Routing Protocol Design Challenges in Terrestrial

Wireless Sensor Network: A Review

Emmanuel Agboola Afolabi Banjo
Department of Computing and Informatics Department of Computing and Informatics
Bournemouth University Bournemouth University
Bournemouth, United Kingdom Bournemouth, United Kingdom
[email protected] [email protected]

Abstract—Terrestrial Wireless Sensor Networks are a critical

technology for industrial applications. The use of wireless sensor
networks in applications like mining, health care, and military op-
erations has generated a lot of interest in the industry. However,
the deployment and maintenance of wireless sensor networks
in an industrial environment can be challenging because of the
need to operate in remote areas and with limited power supply.
Wireless sensor networks are a collection of a large number of
sensor nodes that are typically battery-powered. It is important
to control the node energy consumption to protract the network’s
lifetime and functionality. Several routing protocols in past
literature have attempted to manage the energy dissipation in the
network. However, due to the network structure and formulation
of wireless sensor networks, designing a routing protocol that
efficiently manages energy dissipation while meeting Quality of
Service requirements, such as high throughput and scalability
is quite challenging. This paper explains the infrastructure of Fig. 1. A simplified WSN [6]
Terrestrial Wireless Sensor Networks and reviews the design
challenges of routing protocol. In addition, the paper identifies
the flaws in the design of some of the recent routing protocols. In general, the more Sn that are used to monitor an area
Future research should focus on developing reliable routing or an event, the more precise and trustworthy the data that is
protocols that meet Quality of Service requirements and are
energy efficient. transmitted [7]–[9]. Each Sn conveys the sensed data to the
Keywords— Terrestrial wireless sensor networks (WSNs), Bat- base station or sink in accordance with the chosen routing
tery energy, Network lifetime, Routing Protocol, Quality of protocol Rp . The Rp specifies how theSn exchange sensed
Service (QoS), Sensor nodes data with one another, enabling them to choose the route(s)
with the lowest cost (cost-efficient) to the base station. These
I. I NTRODUCTION Rp can be developed as a load-balanced tree routing algorithm
The advancement of present-day technologies has prompted [10], [11] or a clustering routing algorithm [12], [13]. Irrespec-
the development of low-powered battery devices called sensor tive of the type of routing algorithm, a cost-efficient Rp must
nodes, Sn . In most cases, these Sn are used in large quantities meet some important quality of service (QoS) requirements
to measure physical values in isolated locations to set up a such as high packet throughput, robustness, scalability, and so
wireless sensor network (WSN). WSNs can be used in a wide on. Above all, the Rp must reduce the energy expended to
range of industrial settings such as in health care, mining, prolong the lifetime of the network for a meaningful amount
military operations and so on[1],[2],[3]. WSN technology of time. With regard to energy consumption, because Sn
offers several benefits (such as implementation ease, result are predominantly battery-powered devices, it is necessary to
accuracy, lower implementation costs and scalability) in these manage their energy consumption to extend the lifetime of the
applications over conventional networking solutions [4], [5]. network [4]. Many Rp have been proposed in the literature.
WSNs typically house many Sn that interact with one These protocols make use of different types of algorithms,
another to transmit high-quality data to the base station or including genetic, game theory, swarm intelligence, linear
sink without relying on any existing infrastructure. The Sn programming (LP), meta-heuristic, and heuristic. Most of the
forming a WSN are distributed randomly in a sensor field or Rp have been designed with the intention of minimising
an isolated place where they interact with each other to sense energy consumption while transmitting data over the networks
the physical values as shown in Fig. 1. Furthermore, in many [4]. Yet, the energy consumption issues of terrestrial WSNs are
WSN applications, it is extremely difficult to restore these Sn still a NP-hard problem due to the structure and formulation of
after they have been fully deployed. the network, most especially for large-scale WSNs. Therefore,
this paper discusses the network structure and formulation of there are various definitions for the lifetime of WSNs. Some
terrestrial WSNs and reviews some of the routing protocols’ sources define the network lifetime as the time till the energy
design challenges. Also, the paper identifies the flaws in the in any Sn or all the Sn in the network is drained, while some
design of some of the recent Rp for terrestrial WSNs. Besides, authors define it as the time until the energy in a specified
potential future research avenues for developing reliable Rp percentage of the network’s Sn is exhausted [19]. In General,
for WSNs are identified. The remainder of this paper is the lifetime of a WSN is defined based on its formulation and
structured as follows. Section II explains the infrastructure application, but the concept remains constant.
and main terminologies used in WSNs. In Section III, the C. Residual energy, ER
energy dissipation model used to theoretically evaluate the The residual energy, ER of a Sn is the amount of energy
performance of terrestrial WSNs is explained. The design left in it after each network packet transmission round. A
challenges in developing a reliableRp for WSNs are carefully sensor node’s lifetime is measured typically by its current ER
analysed in Section IV. Section V identifies the flaws in the following a given packet transmission cycle. If the quantity
design of some of the recentRp for terrestrial WSNs. In of energy in a particular Sn is given as Ei and the energy
Section VI, the review findings and potential future research consumed during P transmission rounds is given as Ep , then
avenues for developing reliable Rp for WSNs are identified. the ER of a Sn is defined as [4]:
Finally, the paper is concluded with discernible remarks in
P
Section VII. X
ER = Ei − Ep , if p = 0; ER = Ei (1)
II. WSN I NFRASTRUCTURE AND T ERMINOLOGIES p=0

D. Distance
One crucial factor taken into account when designing Rp
is the distance k. The distance indicates how far apart the Sn
are from one another or how far away a Sn is from the sink.
To minimise transmission energy, it is encouraging for k to be
as small as possible when designing most routing algorithms
because the transmission energy is directly proportional to
distance. Accordingly, most Rp attempt to select the route that
keeps down the value of k from the originating node to the
destination [4].
E. Hop count,h
Another common decision variable in developing Rp for
Fig. 2. Simplified diagram of heterogeneous and homogeneous WSNs [14] terrestrial WSNs is the hop count, h. A few studies use the
hop count as a performance metric to evaluate the performance
A. Heterogeneous and Homogeneous WSNs of divergent Rp [20], [21]. The number of forwarding nodes
WSNs can be categorised as heterogeneous or homogeneous the packet passes through on its way from the originating node
WSNs depending on their infrastructure. All of the Sn in to the destination is known as the hop count. To maximise the
homogeneous WSNs share the same hardware, including the lifetime of WSNs, it is ideal to reduce the h during the packet
radio, processing, sensing, and power supply subsystems. transmission round; however, there is a trade-off between
However, in heterogeneous WSNs, the sensor nodes’ hardware doing so and increasing the lifetime of WSNs. Minimising
components differ from one another. One of the principal the h between the source and destination nodes forces some
reasons for designing heterogeneous WSNs is to enable longer nodes to transmit a lot of messages, which can quickly drain
transmission times by giving some Sn greater sensing range their energy. Thus, when developing a Rp for terrestrial WSNs
and more battery power [15]. Homogeneous WSN deploy- that will balance the network load, the h is generally regarded
ments can be relatively simple; yet, heterogeneous WSN as an important decision variable.
deployments are more beneficial in real-time deployments
III. WSN ENERGY MODEL
because they are closer to practical applications [16], [17].
Fig. 2 illustrates the distinction between heterogeneous and
homogeneous WSNs at the initial deployment stage. In a het-
erogeneous structure, the energy levels of the sensor node Sn
vary, with nodes closer to the sink being equipped with more
battery power to facilitate their participation in all transmission
rounds back to the sink. However, in a homogeneous structure,
all Sn are equipped with the same battery power. Fig. 3. A simplified energy consumption model for WSNs [22]
B. Network Lifetime
The network lifetime is a fundamental parameter used to Fig. 3 depicts a simple WSN energy consumption model
evaluate the energy efficiency of WSNs [18]. In the literature, [22]. The transmitting and receiving Sn consume energy
in powering the radio subsystem that comprises the power The TDMA scheduling saves energy in this way by enabling
amplifier and radio electronics as depicted in Fig. 3. The a Sn to sleep continuously outside of its transmission window.
channel model employed by the power amplifier is determined When utilising single-hop communication, S8 sends the
by the distance k between the transmitting and receiving Sn . b bit packet straight to the sink, without the need for a
The multipath model is adopted if k is more than a predefined relay node. It is assumed that there is a sizeable distance
threshold k0 . Whereas, if k ¡ k0 , the free space channel k between S8 and the sink as depicted in the figure. This
model is adopted [22]. Thus, the energy dissipated by a Sn in implies that S8 will dissipate plenty energy to convey theb bit
transmitting b bit message is given as [23]: packet to sink employing Eqns. 2-4. Therefore, forwarding the
( sensed message to the sink using single-hop communication
bEe + bEs k 2 if k < k0 is not advisable because the amount of energy dissipated
Et (b, k) = (2)
bEe + bEm k 4 if k ≥ k0 , during transmission is directly proportional to the transmission
distance [4]. The Sn with large distance to the sink will drain
where Ee is the energy dissipated to power the radio the battery power quickly which will limit the functionality of
electronics, where Es is the energy dissipated to power the the network.
radio electronics, is the free space power loss, Em is the To conserve the energy of the Sn with large distances from
multipath power loss and K0 is the predefined transmission the sink, some Rp divides the Sn into clusters and alternates
threshold distance given as: the cluster head duties. Any clustering Rp primary function
r is to select a subset of the nodes as cluster heads (CH) and
Es structure the other nodes around these heads as non-cluster
k0 = (3)
Em heads [25].
Conversely, the energy dissipated by a Sn in receiving b bit The nCH nodes forward the measured physical values to
message is given as: their specific CH nodes while the CH nodes aggregate all the
messages from the nodes in their clusters and forward them to
Er (b) = bEe (4) the sink [26]. Nonetheless, the distance problems with single-
hop communication are not solved completely by adopting the
Because of the simplicity of these equations (Eqns. (2)-(4)), clustering approach. A major setback is that the CHs forward
they are used to theoretically evaluate the performance of Rp . the aggregated packet to the sink directly using single-hop
Nonetheless, in reality, it is quite difficult to model the fast- communication. As such, the CHs with large distances from
changing radio wave propagation using these equations. the sink will speedily drain their energy which can lead to
untimely network partitioning [4].
Due to this obvious distance issue with single-hop commu-
nication, modern WSN research employs multi-hop commu-
nication techniques. Using multi-hop communication strategy
is an energy-efficient way to get around this restriction as
described by Fedor and Collier [27]. In multi-hop commu-
nication, the packet is sent from the originating Sn to the
sink using at least one relay node. Researchers predicted that,
in contrast to the single-hop communication technique, the
energy consumed by the Sn at great distances from the sink
will be reduced.

Fig. 4. A simplified heterogeneous WSN scenario (Adopted from Ogundile

and Alfa [4])

IV. WSN ROUTING P ROTOCOL D ESIGN C HALLENGES

Consider the heterogeneous WSN shown in Fig. 4. After
deployment, it is assumed that the Sn are static and the
position of the Sn and sink are predefined. Also, the figure
shows the ER of all the Sn . Suppose a Sn , S8 at its assigned
time-division multiple-access (TDMA) sends a packet to the Fig. 5. A simplified example of a Multi-hop WSN communication (Adopted
sink either employing the single-hop communication (directly) from Ogundile and Alfa [4])
or the multi-hop communication (through relay nodes). It
should be noted that the TDMA schedule makes certain that Nevertheless, the Rp that employs the multi-hop communi-
there are no data collisions while the network is operating [24]. cation techniques has not also been able to solve the energy
dissipation problems. Finding a balance between the most energy (6 times) to receive the transmitted packet than the path
energy-efficient route and the most distance-efficient route is in Fig. 5(a) (five times). The h is a crucial decision factor that
a significant issue with most routing algorithm that employs WSN Rp developers employ to maximise the energy usage of
multi-hop communication techniques [4]. For instance, from the Sn in the WSN [32]. With multi-hop communication, it
Fig. 4, presume that the shortest path to the sink from S8 can also be difficult to figure out the ideal h while transmitting
is through S13 , S23 , S27 , S15 and S7 as shown in Fig. a message from the originating node to the sink. The smaller
5(a). Observe from Fig. 5(a) that the ER of S27 is small as the value of h the lower the energy dissipated in receiving and
compared to the other Sn in this routing path. This means that forwarding the packet from the originating node [33].
S27 will deplete all of its battery energy if used continuously For this reason, it is ideal to minimise the number of hops
as a forwarding node to relay the message from S8 . In such between a source node and the sink. However, doing so does
situation, it is ideal to conclude that this shortest distance path not promise a balance in the energy expended in the network.
to the sink from S8 is not an energy-efficient path. However, To justify, suppose that the path in Fig. 5(a) provides the least
as mentioned earlier, the distance from the originating node h from S8 to the sink. As depicted, S27 is part of the nodes
to the sink plays a key row in reducing the energy expended relaying the message from S8 to the sink. Whereas, S27 is
in the network. In this scenario, the shortest distance path low on ER as compared to the other nodes in the relay chain.
does not balance efficiently the energy expended by the To preserve the entire network functionality at this stage, S27
Sn in the network. Therefore, due to network requirements should not participate in the relay chain to forward the packet
and formulations, determining the most distance-efficient path from S8 to the sink. In the actual sense, S8 should sleep
from the source node to the sink that maximises the energy to conserve its energy unless it is forwarding its message.
dissipation by Sn can be an NP-hard problem. Although the path depicted in Fig. 5(a) has the lowesth from
Furthermore, to determine the most energy-efficient path(s) S8 to the sink, it can be concluded that this path is not optimal.
from the originating node to the sink, some Rp take advantage Accordingly, there exists a path with an optimal h from S8 to
of the ER of the Sn in the network. Researchers use the sensor the sink that will balance the energy expended by the Sn in
node’s ER as a crucial design component when modelling the network.
routing algorithms for WSNs [28], [29]. Some of these Rp To justify, suppose that the path in Fig. 5(a) provides the
choose the Sn with highER to serve as part of the relay chain least h from S8 to the sink. As depicted, S27 is part of the
that connects the originating node to the sink [30]. nodes relaying the message from S8 to the sink. Whereas, S27
For instance, presume that the relay nodes for S8 to forward is low on ER as compared to the other nodes in the relay chain.
its sensed message to the sink based on their ER is given in To preserve the entire network functionality at this stage, S27
Fig. 5(b). As depicted, these Sn serving as the forwarding should not participate in the relay chain to forward the packet
nodes have a significant ER , so they can send the message from S8 to the sink. S8 should however sleep to conserve its
from S8 to the sink for a considerable number of data energy unless it is forwarding its own message. Although the
transmission phases before running out of energy. This path path depicted in Fig. 5(a) has the lowest h from S8 to the sink,
depicted in Fig. 5(b) will surely solve the problem with the it can be concluded that this path is not optimal. Accordingly,
path depicted in Fig. 5(a) because all the Sn in the relay there exists a path with an optimal h from S8 to the sink that
chain have sufficient ER and can participate in a considerable will balance the energy expended by the Sn in the network.
number of data transmission phases. However, this path (Fig. The energy-hole problem is another evident drawback of
5(b)) does not balance efficiently the energy expended in the the multi-hop communication approach [34], [35]. When using
network as a whole. To buttress, the path depicted in Fig. 5(b) multi-hop communication, the Sn localised closer to the sink
will transmit the message to the sink over a large distance dissipates more energy to receive and convey the packet to the
in comparison to the shortest distance path (Fig. 5(a)). This sink [4]. Most of the load traffic to the sink goes through these
simply means that more energy is dissipated in the network Sn ; as such, they tend to expend more energy as compared to
to transmit the sensed packet to the sink using the high ER the other Sn in the network. In fact, the closer the Sn to the
path (Fig. 5(b)). In this given case, even though the message is sink the more energy they expend. To illustrate, most of the
delivered using nodes with high ER , it can be concluded that messages to the sink in Fig. 4 will pass through any of the
this highER path from S8 to the sink is not the optimal energy- following nodes: S12 , S15 , S7 , and S5 .
balanced route. Consequently, there exists a path from S8 to These nodes will deplete their battery power quickly when
the sink that will balance the energy expended by the Sn in compared to the other Sn with large distance to the sink.
the network as a whole. The path can be highly unpredictable This means that the energy consumption in the network is
and may vary with every data transmission phase based on the not balanced, which can cause the Sn closer to the sink to
network’s sensor nodes’ ER and the communication distance die quickly, thereby resulting in a phenomenon called energy-
[31]. hole [36]. Hence, in developing a Rp that involves multi-hop
One noteworthy finding is that, in contrast to the shortest communication, the algorithm must cater for the energy-hole
path depicted in Fig. 5(a), the high ER path depicted in Fig. problem at all times, which can be a difficult task to achieve.
5(b) needs more relay nodes (h= 6) to transmit the message. In summary, decision variables such as h, k and ER must be
As indicated in Eqn. 4, the path in Fig. 5(b) will expend more taken into consideration in the design of any Rp . Nonetheless,
balancing the energy dissipation in a WSN can be uncertain low ER cannot be the CH in that particular transmission round.
because of the trade-off in choosing these decision variables Besides, the CH nodes use multi-hop communication among
[37]. Nonetheless, it is desirable that the Rp efficiently manage themselves to forward the aggregated message to the sink,
the energy expended to prolong the lifetime and functionality which is located at the top of the sensing field. Technically, the
of the network for a reasonable amount of time. In addition to CH nodes closer to the sink will forward messages throughout
energy efficiency, routing protocols Rp should also strive to that particular data transmission round because they do not
achieve other design benefits or objectives such as high packet only forward their message to the sink, they forward all the
throughput, accuracy, and scalability [38]. messages coming from the CH nodes at a far distance from
the sink. Therefore, the CH nodes closer to the sink will die
V. ROUTING P ROTOCOL D ESIGN F LAWS untimely, while the other CH nodes using these CH nodes
Walter et al [39] proposed a clustering energy-efficient Rp closer to the sink as relay nodes will not be able to forward
for WSNs. Their routing algorithm was formulated based their aggregated messages to the sink. This will result in the
on the knowledge of the ER , number of neighbours and energy-hole phenomenon explained earlier.
the average received signal strength (RSS). In addition, they
assumed that the base station is located at the summit of VI. R EVIEW F INDINGS AND P OSSIBLE R ESEARCH
the sensing field. Their clustering routing algorithm does not D IRECTIONS
consider the distance, k which is a key parameter in the design
of Rp and CH rotation is not enabled. Accordingly, their Various research works are being conducted to provide cred-
algorithm will suffer untimely network partitioning because ible solutions to the energy dissipation challenge associated
the CHs that are far from the base station will transfer their with WSNs, most especially for large-scale terrestrial WSNs.
aggregated messages over a long distance. This in turn will Selecting the design decision variable(s) is still a study topic
reduce the packet throughput to the base station and limit other in progress. In most circumstances, adopting a single decision
functionalities of the network. variable such as distance, k cannot ensure an optimal balance
A heterogeneous energy and traffic-aware sleep-awake in the energy dissipation in the network because each of these
cluster-based Rp for WSNs was proposed by Shagari et al [40]. decision variables has benefits and drawbacks. In comparison
This routing algorithm improves on the traditional TDMA to employing just one decision variable, efficiently combining
scheduling used in the studies of Sharma and Bhondekar, and several design choice variables can provide a superior solution
Ahmed et al [41],[42]. Likewise, a new CH selection process to the energy dissipation problem. Nevertheless, it can be NP-
is presented by Shagari et al [40]. whereby the CHs were hard to effectively combine these design decision variables to
selected based on the ER , load traffic, and Sn with the high guarantee an optimal balance in the energy dissipation in the
number of nearest neighbours [40]. The transmission distance, network.
k between the sink and the CHs was ignored in the CHs’ In addition, as documented by Ogundile and Alfa [4],
selection process. If the selected CH is at a large distance from over 50 percent of academics in this field formulate their
the sink, it will expend all its battery energy in sending the routing algorithm using the ER , h, and k decision variables.
aggregated data to the sink. Therefore, that particular cluster Researchers use the Er to ensure that all the relaying Sn
will be partitioned from the whole network in that transmission have adequate energy to receive and transmit the message
round resulting in lower packet throughput. to its destination during that transmission round. The h is
Yao et al [43] proposed an energy-efficient Rp based on used to determine the number of forwarding nodes from the
multi-threshold segmentation to improve the cluster formation originating node to the destination. Keeping the h at minimum
and CH selection process in clustering WSNs [43]. The cluster is preferable to minimise the energy expended in receiving
formation was based on a multi-threshold image segmentation and relaying the message. while the k is an important variable
while the CHs were selected based on the ER and the location used to select the next hop. As shown in Eqns. 2-4, a small
of the CH to the base station. Note that the algorithm assumed k is desirable because it’s directly proportional to the energy
that the base station is located in the middle of the sensing dissipated during transmission. Nevertheless, it is emphasised
field; as such, their design calculation is centred on this that other decision variables such as the average network
assumption. If the position of the base station is changed, for energy, RSS, transmission power and so on, can also be
example, to the top of the sensing field, the performance of used in developing the Rp mathematical formula. Nonetheless,
this routing algorithm will plummet because the distance of looking closely at Eqns. 2-4, it can be noted that the b bit
the CH nodes to the base station has changed. That is, the message is a scaling factor. Accordingly, the energy required
CH nodes at the bottom of the field will pay the increased to receive or convey the message to its destination increases
transmission distance cost to convey their aggregated message with the value ofb. Interestingly, most academics have mostly
to the base station. Therefore, the performance of this proposed overlooked the weight of the q when developing their Rp . The
Rp is subjected to the network configuration. understanding of b can also be used as a significant decision
In a recent study by Pedditi and Debasis [44], a clustering variable that can be combined with the other decision variables
energy-efficient Rp was proposed for IoT-based WSNs. The while developing a Rp that will balance the energy dissipation
algorithm selected CHs based on the ER , such that Sn with in the network.
 Another important observation in designing a reliable Rp [9] J. N. Al-Karaki and A. E. Kamal, “Routing techniques in wireless sensor
is the position of the sink or base station. There should be networks: a survey,” IEEE Wireless Communication, vol. 11, pp. 6-28,
2004.
a standard for the position of the sink based on applications. [10] M. Masdari and M. Tanabi, “Multipath Routing Protocols in Wireless
Researchers assume different positions for the sink in their Rp Sensor Networks: A Survey and Analysis,” International Journal of
design. If the position of the sink changes, the performance Future Generation Communication and Networking, vol. 6, pp. 181-192,
2013.
of their routing algorithm will plummet. This implies that the [11] A. A. Anasane and R. A. Satao, “A Survey on various Multipath Routing
performance of their Rp relies heavily on the sink’s position. protocols in Wireless Sensor Networks,” Procedia Computer Science,
Furthermore, in Rp such as highlighted in the works of Taleb vol. 79, pp. 610-615, 2016.
[12] X. Liu, “A Survey on Clustering Routing Protocols in Wireless Sensor
et al [ 45] and Dholey et al [46], the sink is assumed to be Networks,” Sensors, vol. 12, pp. 11113-11153, 2012.
mobile, which of course enhances the performance of their [13] O. Younis, M. Krunz and S. Ramasubramania, “Node Clustering in
Rp, most especially in avoiding the energy hole problem Wireless Sensor Networks: Recent Developments and Deployment Chal-
lenges.” IEEE Network, vol. 20, pp. 20-25, 2006.
mentioned earlier. Therefore, the sink’s position should play [14] Rowayda A. Sadek, Doha M. Abd-alazeem, and Mohamed M. Abbassy,
an important role in the design of Rp for WSNs. It is ‘A New Energy-efficient Multi-hop Routing Protocol for Heterogeneous
acknowledged that there are ongoing research projects on Wireless Sensor Networks’, IJACSA) International Journal of Advanced
Computer Science and Applications, vol. 12, pp. 1–3, 2021, doi: DOI:
WSN node placement optimisation. 10.14569/IJACSA.2021.0121154.
However, the focus should be on determining the best [15] Y. Wang, X. Wang, B. Xie, D. Wang and D. P. Agrawal, “Intrusion De-
location for the sink because a poorly positioned sink will tection in Homogeneous and Heterogeneous Wireless Sensor Networks,”
IEEE Transactions on Mobile Computing, vol. 7, pp. 698-711, 2008.
hamper the performance of a well-designed Rp. More so, [16] R. Sheikhpour, S. Jabbehdari and A. Khadem-Zadeh, “Comparison of
defining a standard for the position of the sink will create a Energy Efficient Clustering Protocols in Heterogeneous Wireless Sensor
level ground to evaluate the performance of a newly developed Networks,” International Journal of Advanced Science and Technology,
vol. 36, pp. 27-40, 2011.
Rp. [17] S. Al-Sodairi, and R. Ouni, “Reliable and energy-efficient multi-hop
LEACH-based clustering protocol for wireless sensor networks,” J
VII. C ONCLUSION Sustain Comput Inform Syst., vol. 20, pp. 1-13, 2019
[18] Z. Zhao, K. Xu, G. Hui and L. Hu, “An Energy-Efficient Clustering
This paper explains the structure of terrestrial WSNs and Routing Protocol for Wireless Sensor Networks Based on AGNES with
reviews some of the problems in designing an energy-balanced Balanced Energy Consumption Optimization,” Sensor, vol. 18, no. 3938,
pp. 1-27, 2018.
routing protocol, Rp . More so, the design flaws with some [19] P. Kuila and P. Jana “Energy Efficient Clustering and Routing Al-
recently developed routing protocols for terrestrial WSNs were gorithms for Wireless Sensor Networks: Particle swarm optimization
identified. The paper emphasised that the design of routing approach,” Eng. Appl. Artif. Intell., vol. 33, pp. 127-140 2014.
[20] J. Kim, X. Lin, N. Shroff and P. Sinha, “Minimizing delay and maxi-
protocols for terrestrial WSNs can be an NP-hard problem. mizing lifetime for wireless sensor networks with anycast,” IEEE/ACM
Nevertheless, the paper created awareness of the challenge in Transactions on Networking, vol. 18, pp. 515-528, 2010.
developing an efficient Rp and the potential future research [21] A. Capone, M. Cesana, D. De Donno and I. Filippini, “Deploying multi-
ple interconnected gateways in heterogeneous wireless sensor networks:
avenues for developing a reliable Rp for WSNs. An optimization approach,” Computer Communications, vol. 33, pp.
1151-1161, 2010.
R EFERENCES [22] S. Tyagi, and N. Kumar, “A systematic review of clustering and
routing techniques based upon the LEACH protocol for wireless sensor
[1] N. Atitallah, O. Cheikhrouhou, K. Mershad, A. Koubaa, and F. Hajjej, networks.” Journal of Networks and Computer Applications, vol. 36, pp.
“CERP: Cooperative and Efficient Routing Protocol for Wireless Sensor 623-645, 2013.
Networks,” Sensors, vol. 23, no. 8834, pp. 1-19, 2023. [23] T. Rappaport, Wireless Communications: Principles Practice.Englewood
[2] A. Manhar, and D. Dembla, “Improved Hybrid Routing Protocol (IHRP) Cliffs, NJ: Prentice-Hall, 1996.
in MANETs Based on Situation Based Adaptive Routing,” International [24] L. Clare, G. Pottie and J. Agre, “Self-organizing distributed sensor
Journal of Electrical and Electronics Research, vol. 11, no. 1, pp. 15-24, networks,” Unattended Ground Sensor Technologies and Applications,
2023. vol. 3713, pp. 229-237, 1999.
[3] M. Ibrahim, and A. E. Ahmed, “Energy-aware intelligent hybrid routing [25] M. Youssef, A. Youssef, and M. Younis, “Overlapping Multi-
protocol for wireless sensor networks. Concurrency and Computation: hop Clustering for Wireless Sensor Networks,” 2009, DOI:
Practice and Experience,” vol. 34, 2021. 10.1109/TPDS.2009.32.
[4] O. O. Ogundile and A. S. Alfa, “A Survey on Energy-Efficient and [26] C. Gherbi, Z. Aliouat, and B. Mohamed, “An adaptive clustering
Energy-Balanced Routing Protocol for Wireless Sensor Networks,” approach to dynamic load balancing and energy efficiency in wireless
Sensor, vol. 17, no. 1084, pp. 1-43, 2017 sensor networks,” 2016.
[5] P. Rawat, K. D. Singh, H. Chaouchi and J. M. Bonnin, “Wireless sensor [27] S. Fedor and M. Collier, “On the Problem of Energy Efficiency of
networks: a survey on recent developments and potential synergies,” The Multi-Hop vs One-Hop Routing in Wireless Sensor Networks,” in 21st
Journal of Supercomputing vol. 68, pp. 1-30, 2014. International Conference on Advanced Information Networking and
[6] S. Gul, A. B. Kamalrulnizam, R. Sabit, A. K. K. K. Naveed, and B. Applications Workshops (AINAW’07), pp. 380–385, 2007.
Tehmina, ‘Recent Advancement of Data-Driven Models in Wireless [28] J. Wang, Y. Gao, W. Liu, Arun Kumar Sangaiah, and Hyung Sik Kim,
Sensor Networks: A Survey’,MDPI, vol. 9, no. 4, pp. 2–4, 2021, “Energy Efficient Routing Algorithm with Mobile Sink Support for
https://doi.org/10.3390/technologies9040076. Wireless Sensor Networks,” MDPI, vol. 19, no. 7, pp. 1494–1494, 2019,
[7] D. Hemanand, C. Senthilkumar, Omar S. Saleh, B. Muthuraj, A. Anand doi: https://doi.org/10.3390/s19071494.
and V. Velmurugan, “Analysis of power optimization and enhanced [29] A. R. Affane, H. Satori, F. Sanhaji, Y. Boutazart, and K. Satori,
routing protocols for wireless sensor networks,” Measurement: Sensors, “Energy enhancement of routing protocol with hidden Markov model in
vol. 25, pp. 1-6, 2023. wireless sensor networks,” Neural Computing and Applications, 2022,
[8] O. O. Ogundile, M. B. Balogun, O. E. Ijiga and E. O. Falayi, “Energy doi: https://doi.org/10.1007/s00521-022-07970-3
balanced and energy-efficient clustering routing protocol for wireless [30] R. Arroyo-Valles, A. Simonetto, and G. Leus, “Consistent Sensor,
sensor networks,” IET Communications, vol. 13, no. 10, pp. 1449-1457, Relay, and Link Selection in Wireless Sensor Networks,”, 2017. doi:
2019. https://doi.org/10.48550/arXiv.1705.02298.
[31] S. Rajaram, A. B. Karuppiah, and K. V. Kumar, “Secure Routing Path
Using Trust Values for Wireless Sensor Networks,” International Journal
on Cryptography and Information Security, vol. 4, pp. 27–36, 2014, doi:
https://doi.org/10.5121/ijcis.2014.4203
[32] C. Narmadha, V. Hamsadhwani, R. Poonguzhali and T. Kavitha,
”Optimizing Techniques for Energy Efficiency in Wireless Sensor
Network,”6th International Conference on Computing Methodologies
and Communication (ICCMC), pp. 337-340, 2022, doi: 10.1109/IC-
CMC53470.2022.9753748
[33] D. Rajan, “Optimum number of hops in linear multihop wire-
less networks,” International Journal of Advances in Engineering
Sciences and Applied Mathematics, vol. 5, pp. 32–42, 2013, doi:
https://doi.org/10.1007/s12572-013-0080-8
[34] D. R. Edla, A. Lipare and S. R. Parne, “Load balanced cluster formation
to avoid energy hole problem in WSN using fuzzy rule-based system,”
Wireless Network, vol. 29, pp. 1299-1310, 2023
[35] T. M. Behera and S. K. Mohapatra “A novel scheme for mitigation of
energy hole problem in wireless sensor network for military application,”
International Journal of Communication systems vol. 34, no. 11, e4886,
2021.
[36] K. Mahendrababu and K. L. Joshitha, ”A solution to energy hole
problem in Wireless Sensor Networks using WITRICITY,” Interna-
tional Conference on Information Communication and Embedded Sys-
tems (ICICES2014), Chennai, India, pp. 1-6, 2014, doi: 10.1109/ICI-
CES.2014.7034150.
[37] M. K. Khan et al., “Hierarchical Routing Protocols for Wireless Sensor
Networks: Functional and Performance Analysis,” Journal of Sensors,
2021. Available: https://www.hindawi.com/journals/js/2021/7459368/
[Accessed: Nov. 27, 2023]
[38] L. Garcı́a Villalba, A. Sandoval Orozco, A. Triviño Cabrera,
and C. Barenco Abbas, “Routing Protocols in Wireless Sensor
Networks,” Sensors, vol. 9, no. 11, pp. 8399–8421, 2009, doi:
https://doi.org/10.3390/s91108399.
[39] A. Walter, M. Singh and S. S. Tripathi, “Energy Efficient Routing Pro-
tocol in Wireless Sensor Network,”International Journal of Computer
Engineering and Technology (IJCET), vol. 11, no. 2, pp. 31-48, 2020.
[40] N. M. Shagari, M. Y. I. Idris, S. B. Salleh, I. Ahmedy, G. Murtaza and H.
A. Shehadeh, “Heterogeneous Energy and Traffic Aware Sleep- Awake
Cluster-Based Routing Protocol for Wireless Sensor Network,”IEEE
Access, vol. 8, pp.12232-12252, 2020.
[41] D. Sharma and A. P. Bhondekar, “Traffic and energy aware routing for
heterogeneous wireless sensor networks,” IEEE Commun. Lett., vol. 22,
no. 8, pp. 1608-1611, 2018.
[42] G. Ahmed, J. Zou, M. M. S. Fareed and M. Zeeshan, “Sleep-awake
energy efficient distributed clustering algorithm for wireless sensor
networks,” Comput. Electr. Eng., vol. 56, pp. 385-398, 2016.
[43] Y. -D. Yao, X. Li, Y. -P. Cui, J. -J. Wang and C. Wang, “Energy- Efficient
Routing Protocol Based on Multi-Threshold Segmentation in Wireless
Sensors Networks for Precision Agriculture,” IEEE Sensors Journal, vol.
22, no. 7, pp. 6216-6231, 2022.
[44] R. B.Pedditi and K. Debasis, “Energy Efficient Routing Protocol for an
IoT-Based WSN System to Detect Forest Fires,” MDPI, Appl. Sci, vol.
13, 3026, pp. 1-22, 2023 https://doi.org/10.3390/app13053026
[45] A. A. Taleb, Q. A. Al-Haija, and A. Odeh, “Efficient Mobile Sink
Routing in Wireless Sensor Networks Using Bipartite Graphs,” Future
Internet, vol. 15, no. 5, 182, pp. 1-17, 2023.
[46] M. K. Dholey, D. Sinha and S. Mukherjee, “MSHRP: Mobile Sink Based
Limited Hop Routing Protocol for Wireless Sensor Networks,” Wireless
Pers Commun, 2023.