A comparative study in wireless sensor networks

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 6, No. 1, February 2014
DOI : 10.5121/ijwmn.2014.6105 61
A COMPARATIVE STUDY IN WIRELESS
SENSOR NETWORKS
1 Hasan Al-Refai 2 Ali Alawneh 3Khaldoun batiha
1 Department of Computer Information Systems, Philadelphia University, Amman Jordan
2Department of Management Information Systems, Philadelphia University, Amman
Jordan
3Dean of IT, Philadelphia University, Amman Jordan
ABSTRACT
Sensor networks consist of a large number of small, low-powered wireless nodes with limited computation,
communication, and sensing abilities, in a battery-powered sensor network, energy and communication
bandwidth are a precious resources. Thus, there is a need to adapt the networking process to match the
application in order to minimize the resources consumed and extend the life of the network. In this paper,
we introduce a comparative study in different routing algorithms that propose vital solutions to the most
important issues that should be taken into account when designing wireless network which are reliability,
lifetime, communication bandwidth, transmission rand, and finally the limited energy issue, so we will
introduce their algorithms and discuss how did they propose to solve such of these challenges and finally
we will do some evaluation to each approach.
KEYWORDS
Wireless sensor networks; Geographic routing; Localized routing; Power-aware routing, reliability,
timeliness, query processing, redundancy, energy conservation, QoS.
1. INTRODUCTION
Recent advances in technology have made low-cost, low-power wireless sensors a reality. Sensor
networks formed from such sensors can be deployed in an ad hoc fashion and cooperate to sense
and process a physical phenomenon. As each sensor has a finite battery source, an important
feature of sensor network is energy efficiency to extend the network’s lifetime.
Thus, the unique characteristics and constraints present many new challenges for the development
and application of WSNs. Due to the severe energy constraints of large number of densely
deployed sensor nodes, it requires a suite of network protocols to implement various network
control and management functions such as synchronization, node localization, and network
security.
There are a common routing protocols that have several shortcomings when applied to WSNs,
which are mainly due to the energy- constrained nature of such networks [3]. In most cases it is
very difficult and even impossible to change or recharge batteries for the sensor nodes. WSNs are
characterized with denser levels of sensor nodes deployment, higher unreliability of sensor nodes,
and sever power, computation, and memory constraints.
In [1] the authors have proposed an algorithm that investigate how to make on-demand routing
protocols adapt quickly to topology changes, this problem is important because such protocols use

62
route caches to make routing decisions; it is challenging because topology changes are frequent.
They propose proactively disseminating the broken link information to the nodes that have that
link in their caches. Proactive cache updating is key in making route caches adapt quickly to
topology changes, it is also important to inform only the nodes that have cached a broken link to
avoid unnecessary overhead. Thus, when a link failure is detected, the goal is to notify all
reachable nodes that have cached the link about the link failure. They define a new cache
structure called a cache table to maintain the information necessary for cache updates. A cache
table has no capacity limit; its size increases as new routes are discovered and decreases as stale
routes are removed.
The algorithm is fully-distributed, adaptive to the topology changes and fault-tolerant, since there
is always a way to handle broken link or node failure.
Moreover, this algorithm eliminates the overhead of flooding route request into the network and
can construct the multipaths on the fly. Thus, this algorithm is scalable and efficient and it has
less communication and storage overhead than other algorithms while providing greater resilience
to node failures.
In [3, 9] the authors have proposed a fast, randomized, distributed algorithm for organizing the
sensors in a wireless sensor network in a hierarchy of clusters with an objective of minimizing the
energy spent in communicating the information to the information processing center, but in the
other hand, they assume that the communication environment is contention and error-free, so
sensors do not have to retransmit any data, in the other words the overall reliability of the system
is on a stake, since there is no redundancy mechanism to cope with network failures.
So the main concentration in this approach is how to reduce the overall power consumption in the
system to cope with limited battery power or battery constraints.
In [4, 7] the authors have proposed an algorithm that handle the fault tolerant and reliability
requirements for data information in sensor networks, so in this approach each node has a
maximum transmission range to avoid link failure due to the power limitations of sensors(small
battery), also it uses Total Time to Live (TTL) to adjust the level of data reliability, so that the
more TTL the more possible paths used by the data the more robust it would be in the face of link
or node failures, and TTL is always chosen to be slightly higher than the diameter of a network
for the sake of necessary redundant packet.
In [5, 8] the authors have proposed an algorithm that is very based on query-based sensor
systems, in which a user would issue a query with QoS requirements in term of reliability and
timeliness and expect a response within a specific time which is called the deadline, where the
fault-tolerant for the overall system can be achieved by redundancy of path and source and the
optimal path and source redundancy is chosen depends on the routing algorithm they mentioned.
They also determine the optimal redundancy level that could satisfy the QoS requirements while
prolonging the lifetime of the wireless sensor network. Since sensor networks are constrained
with resources, they develop a hop-by-hop data dissemination paradigm to dynamically form
multiple paths for data delivery, instead of incurring extra overhead to first formulate multiple
paths before data delivery.

63
2. THE PROPOSED ROUTING ALGORITHM
In [13] the authors have proposed a dynamic source routing protocol to address the cache
staleness issue which limits the mobility of the overall network because of the delay that is, a
node is not notified when a cached route breaks until it uses the route to send packets.
When a node detects a link failure, the goal is to notify all reachable nodes that have cached that
link to update their caches. To achieve this goal, the node detecting a link failure needs to know
which nodes have cached the broken link and needs to notify such nodes efficiently. This goal is
very challenging because of mobility and the fast propagation of routing information.
The solution is to keep track of topology propagation state in a distribution manner. Topology
propagation state means which node has cached which link. In a cache table, a node not only
stores routes but also maintains two types of information for each route which are how well
routing information is synchronized among nodes on a route and which neighbor has learned
which links through a route reply. Each node gathers such information during route discoveries
and data transmission, without introducing additional overhead.
This approach enables dynamic source routing (DSR) to quickly remove stale routes, thus
reducing packet losses, delivery latency, and replies.
In [14,17] the authors have proposed the geographical power efficient routing (GPER) protocol
where each node knows whereabouts of the nodes within its power range and each intermediate
node locally identifies the most power efficient path to the sub destination within its reach and
forwards the packet to the next node on this locally identified path. The packet however, may not
follow this path exactly since the sub destination changes
from node to node. Two mechanisms, force routing and perimeter routing, are applied in GPER to
prevent infinite loops.
So, each sensor node makes local decisions as to how far to transmit; therefore, the protocol is
power-efficient, localized, highly distributed, and scalable, in GPER, given a final destination,
each node first establishes a sub destination within its maximum radio range, the node, however,
may decide to relay the packet to this sub destination through an intermediary node or alter the
sub destination if this will preserve power.
Three geographic multipath routing algorithms have been introduced (MGPSR, MGPERsub, and
MGPERhop), which generate alternative paths probabilistically based only on local information.
These algorithms identify and use different routes for different messages between a source and a
destination, they take away from the load on the nodes that are on the preferred path between the
end-points and distribute this load on the other nodes that are in the neighborhood. Of course, as
these paths are less perfect than the preferred path, they are expected to introduce a degree of
overhead, but experiments show that multipath routing improves the lifetime of the network,
especially in cases where the communication is non-uniformly distributed in the network
In [1,20] the authors have proposed a distributed, randomized, clustering algorithm to organize
the sensors in a wireless sensor network into clusters. Then they extend this algorithm to generate
a hierarchy of cluster heads and observe that the energy savings increase with the number of
levels in the hierarchy, they assume that sensors transmit at the same power level and hence they
all have the same radio range, moreover they suppose that each sensor uses 1 unit of energy to
transmit or receive I unit of data and finally, they assume that the network is fault-tolerant and the

64
communication environment is contention and error-free and sensor do not have to retransmit any
data.
A routing infrastructure has been used in this approach, hence when a sensor communicates data
to another sensor, only sensors on the routing path forward the data, which means the route
should be established first and then data can be transmitted which adds time overhead to the
system at all.
The algorithm states that each sensor in network becomes a cluster head and advertises itself to
the sensors within its radio range, and it is called the volunteer cluster head, this advertisement is
forwarded to all the sensors that are no more than k hops away from the cluster head where k is
chosen by the algorithm and any sensor that receives such advertisements and is not itself a
cluster head joins the cluster of the closest cluster head, any sensor that is neither a cluster head
no has joined any cluster itself becomes a cluster head and then it is called the forced cluster head
and since this algorithm is a distributed one, there is no need for clock synchronization between
the sensors at all even if they were in the same cluster. Since this algorithm is a hierarchy based
then each level does the same and elects the cluster head in the same manner.
In [15,18] the authors have proposed an algorithm based on majority voting in sensor network,
under the assumptions that the topology of the network is fixed, each node has a maximum
transmission range to overcome the power limitations of sensors, the propagation delay on each
functioning link is equivalent, and the processing delay in each node is non zero.
In this approach, when nodes are first deployed, the sink node sends a control packet informing
each node of the value of TTL (total time to live) where TTL is chosen under a certain criterion
that is, the more TTL the more possible paths used by the data the more robust to cope with link
or node failures, on the other hand, TTL Is always taken to be equal to the diameter of the
network to eliminate redundant traffic while broadcasting, but for the sake of necessary redundant
packet to perform majority voting, TTL should be set to be slightly higher the diameter. The
majority voting could be trigger whenever 3 packets with the same sequence number arrive.
In [16,19] the authors have proposed an algorithm based on a hop-by-hop data dissemination
paradigm to dynamically form multiple paths for data delivery, instead of incurring extra
overhead to first formulate multiple paths before data delivery. They assume that sensors have the
same initial energy level and sensors group themselves into distinct clusters in the system for
energy conservation purpose, each cluster has a cluster head (CH) which is responsible for data
reading and aggregating from sensor within same cluster and the role of cluster head is rotated
fairly among sensors in same cluster using a specific algorithm such as HEED or LEACH which
will help to conserve energy and prolong the lifetime of the network so that sensors will not
quickly deplete their energy, not only that but also the transmission power of a sensor even when
it is a cluster head is reduced to a minimum level to enable it to communicate within one-hop
radio range, thus a sensor node should use a multi-hop route protocol to enable it to communicate
with another sensor distant away.
Two forms of redundancy are used in this approach to help fault tolerance and lifetime reasons
which are path and source redundancy, that is, instead of having a single path to connect a source
cluster to the processing center, m disjoint paths may be used, on the other hand, they use m
sensor nodes to return readings to cope with sensor hardware and software faults.
In cluster network a source cluster head must send data information to the processing center and
this will consume energy depends on the length of the path between CH and the processing
cluster, if the overall system energy falls below a threshold value say threshold energy then the
system is considered as having depleted its energy. That is, the consumed energy can be

65
computed in terms of the energy to run the transmitter and receiver circuitry, the amplification
energy that is used to achieve acceptable signal to noise ratio, the loss energy due to channel
transmission.
Thus, the system lifetime can be defined here as the total number of queries the system can
answer correctly until it fails due to either channel or sensor faults or when the system energy
approaches the threshold level energy.
In [23] their work based on [21] to conduct a measurement study on a large-scale operating
sensor network system, GreenOrbs, with up to 330 nodes deployed in the wild. they aim to
understand how the sensor network performs when it scales to contain hundreds or even
thousands of nodes.
3. EVALUATION
The approach in [13] achieves fault-tolerance to the wireless network against link or node failures
by using caching technique in dynamic source routing which quickly removes stale routes, thus it
reduce packet losses, delivery latency, and replies.
The approach in [14,17] achieves power efficient routing method which enables the nodes to
adaptively chooses the transmission range to minimize the power consumption of the sensor as
well as of the non-participating listeners, and by using three families of probabilistic geographical
multipath routing protocols, they improve the network lifetime of wireless network, in addition,
all of these protocols are highly power efficient, distributed and scalable, on the other hand, the
reliability of the network still not handled in this approach.
The approach in [1,19] achieves conservation to the total energy by minimizing the total energy
spent in the system to communicate the information gathered by these sensors to the information
processing center and even this approach has a significant improvement on a time complexity
over many clustering algorithms and this makes this approach suitable for networks of large
number of nodes, it is still has a problem with reliability since it states that the communication
environment is contention and error free, moreover it neglects the fact that the cluster heads in
this architecture spend relatively more energy than other sensors because they have to receive
information from all the sensors within their clusters, aggregate this information and then
communicate to the higher level cluster heads or the information processing center.
The approach in [15,18] achieves reliability and fault tolerance for a simple network with fixed
topology which is free of crash failures but it is still unable to cover dynamic and sophisticated
networks reliabilities which are changing very quickly, on the other hand it, didn’t handle the
problem of power limitations of sensors or manage energy usage.
The approach in [16,19] achieves quality of service requirements in terms of reliability and
timeliness while maximizing the lifetime of query-based sensor networks by using hop-by-hop
packet broadcasting protocol, in the other hand this approach is based on the average behavior
without considering dynamics such as more energy may be consumed by some sensors over
others or some sensors may fail earlier than others. It doesn’t handle multiple QoS levels from
multiple queries and the bandwidth here is a bottle-neck, since much redundancy is achieved by
hop-by-hope packet broadcasting protocol.
The approach in [23] achieves two contributions, first, they argue that they are the first to conduct
a long term and large-scale measurement study on an operating sensor network in the wild. Where
they present observations across a variety of layers in the network that provide research

66
community empirical experiences on how practical problems affect when the sensor network
scales.
Second, based on their basic findings from the system measurement, they propose and initially
attempt to validate three conjectures that provide guidelines for future algorithm and protocol
designs with larger scale sensor networks. That is, 1- they thought it possible that some of the
intermediate nodes bottleneck the entire network, and most of currently used indicators may not
accurately capture them; 2- most of the wireless links in large scale sensor networks are
physically stable. But due to the dynamic features mainly come from the inherent concurrency of
network operations that need to be further investigated and considered in designing scalable
network protocols. 3- the environment, while with insignificant dynamics, has an unpredictable
impact on thesensor network under it. They suggest that an event based routing structure can be
trained optimized and thus better adapt to the wild environment when building a large-scale
sensor network.
REFERENCES
[1] Seema Bandyopadhyay and Edward J, Coyle, An Energy Efficient Hierarchical Clustering Algorithm
for Wireless Sensor Networks.
[2] D.B. Johnson, D.A. Maltz, Dynamic source routing in ad hoc wireless networks, in: Mobile
Computing, 1996, pp. 153–181.
[3] S. Murthy, J.J. Garcia-Luna-Aceves, An efficient routing protocol for wireless networks, Mobile
Networks and Applications 1 (2) (1996) 183–197.
[4] V.D. Park, M.S. Corson, A highly adaptive distributed routing algorithm for mobile wireless
networks, in: INFOCOM, 1997, pp. 1405–1413.
[5] C. Perkins, Ad hoc on-demand distance vector routing, in: MILCOM ’97 Panel on Ad Hoc Networks,
1997.
[6] C. Perkins, P. Bhagwat, Highly dynamic destinationsequenced distance-vector routing (DSDV) for
mobile computers, in: SIGCOMM, 1994.
[7] A. Rao, S. Ratnasamy, C. Papadimitriou, S. Shenker, I. Stoica, Geographic routing without location
information, in: MobiCom, 2003, pp. 96–108.
[8] V. Rodoplu, T. Meng, Minimum energy mobile wireless networks, in: ICC, 1998, pp. 1633–1639.
[9] L. Li, J. Halpern, Minimum energy mobile wireless networks revisited, in: ICC, 2001.
[10] Q. Li, J. Aslam, D. Rus, Distributed energy-conserving routing protocols for sensor networks, in:
IEEE HICSS, 2002
[11] J.H. Chang, L. Tassiulas, Energy conserving routing in wireless ad hoc networks, in: INFOCOM,
2000, pp. 22–31.
[12] Y.B. Ko, N.H. Vaidya, Location-aided routing (LAR) in mobile ad hoc networks, in: MobiCom,
1998, pp. 66–75.
[13] Xin Yn, Distributed cache updating for the dynamic source routing protocol, IEEE TRANSACTIONS
ON MOBILE COMPUTING, VOL. 5, NO. 6, JUNE 2006
[14] Shibo Wu, K. Selc, uk Candan, Power-aware single and multipath geographic routing in sensor
networks. Elsevier, Ad Hoc Networks 5 (2007) 974–997
[15] Jen-Yeu Chen, Yen-Shiang Shue, Hakeem Ogunleye, Saurabh Bagchi, A comparative study on data
fault tolerant requirements for data propagation in sensor networks.
[16] Anh Phan Speer Ing-Ray Chen Virginia Tech, On Optimal Path and Source Redundancy for
Achieving QoS AND Maximizing Lifetime of Query-Based Wireless Sensor Networks
[17] Haojun Huang*, Guangmin Hu, Fucai Yu. Energy-aware multipath geographic routing for detouring
mode in wireless sensor networks. European Transactions on Telecommunications. Volume 22, Issue
7, pages 375–387, November 2011
[18] Ehsan, S. A Survey on Energy-Efficient Routing Techniques with QoS Assurances for Wireless
Multimedia Sensor Networks. Communications Surveys & Tutorials, IEEE. Second Quarter 2012
[19] Debnath Bhattacharyya , Tai-hoon Kim , and Subhajit Pal. A Comparative Study of Wireless Sensor
Networks and Their Routing Protocols. Journal of MDPI Sensors. 2010.

67
[20] Guangjie Han, Lei Shu, Jianhua Ma, Jong Hyuk Park, Jianjun Ni. Power-Aware and Reliable Sensor
Selection Based on Trust for Wireless Sensor Networks . Journal of Communications, Vol 5, No 1
(2010)
[21] T. He, P. Vicaire, T. Yan, Q. Cao, G. Zhou, L. Gu, L. Luo, R. Stoleru, J. A. Stankovic, and T. F.
Abdelzaher, “Achieving Long-Term Surveillance in VigilNet,” TOSN, vol. 5, no. 1, pp. 1–39, 2009.
[22] Shio Kumar Singh , M P Singh , and D K Singh, “Routing Protocols in Wireless Sensor Networks –A
Survey”, International Journal of Computer Science & Engineering Survey (IJCSES), November
2010.
[23] Does Wireless Sensor Network Scale? A Measurement Study on GreenOrbs Yunhao Liu, Yuan He1,
Mo Li, Jiliang Wang1, Kebin Liu, Lufeng Mo, Wei Dong, Zheng Yang, Min Xi, Jizhong Zhao,
Xiang-Yang Li. 2013 Zhejiang University. Singapore.
AUTHORS
Ass. Prof. Dr. Hasan Alrefai
Dr. Hasan Al-Refai is an assistant professor and a head of Computer Information Systems
department for three years 2009 to 2012 at Philadelphia University, Jordan. He got his
PhD from National University of Malaysia, 2005. He joined Philadelphia in the fall (first)
semester 2006 after one year e xperience as assistant professor at Yarmouk University,
Jordan. His research interests include Cryptography, Mobile Cryptographic protocols, E-
Commerce Security, Formal Methods, Multimedia, Mobile & Distributed Computing. He has written a
number of journal articles and conference papers. He is a member of IJOPCM Editorial Board,
International Journal of Open Problems in Computer Science, Program Committee of the Third
International Symposium on Innovation in Information & Communication Technology - ISIICT 2009
(From 15 - 17 December, 2009) Jordan. Reviewer at Third International Symposium on Innovation in
Information & Communication Technology - ISIICT 2009 (From 15 - 17 December, 2009), Member of the
steering committee of the fourth International Symposium on Innovation in Information & Communication
Technology - ISIICT 2011 November, 2011.
Dr. ali Alawwneh
Dr. ali Alawwneh is an associate professor and a head of Management Information
Systems department at Philadelphiad University, Jordan. He got his PhD from Arab
Academy for Banking and Financial Sceiences, 2008. his Phd involved a study in e-
business in which he investigated e-business value creation at Jordan Banking Sector
through developing a model named e-TOEECLN. He has written a number of journal
articles and conference papers.
Dr. Khaldoun Mohammad Batiha
Dr. Khaldoun Mohammad Batiha is an associate professor and Dean of Information
Technology Faculty at Phialdelphia University, Jordan. Has research interests include
Computer Networks and Artificial Intelligence. He has written a number of journal
articles and conference papers. He has participated in authoring two books are Computer
Skills and Analysis and Design Algorithms.

A comparative study in wireless sensor networks

More Related Content

What's hot (15)

Viewers also liked (19)

Similar to A comparative study in wireless sensor networks (20)

Recently uploaded (20)

A comparative study in wireless sensor networks