pass their data through the network
to a main location[1,2]. The more modern
networks are bi-directional, also enabling control
of sensor activity. The development of wireless
sensor networks was
motivated by military applications such as
battlefield surveillance; today such networks are
used in many industrial and consumer
applications, such as industrial process monitoring
and control, machine health monitoring, and so
on. Figure 1 shows an example of a wireless
sensor network.
Figure 1: An example of Wireless Sensor Network
Recent advancement in mobile sensor platform
technology has been taken into attention that
mobile elements are utilized to improve the
WSN’s performances such as coverage,
connectivity, reliability and energy efficiency.
The concept of mobile relay is that the mobile
nodes change their locations so as to minimize
the total energy consumed by both wireless
transmission and locomotion. The conventional
methods, however, do not take into account the
energy level, and as a result they do not always
prolong the network lifetime.
Proceedings of International Conference on Developments in Engineering Research
ISBN NO : 378 - 26 - 13840 - 9
www.iaetsd.in
INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT
105

2. Related Work
Analyzing the three different approaches: Mobile
base stations, data mules and mobile relays. All
the three approaches use mobility to reduce
energy consumption in wireless networks.
A. Mobile Base Station
A mobile base station is a sensor node collects the
data by moving around the network from the
nodes [4]. In some work, in order to balance the
transmission load, all nodes are performing
multiple hop transmissions to the base station. The
goal is to rotate the nodes which are close to the
base station. Before the nodes suffer buffer
overflows, the base station computes the mobility
path to collect data from the visited nodes. Several
rendezvous based data collection algorithms are
proposed, where the mobile base station only
visits a selected set of nodes referred to as
rendezvous points within a deadline and the
rendezvous points buffer the data from sources.
High data traffic towards the base station is
always a threat to the networks life time [5]. The
battery life of the base station gets depleted very
quickly due to the sensor nodes which are located
near to the base station relay data for large part of
the network. The proposed solution includes the
mobility of the base station such that nodes
located near base station changes over time. All
the above approaches incur high latency due to the
low to moderate speed of mobile base stations.
Figure 2 shows Mobile base station
Figure 2: Mobile base station
B. Data Mules
Data mules are another form of base stations.
They gather data from the sensors and carry it to
the sink. The data mule collects the data by
visiting all the sources and then transmits it to the
static base station through the network. In order to
minimize the communication and mobility energy
Consumption the mobility paths are determined.
In paper [6] the author analyses an architecture
based on mobility to address the energy efficient
data collection problem in a sensor network. This
approach utilizes the mobile nodes as forwarding
agents. As a mobile node moves in close
propinquity to sensors, data is transmitted to the
mobile node for later dumps at the destination. In
the MULE architecture sensors transmit data only
over a short range that requires less transmission
power. However, latency is increased because a
sensor has to wait for a mule before its data can be
Delivered. Figure 3. The three tiers of the MULE
architecture. The Mule architecture has high
latency and this limits its applicability to real time
applications (although this can be mitigated by
collapsing the MULE and access point tiers). The
system requires sufficient mobility. For example,
mules may not arrive at a sensor or after picking
the data may not reach near an access-point to
deliver it. Also, data may be lost because of radio-
communication errors or mules crashing. To
improve data delivery, higher-level protocols need
to be incorporated in the MULE architecture. Data
mules also introduce large delays like base
stations since sensors have to wait for a mule to
pass by before initiating their transmission.
Figure 3: The three tiers of the MULE architecture
ISBN NO : 378 - 26 - 13840 - 9
www.iaetsd.in
106

C. Mobile Relay
In this approach, the network consists of three
nodes such as mobile relay nodes along with static
base station and data sources. To reduce the
transmission cost relay nodes do not transport data
rather it will move to different locations. We use
the mobile relay approach in this work. In [7]
author showed that an iterative mobility algorithm
where each relay node moves to the midpoint of
its neighbors converges on the optimal solution
for a single routing path This paper presents
mobility control scheme for improving
communication performance in WSN. The
objectives of the paper are
1) Analyze when controlled mobility can improve
fundamental networking performance metrics
such as power efficiency and robustness of
communications
2) Provide initial design for such networks.
Mobile nodes move to midpoint of the neighbors
only when movement is beneficial.
Unlike mobile base stations and data mules, our
approach reduces the energy consumption of both
mobility and transmission. Our approach also
relocates each mobile relay only once
immediately after deployment. The paper study
the energy optimization problem that accounts for
energy costs associated with both communication
and physical node movement. Unlike previous
mobile relay schemes the proposed solution
consider all possible locations as possible target
locations for a mobile node instead of just the
midpoint of its neighbors.
3. Proposed work
We use low-cost disposable mobile relays to
reduce the total energy consumption of data-
intensive WSNs. Different from mobile base
station or data mules, mobile relays do not
transport data; instead, they move to different
locations and then remain stationary to forward
data along the paths from the sources to the base
station. Thus, the communication delays can be
significantly reduced compared with using mobile
sinks or data mules. Moreover, each mobile node
performs a single relocation unlike other
approaches which require repeated relocations.
Figure 4 shows Proposed Network.
Figure4: Proposed Network
The network consists of mobile relay nodes along
with static base station and data sources. Relay
nodes do not transport data; instead, they move to
different locations to decrease the transmission
costs. We use the mobile relay approach in this
work. Goldenberg et al. [13] showed that an
iterative mobility algorithm where each relay node
moves to the midpoint of its neighbors converges
on the optimal solution for a single routing path.
However, they do not account for the cost of
moving the relay nodes. In mobile nodes decide
to move only when moving is beneficial, but the
only position considered is the midpoint of
neighbors.
The sink is the point of contact for users of the
sensor network. Each time the sink receives a
question from a user, it first translates the question
into multiple queries and then disseminates the
queries to the corresponding mobile relay, which
process the queries based on their data and return
the query results to the sink. The sink unifies the
query results from multiple storage nodes into the
final answer and sends it back to the user.
The source nodes in our problem formulation
serve as storage points which cache the data
gathered by other nodes and periodically transmit
to the sink, in response to user queries. Such a
ISBN NO : 378 - 26 - 13840 - 9
www.iaetsd.in
107

network architecture is consistent with the design
of storage centric sensor networks. Our problem
formulation also considers the initial positions of
nodes and the amount of data that needs to be
transmitted from each storage node to the sink.
We consider the sub problem of finding the
optimal positions of relay nodes for a routing tree
given that the topology is fixed. We assume the
topology is a directed tree in which the leaves are
sources and the root is the sink. We also assume
that separate messages cannot be compressed or
merged; that is, if two distinct messages of lengths
m1 and m2 use the same link (si, sj ) on the path
from a source to a sink, the total number of bits
that must traverse link (si, sj ) is m1 + m2.
a) Energy Optimization Framework
In this section, we formulate the problem of
Optimal Mobile Relay Configuration (OMRC) in
data-intensive WSNs. Unlike mobile base stations
and data mules, our OMRC problem considers the
energy consumption of both mobility and
transmission. The Optimal Mobile Relay
Configuration (OMRC) problem is challenging
because of the dependence of the solution on
multiple factors such as the routing tree topology
and the amount of data transferred through each
link. For example, when transferring little data,
the optimal configuration is to use only some
relay nodes at their original positions.
Assume the network consists of one source si −1,
one mobile relay node si and one sink si +1. Let
the original position of a node sj be oj = (pj , qj),
and let uj = (xj , yj) its final position in
configuration U. According to our energy models,
the total transmission and movement energy cost
incurred by the mobile relay node si is
ci(U) = k║ui – oi ║ + am + b║ui+1 – ui ║2
m
Now We need to compute a position ui for si that
minimizes Ci(U) assuming that ui−1 = oi−1 and
ui+1 = oi+1; that is, node si’s neighbors remain at
the same positions in the final configuration U.
We calculate position ui = (xi, yi) for node si by
finding the values for xi and yi where the partial
derivatives of the cost function Ci(U) with respect
to xi and yi become zero. Position Ui will be
toward the midpoint of positions ui−1 and ui+1.
The partial derivatives at xi and yi, respectively
are defined as follows:
δ Ci(U)
---------- = −2bm(xi+1 − xi) + 2bm(xi − xi−1)
δxi
(xi − pi)
+ k --------------------------------------
√ (xi - pi)2
+ (yi - qi)2
δ Ci(U)
---------- = −2bm(yi +1 −yi) + 2bm(yi − yi−1)
δ yi
(yi − qi)
+ k --------------------------------------
√ (xi - pi)2 + (yi - qi)2
and rms do not have to be defined. Do not use
abbreviations in the title or heads unless they are
unavoidable.
b) Tree Optimization Algorithm
In this section, we consider the sub problem of
finding the optimal positions of relay nodes for a
routing tree given that the topology is fixed. We
assume the topology is a directed tree in which the
leaves are sources and the root is the sink. We
also assume that separate messages cannot be
compressed or merged; that is, if two distinct
messages of lengths m1 and m2 use the same link
(si, sj ) on the path from a source to a sink, the
total number of bits that must traverse link(si, sj )
is m1 + m2. Let the network consists of multiple
sources, one relay node and one sink such that
data is transmitted from each source to the relay
node and then to the sink. We modify our solution
as follows. Let si be the mobile relay node, S(si)
the set of source nodes transmitting to si and sdi
the sink collecting nodes from si. The cost
incurred by si in this configuration U is:
ci(U) = k║ui – oi ║ + ami + bmi║ui+1 – ui ║2
Algorithm 1
procedure OPTIMALPOSITIONS(U0)
converged ← false;
j ← 0;
repeat
anymove ← false;
ISBN NO : 378 - 26 - 13840 - 9
www.iaetsd.in
108

j ← j + 1;
⊳ Start an even iteration followed by an odd
iteration
for idx = 2 to 3 do
for i = idx to n by 2 do
(uj
i ,moved) ← LOCALPOS(oi, S(si), sdi );
anymove ← anymove OR moved
end for
end for
converged ← NOT anymove
until converged
end procedure
Our algorithm starts by an odd/even labeling step
followed by a weighting step. To obtain consistent
labels for nodes, we start the labeling process
from the root using a breadth first traversal of the
tree. The root gets labeled as even. Each of its
children gets labeled as odd. Each subsequent
child is then given the opposite label of its parent.
We define mi, the weight of a node si, to be the
sum of message lengths over all paths passing
through si. This computation starts from the
sources or leaves of our routing tree. Initially, we
know mi = Mi for each source leaf node si. For
each intermediate node si, we compute its weight
as the sum of the weights of its children. Once
each node gets a weight and a label, we start our
iterative scheme. In odd iterations j, the algorithm
computes a position uj
i for each odd-labeled node
si that minimizes Ci(Uj) assuming that
uj
i −1 =uj−1
i −1 and uj
i+1 = uj−1
i+1 that is, node
si’s even numbered neighboring nodes remain in
place in configuration Uj . In even-numbered
iterations, the controller does the same for even-
labeled nodes. The algorithm behaves this way
because the optimization of uj
i requires a fixed
location for the child nodes and the parent of si.
By alternating between optimizing for odd and
even labeled nodes, the algorithm guarantees that
the node si is always making progress towards the
optimal position ui. Our iterative algorithm is
shown in algorithm1.
4. Conclusion
The main objective of this paper is energy
conservation which is holistic in that the total
energy consumed by both mobility of relays and
wireless transmissions is minimized, which is in
contrast to existing mobility approaches that only
minimize the transmission energy consumption.
The tradeoff in energy consumption between
mobility and transmission is exploited by
configuring the positions of mobile relays. We
develop two algorithms that iteratively refine the
configuration of mobile relays. The first improves
the tree topology by adding new nodes. It is not
guaranteed to find the optimal topology. The
second improves the routing tree by relocating
nodes without changing the tree topology. It
converges to the optimal node positions for the
given topology. Our algorithms have efficient
distributed implementations that require only
limited, localized synchronization.
References
[1] I.F. Akyildiz, W. Su, Y. Sankarasubramaniam,
and E.Cayirci, “A Survey on Sensor Networks,”
IEEE Comm.Magazine, vol. 40, no. 8, pp. 102-
114, Aug. 2002.
[2] Shio Kumar Singh, M.P. Singh, and D.K.
Singh,“Applications, Classifications, and
Selections of Routing Protocols for Wireless
Sensor Networks” International Journal of
Advanced Engineering Sciences and Technologies
(IJAEST), November 2010, vol. 1, issue no. 2, pp.
85-95.
[3] Ganesan, B. Greenstein, D. Perelyubskiy, D.
Estrin, and J.Heidemann, “An Evaluation of
Multi-Resolution Storage for Sensor Networks,”
Proc. First Int’l Conf. Embedded Networked
Sensor Systems (SenSys), 2003.
[4] Fatme El-Moukaddem, Eric Torng, Guoliang
Xing" Mobile Relay Configuration in Data-
intensive Wireless Sensor Networks" in IEEE
Transactions on Mobile computing, 2013.
[5] J. Luo and J.-P. Hubaux, Joint Mobility and
Routing for Lifetime Elongation in Wireless
Sensor Networks, in INFOCOM, 2005.
[6] S. Jain, R. Shah, W. Brunette, G. Borriello,
and S. Roy,Exploiting Mobility for energy Cient
Data Collection in Wireless Sensor Networks,
MONET,vol. 11, pp. 327339, 2006.
[7] K. Goldenberg, J. Lin, and A. S. Morse,
Towards Mobility as a Network Control
Primitive, in MobiHoc, 2004, pp. 163174.
[8] Tang and P. K. McKinley, Energy
Optimization Under Informed Mobility, IEEE
ISBN NO : 378 - 26 - 13840 - 9
www.iaetsd.in
109

Trans. Parallel Distrib. Syst., vol. 17,pp. 947962,
2006.
[9] Fatme El-Moukaddem, Eric Torng, and
Guoliang Xing, Member, IEEE "Mobile Relay
Configuration in Data- Intensive Wireless Sensor
Networks".
ISBN NO : 378 - 26 - 13840 - 9
www.iaetsd.in
110

Iaetsd use of mobile communication in data-intensive

More Related Content

What's hot (18)

Similar to Iaetsd use of mobile communication in data-intensive (20)

More from Iaetsd Iaetsd (20)

Iaetsd use of mobile communication in data-intensive