Data-Centric Routing Protocols in Wireless Sensor Network: A survey

Yousef et al. / Journal of Computing Technologies

Vol 2, Issue 8

ISSN 2278 – 3814

Data-Centric Routing Protocols in Wireless Sensor
Network: A survey
Yousef Emami1, Ali Habeeb2, Mohammad Jalil Piran3, Manijeh Keshtgari4
1

Department of Information Technology, Shiraz University of Technology, Shiraz, Iran
Yousef.emami@ieee.org
2
College of Engineering, JNT University, Hyderabad, India
Ali_hb@ymail.com
3
Department of Electronics and Radio Engineering, Kyung Hee University, South Korea
piran@khu.ac.kr
4
Department of Information Technology,Shiraz University of Technology, Shiraz, Iran
keshtgari@sutech.ac.ir

Abstract- Energy consumption is one of the most important
factors in designing Wireless Sensor Networks which can not be
neglected. The required energy for transmission of one KB of
data is equal to that of processing of 3,000,000 instructions.
Therefore, to prolong the life time of wireless sensor networks,
energy consumption should be efficient and the efficiency of
energy consumption depends on reducing the number of
transmissions. Optimal routing techniques manage the number
of transmissions within networks. Among different routing
algorithms classifications, data-centric routing algorithms do
not use ID hence overhead will be reduced and energy saved. In
this paper, we outline challenges of routing in wireless sensor
networks and survey state-of-the-art in research on data-centric
routing protocols in wireless sensor networks.
Keywords -wireless sensor network; energy consumption;
routing protocols; data-centric routing protocol.

I. Introduction
Routing in wireless sensor network (WSN) is divided in
four categories: data-centric protocols, hierarchical protocols,
geographical protocols, Quality of service (QoS) based
protocols. In WSN, because of high density and overhead it
is not feasible to assign a global identification to a node;
therefore address-based routing protocols cannot be applied
in WSN and data-centric routing protocols have been
proposed. In data-centric routing instead of ID, attributebased naming is used, in attribute-based naming data related
to a particular node are not desired and data related to
attribute of the phenomenon are desired [6]. In data-centric
routing, the sink which is responsible for gathering data and
sending to the base station, issues a query for finding target
data stored in the other nodes of WSN. Data-centric routing
protocols are composed of two phases: route discovery and

© 2013 JCT JOURNALS. ALL RIGHTS RESERVED

communication .In route discovery phase, the optimal route
between sink and the target node is established and in
communication phase data is transferred from the destination
to the sink [7].
In this paper, data-centric routing protocols are
investigated. We aim to update the reader with the current
status of research in this realm. The rest of this paper is
organized as follows: in section II, we state challenges for
routing, section III mainly describe data-centric routing
protocols, and finally section IV concludes the paper.
II. Challenges for Routing
Since WSNs have their own challenges utilizing
traditional routing protocol is not workable .In following
these challenges are discussed:

A. Energy Consumption
The main concern in developing routing protocols for
WSNs is energy consumption. Due to limited energy
resource, data shall be delivered in an energy-efficient
manner. Thus, Conventional routing protocols are not
suitable.

B. Scalability
Scalable routing protocol can expand to support
increasing workloads. To provide scalability in WSN,
distributed protocols are needed. Due to high density of
nodes in WSN, full image of topology cannot be obtained in
a node; therefore distributed protocols which rely on a
limited knowledge of topology are preferred.

27


C. Addressing
Regarding to the high number of sensor nodes in WSN
assigning a unique address to each node is not viable.
Therefore, address-based routing protocols are not suitable
for this type of networks. In addition, in WSN, information
from a collection of sensors are preferred over the
information from individual sensor nodes. Addressing
mechanisms which do not rely on unique ID are used.

Vol 2, Issue 8

ISSN 2278 – 3814

route discovery algorithm, despite of these advantages it
suffers from the followings [6]:
Implosion: A node is likely to receive a packet from many
nodes, thus due to extra transmit-receive operation the
resources are wasted.
Overlap: It is possible for two nodes to sense an in common
region and then each one send the sensed data to their
common neighbors, thus the information are sent twice to the
neighbor node [8].

D. Robustness
There is no dedicated router in WSNs; consequently
routing protocols operate on sensor nodes. Regarding the
high probability of node failure in WSNs, it is intrinsic for
routing protocols to provide robustness to node failure.

E. Topology
WSN topology is deemed to be static, but failing nodes
can alter the topology especially when failed node is an
articulation point. Routing protocol should be aware of
topology so that the best routing decision can be taken.

F. Application
To select a right routing protocol, the type of application
shall be considered. For instance, static routes are good
option for monitoring application since information are sent
to the sink in periodic manner. Totally, different type of
applications requires different type of routing protocols [6].
III. Data-Centric Routing Protocols
In this part, data-centric protocols are discussed as
follow: Flooding, Directed Flooding, Constrained Flooding
Gossiping,Fuzzy-Gossip, LGossiping, ELGossiping, FELGossiping,
SGDF,Flossiping, Rumor, Spin, SPMS,M-Spin, EEM-SPIN,
Directed diffusion, Energy-efficient directed diffusion ,Mobile Sink
Directed Diffusion, COUGAR, ACQUIRE .

A. FLOODING
This algorithm is an old and simple one, it works as
follows: the received packet by a node is broadcasted for its
neighbors, and this process is repeated until all nodes in the
network receive a copy of the packet. No need for a node to
keep information about its neighbors and utilize complex


Resource blindness: the amount of available energy of a node
is not considered in this algorithm; consequently the behavior
is not adopted according the amount of available energy [6].
1)

Directed Flooding:

In this protocol, data are sent in a specific directional
virtual aperture instead of broadcasting. To be specific, the
information related to the current transmission virtual
aperture (it represents which neighbor nodes are allowed to
receive a packet sent from this node) and X and Y
coordinates are put in the corresponding fields and then the
constructed packet is sent. Whenever one of the neighbors
receive the packet, the transmission virtual aperture of the
packet’s sender is determined, if both are in the same
transmission virtual aperture the packet is relayed otherwise
it is discarded. This protocol, consumes less energy while
maintaining high levels of fault-tolerance [9].

2)

Constrained Flooding

Constrained Flooding is an energy efficient and robust
version of the flooding. Because of utilizing real time
reinforcement learning technique: initial flooding from the
sink is not required; the change in network does not result in
flooding from the sink. In this protocol, as long as the
packet has not reached to its destination, first, the cost and
temperature values of the packet are updated; temperature
controls the number of transmissions and energy efficiency.
Next, the cost difference of node and received packet is
computed; to retransmit the packet the obtained value must
be less than temperature. Then, to avoid collision, based on
the cost difference, random delay is added to forwarding
packets. Last, a probabilistic policy is utilized to arbitrate
retransmission; this step is special to the constrained
flooding.. Probabilistic policy controls number of
transmitted packets and energy costs. If the same packet is
frequently heard, it is less likely the packet will be
retransmitted [10].

28


B. Gossiping
The main objective of this protocol is to address the
implosion deficiency in flooding. To address this problem,
when a node receives the packet, unlike flooding, the packet
is not broadcasted, instead one of the neighbor nodes is
randomly selected and the packet is forwarded to it. When
the neighbor node receives the packet, this trend is repeated
until the packet reaches to the destination. Compared to the
flooding, energy consumption is less but latency in
propagating the message is high [6].

1) Fuzzy-Gossip :
This protocol is a modification of the gossip protocol
using fuzzy logic. In this protocol, when an event is detected
by a node, to determine the optimal neighbor for relaying
data toward sink the fuzzy logic is utilized. The goal of fuzzy
logic is to determine the best neighbor regarding residual
energy and distance to the sink. Hence the residual energy
and distance to the sink for each neighbor of node are
selected as inputs to fuzzification then the fuzzified values
are processed by the inference engine and finally
defuzzification output is selected as fuzzy value .Node with
the highest fuzzy value is selected as next hope and data is
relayed to it [1].

2) Location-based gossiping (LGossiping ):
In location-based gossiping, global positioning system
is utilized to relay data in the WSN. When a node has data for
transmitting, an in range neighbor is randomly selected and
data is transmitted to it, this trend is repeated until data reach
to the sink. In LGossiping data can be sent to one of the
known neighbors in reliable manner using position
knowledge of the nodes whereas in Gossiping data is blindly
sent to one of the neighbors which might be far from source
and thus data to be lost .LGossiping is a reliable version of
Gossiping [11].

Vol 2, Issue 8

ISSN 2278 – 3814

This protocol, aims to prolong network life time by
choosing a node with maximum residual energy and less hop
count to the sink. In the beginning, the hello messages are
broadcasted by sink for its neighbors. When a hello message
is received by a neighbor, the value of hop count variable is
stored in the memory and before rebroadcasting the message
to its neighbors the value of hop count one unit is
incremented. This process is repeated until all nodes receive a
copy of the hello message. At the end, each node is cognizant
of its distance to the sink. When an event is detected by a
node a request message is sent to its neighbor. Each neighbor
node which receives the message sends back its information
to the source. Regarding these gathered data, two nodes with
minimum hop count to the sink are selected, and then node
with maximum residual energy is selected and data is relayed
for it [3].

5) Single Gossiping with Directional Flooding (SGDF)
SGDF is divided into network topology and routing
scheme phases. In network topology phase, the sink
broadcast the hello message contained the hop-count and
threshold values, upon receiving the hello message by the
immediate neighbors, the hop-count value of the hello
message is saved and incremented then hello message is
rebroadcasted until all sensors receive one copy of the hello
message. In routing scheme ,prior to sending data by a node
a request message is broadcasted to its neighbors, all
neighbors send back their distance to the sink ,then to select
the next hop one of the neighbors with smaller hop-count
is randomly selected .No such neighbor, one with the equal
hop-count is randomly selected if not last chance is given to
the neighbors with larger hop-count and one of them is
blindly chosen. After, the next hop determination, source
transmits the packet to the destination in broadcast.
Destination receive the packet and relay it, simultaneously the
other neighbors listening to the channel generate a random
number between 0 and 1 ,if the generated number is less than
threshold for a special neighbor, it receives the packet and
starts directed flooding [12].

C. Flossiping
3) ELGossiping
In ELGossiping, an additional parameter called distance
is considered. when a node has data for sending, a node in
its transmission radius with lowest distance to sink is
selected for relaying data toward sink. This trend is repeated
until data reach to the sink [5].

4) Fair efficient location-based gossiping (FELGossiping):


The ingredients of flossiping are gossiping and lowprobability random selective relaying (LPRSR).When a
node has a packet to send, along with other fields it sets the
threshold and mode fields of the packet and broadcasts
the packet to its next hop neighbor in gossip mode
,meanwhile, other neighbors of next hop neighbor receive
the packet and start LPRSP process ,each one generate a
random number ,if generated number is less than specified

29


Vol 2, Issue 8

threshold in packet header ,packet is flooded by a specified
TTL otherwise packet is gossiped [13].

3) SPIN for broadcast networks (SPIN-BC)::

D. Rumor Routing
In rumor routing, in addition to neighbor list an event
table is maintained by each node .When an event occurs,
event is added to event table and an agent is generated. The
data related to the occurred event is propagated by generated
agent, when agent arrives in a node, event table of node can
be updated by the contents of agent. When life cycle of agent
is finished the path leading event is created. These agentgenerated paths are used to route queries. When a query is
generated by a node for a particular event, if a route is
available for target event in event table, query is forwarded to
the neighbor indicated by the event’s table entry, otherwise a
random neighbor is selected and the query is passed to it [8].

E. Sensor Protocols for Information via Negotiation (SPIN):
In spin, negotiation and resource adaptation are utilized
to address the deficiencies in flooding. Negotiation is utilized
for handling implosion and overlap. Spin nodes negotiate
with their neighbors before transmitting data, therefore
unnecessary communications are avoided. To handle
resource blindness, each spin node uses an energy manager to
monitor its energy resource. Hence energy-aware decisions
can be taken [6,8].

1) point-to-point SPIN protocol (SPIN-PP):
In SPIN-PP, whenever a node intends to transmit data, an
ADV packet is formed which contains a description of
transmitting data and then the ADV packet is transmitted to
the neighbor nodes .Upon a neighbor receive the packet, it
checks whether it has already received the descripted data .If
not, it replies back with a REQ message to show its interest
in receiving the descripted data. Finally, Data packet is sent
to the node requested it[8]. SPIN-PP does not address the
resource-blindness problem of flooding and just address
implosion and overlap problem. The next variation addresses
the resource-blindness problem [6].

ISSN 2278 – 3814

First.in SPIN-PP, many nodes can send REQ message
and each one receive a DATA packet individually hence,
resource are wasted. Second, when many nodes send REQ
message it may collision occurs, SPIN-PP does not prevent
from this issue. The mentioned issues are addressed by SPINBC. In SPIN-BC, whenever a message is sent by a sender all
nodes within the sender’s radio range will receive a copy of
that message, thus resource are not wasted . Randomized
backoff mechanism is utilized in SPIN-BC for the nodes
before transmitting a REQ packet so that collision to be
avoided [6, 8].

3) SPIN with reliability (SPIN-RL):
SPIN-RL is a reliable version of SPIN-BC , if a node
receive an ADV packet but does not receive DATA packet
,it can solicits the DATA packet from the neighbors that
already have that DATA packet. In SPIN-RL the
retransmission period of the nodes is limited [6, 8].

4) Shortest path minded SPIN (SPMS):
In spin all packets are transmitted at the same power
level and the power level is not adjusted according to
distance parameter. This problem is addressed by SPMS. In
SPMS, the transmitting power level is adjusted based on
distance, in other word, variable transmission power levels
are used. Because of high density in wireless sensor network,
the size of routing table is large therefore maintaining routing
table is not feasible and new solution is required. In SPMS, a
zone is defined as area which covered by a node in maximum
power level. The entries of routing table for each node are
restricted to the zone members. To compute the routes for
each zone, distributed bellman ford algorithm is utilized.
After filling routing tables, data transmission is started. When
a node has data for transmitting, it broadcasts ADV packets
to its zone member. Upon receiving ADV packet by a zone
member, if data described by ADV packet is required a REQ
packet is sent back to the source and data transmission is
initiated [14].

2) SPIN with energy consumption awareness(SPIN-EC)
6.) Modified SPIN (M-SPIN):

The resource-blindness problem of flooding is addressed
by SPIN-EC. In SPIN-EC, the residual energy of a node
before sending a REQ message is checked, if it is higher than
a threshold the operation is continued like SPIN-PP
otherwise the operation is cancelled and node does not
participate in it [6].


In applications, such as alarm monitoring, it is
intrinsic to receive a quick response, this protocol is suitable
for this kind of applications .To achieve a quick response, a
new phase called as distance discovery has been added to the
normal phases of spin protocols. In distance discovery, the
distance of each node from sink is measured and the

30


measured distance is used as a parameter in negotiation
phase. To measure the hop distance for each node, startup
message is broadcasted by sink .Upon receiving startup
message by a node, hop value of startup message is stored as
hop distance and hop value one unit is incremented and then
startup message is rebroadcasted to its neighbor. In the end of
distance discovery phase, each node knows its hop distance
from sink and consequently, negotiation phase is started .In
negotiation phase, whenever an ADV message is received by
a node, the following actions are taken:
1. Node checks whether it has already received the associated
data with ADV or requested it.
2. It replies with REQ message if its hop value is less than
that of ADV’s sender, otherwise request is rejected.
Energy efficiency is achieved with this protocol but some
nodes are frequently used and there is possibility to die
earlier [4].
7) Energy Enhanced M-SPIN (EEM-SPIN) EEM-SPIN:
In M-SPIN, some nodes are frequently used consequently
these nodes may be destroyed earlier than other nodes.
Energy enhanced M-SPIN has been proposed to address this
deficiency. In EEM-SPIN, clusters are used for solving the
problem. For cluster formation, and the election of cluster
heads the weighted cluster algorithm is used. In clusterbased structures, resources are used efficiently [2].
F. Directed Diffusion
The family of directed diffusion has three members: two
phase pull, one phase pull, and push. The involved steps in
simple directed diffusion or two phase pull are: 1. Interest
propagation, 2. Gradient setup, 3. Reinforcement, 4. Data
delivery
To initiate directed diffusion, interest messages are diffused
by sink to its neighbor in interest propagation phase, and this
process is repeated until all sensors receive a copy of interest
message [8].The received interest message is stored by each
sensor in interest cache. The required data from sensor
network is specified in interest messages. A node becomes a
source if it has data matching interest message. In gradient
setup phase, each source establish a gradient (a reply link
toward the neighbor which the interest message was
received) toward sink it may many gradients to be established
for the same interest in this case it is the task of
reinforcement phase to reinforce a particular path with best
link quality and lowest delay by resending the interest
message. Finally, the route between source and sink is
established and the data delivery phase is started. In push


Vol 2, Issue 8

ISSN 2278 – 3814

diffusion, interest propagation phase is ignored and data is
advertised by the sensor to sink. Upon receiving the
advertisements, the sink sends reinforcement packets for
source so that routes to be established [6].
1)

An energy efficient directed diffusion:

As mentioned, in reinforcement phase of directed
diffusion the path with lowest delay and best quality is
selected and reinforced. The selected path is not the path with
the minimum energy consumption and the residual energy of
the nodes on the path is not equal. Whenever source receives
data from its neighbors, it computes the cost function for
each path to its neighbor, then it compares the computed cost
functions and the minimum cost function is selected and the
related path is reinforced. The cost function is based on the
following factor:
Total communication cost of the path, Average residual
energy of the nodes on the path, Minimum node energy on
the path and Node connectivity. The mentioned factors are
emphasized based on application .Since in selected path
nodes have a balanced residual energy; it is less likely a node
with low residual energy cause failure in the path. Compared
to direct diffusion less energy is consumed and the network
has a longer lifetime [15].
2)

Mobile Sink Directed Diffusion:

MSDD is based on two-phase pull directed diffusion
algorithm .The idea in MSDD is to support sink mobility by
utilizing unused routes from source to sink, established in
the second phase of two-phase pull. This protocol is
composed of route completion and re-routing .In route
completion, to follow the new position of sink, the
established path in directed diffusion is extended. To extend
the available route towards sink, a relay node is used. A
relay node is used to monitor sink movements. Upon
detecting a sink movement the required route towards the
new position of sink is completed. Relay node broadcasts
tagged packets with hop-count called as exploratory data
packet within two hops from itself in periodic manner.
When sink receives these packets, it sends back another
tagged packet with hop-count called as
positive
reinforcement packet. If relay node, receive positive
reinforcement packet in its first hop, it means there is no
sink mobility and exploratory data packets are
rebroadcasted, If relay node, receive positive reinforcement
packet in its second hop there is sink mobility therefore
relay node becomes an ordinary one and the node received
positive reinforcement packet in its first hop becomes new
relay node and it broadcasts exploratory data packet. In rerouting phase, constructed path is replaced by new one [16].

31


G.

COUGAR

In cougar a new layer called as query proxy is defined
which interacts with network and application layers. This
layer is responsible for registering the query, creating a local
operator tree and so on . To comprehend cougar, suppose a
typical query is available, the query optimizer which is
located on gateway node optimizes the query and generates a
query plan. The data flow between sensors and computation
plan at each sensor and procedure for determining leader are
specified in query plan. Leader is a node with high remaining
energy and a center for doing computations. The query plan
is disseminated to the query proxy of related sensor nodes,
each participant non-leader node in the query reads its sensed
data and aggregates it with the received data from other
nodes and sends it toward leader. Leader aggregates received
data from non-leader nodes and under specified conditions
send it for gateway [17].

H. ACQUIRE
In ACQUIRE a novel approach for resolving query is
utilized. In traditional approaches, Query is disseminated by
the sink and response is gathered. The clear point in this
approach is distinction between query dissemination and
response generation. ACQUIRE aims to integrate query
dissemination and response generation. As the query
generated by the sink is forwarded in network the query is
resolved by each recipient node and forwarded to next node
until the query is fully solved and the complete response to
be found. In this step response is routed back toward sink
[18,19,20,21].
IV. Conclusion
In this paper, we investigated a comprehensive list of
data-centric protocols. There are many issues need to be
addressed by researchers e.g.: energy efficiency and life
time. Great advantages are achieved because of not using ID
instead a general question regarding a special phenomenon to
be asked and a response to be gathered. The best algorithm
which an implement this idea is ACQUIRE and is strongly
recommended to be used in future.
References:
[1] Imad S. AlShawi, Lianshan Yan, Wei Pan, Bin Luo,“A
Fuzzy-Gossip routing protocol for an energy efficient
wireless sensor networks,” Journal of Sensors IEEE,
pp. 1-4, 28-31 Oct. 2012.


Vol 2, Issue 8

ISSN 2278 – 3814

[2] M. Kavitha, K. Karthikeyan,” Comparison Of Data
Centric Protocols For WSN And Energy Enhanced MSPIN (EEM-SPIN),” International Journal of
Engineering Research & Technology (IJERT), vol. 1
,issue 9,November 2012.
[3] Ali Norouzi, Faezeh Sadat Babamir, Abdul Halim
Zaim, “A Novel Energy Efficient Routing Protocol in
Wireless Sensor Networks,” Journal of Wireless
Sensor Network, vol. 3, pp. 341-350, 2011.
[4] Zeenat Rehena, Sarbani Roy, Nandini Mukherjee,” A
Modified SPIN for Wireless Sensor Networks,” In
Third International Conference on Communication
Systems and Networks (COMSNETS), Bangalore
,India, PP. 1-4, 4-8 Jan. 2011.
[5] Ali
Norouzi,
Mehdi
Dabbaghian,
Alireza
Hatamizadeh, Burak Berk Ustundag, ” An improved
ELGossiping data distribution technique with
emphasis on reliability and resource constraints in
wireless sensor network,” In International Conference
on Electronic Computer Technology (ICECT), Kuala
Lumpur ,pp. 179-183, 7-10 May 2010.
[6] Ian F. Akyildiz, Mehmet Can Vuran, "Wirless sensor
networks,” john Wiley & Sons Ltd, pp. 139-148, 2010.
[7] Huilong Huang, J.H.H., Terril N. Hurst, ”DataCentric
Routing in Sensor Network using Biased Walk,” In
2006 3rdAnnual IEEE Communications Society on
Sensor and Adhoc Communications and Networks,
Reston, VA,vol.1 ,pp. 1-9, 28-28 Sept. 2006.
[8] Waltenegus Dargie, C.P.,” Fundamental of Wireless
Sensor Networks,” john Wiley & Sons Ltd, pp. 168175, 2010.
[9] R. Farivar, M.F., S.G.Miremadi,” Directed flooding: a
fault-tolerant routing protocol for wireless sensor
networks,”In Systems Communications Proceedings,
pp. 395- 399 , 14-17 Aug. 2005.
[10] Ying Zhang and Markus Fromherz, ”Constrained
flooding: a robust and efficient routing framework for
wireless sensor networks,” In 20th International
Conference on Advanced Information Networking and
Applications, vol. 1, 18-20 April 2006.
[11] S. Kheiri,MB. Ghaznavi Goushchi,M. Rafiee,B.
Seyfe, ” An Improved Gossiping Data Distribution
Technique with Emphasis on Reliability and Resource
Constraints,” In
International Conference on
Communications and Mobile Computing , Yunnan,
vol. 2, pp. 247-252, 6-8 Jan. 2009
[12] Wei Yen, Ching-Wei Chen, Cheng-hsiang Yang,
“Single Gossiping with Directional Flooding Routing
Protocol in Wireless Sensor Networks,” In IEEE
Conference
on
Industrial
Electronics
and
Applications, Singapore , pp. 1604-1609, 3-5 June
2008.

32


[13] Yuecheng Zhang and Liaug Cheng, “Flossiping: A
New Routing Protocol for Wireless Sensor Networks,”
In International Conference on Networking, Sensing
and Control IEEE, vol. 2, pp. 1218-1223, 2004.
[14] Gunjan Khanna, Saurabh Bagchi, Yu-Sung Wu,”
Fault tolerant energy aware data dissemination
protocol in sensor networks,” In International
Conference on Dependable Systems and Networks, pp.
795- 804, June 28 2004-July 1 2004.
[15] Shousheng Zhao, Fengqi Yu , Baohua Zhao,” An
Energy Efficient Directed Diffusion Routing
Protocol,” In
International
Conference
on
Computational Intelligence and Security, Harbin, pp.
1067- 1072, 15-19 Dec. 2007.
[16] Azadeh Lajevardi, Abolfazl Toroghi , Haghighat,Arash
, Nasiri Eghbali, ” Extending Directed Diffusion
Routing Algorithm to Support Sink Mobility in
Wireless Sensor Networks,” In International
Conference on Communications (MICC), Kuala
Lumpur, Malaysia, pp. 541- 546, 15-17 Dec. 2009.
[17] Yong Yao, Johhanes Gehrke ,”The Cougar Approach
to In-Network query Processing,”vol. 31,September
2002.
[18] Narayanan Sadagopan, B.K., Ahmed Helmy.” The
ACQUIRE Mechanism for Efficient Querying in


Vol 2, Issue 8

ISSN 2278 – 3814

Sensor Networks,”In Sensor Network Protocols and
Application, 2003.
[19] Mohammd Jalil Piran, G. Ramma Murthy, "A Novel
Routing Algorithm for Vehicular Sensor Networks",
International Journal of Wireless Sensor Networks,
USA, p.p. 919-923, December 2010.
[20] Mohammd Jalil Piran, G. Ramma Murthy, G. Praveen
Babu, "Vehicular Ad Hoc and Sensor Networks;
Principles and Challenges", International Journal of
Ad Hoc, Sensor and Ubiquitous Computing, p.p. 3849, June 2011.
[21] Ahvar, Ehsan, Pourmoslemi, Alireza, Jalil Piran,
Mohammad, “Fear: A Fuzzy-based Energy-aware
Routing Protocol for Wireless Sensor Networks”,
Research Article,
Networking
and Internet
Architecture, 2011.

33

Data-Centric Routing Protocols in Wireless Sensor Network: A survey

More Related Content

What's hot (20)

Viewers also liked (7)

Similar to Data-Centric Routing Protocols in Wireless Sensor Network: A survey (20)

More from Ali Habeeb (20)

Recently uploaded (20)

Data-Centric Routing Protocols in Wireless Sensor Network: A survey