Mathematical Statistician and Engineering Applications

ISSN: 2094-0343
2326-9865
Energy-Efficient Hybrid Routing Protocol to Extend the Network
Lifetime in IoT Applications
Jyoti Sharma1, Surendra Kumar Patel2*, V. K. Patle3
1
Research Scholar SoS in Computer Science & IT

Pt.Ravishankar Shukla University

Raipur, Chhattisgarh, India.

2
Assitant Professor

Dept.of Information Technology,

Govt.Nagarjuna P.G. College of Science, Raipur, India

3
Assistant Professor SoS in Computer Science & IT

Pt.Ravishankar Shukla University

Raipur, Chhattisgarh, India.

*corresponding author
Article Info Abstract - A Wireless Sensor Network (WSN) is made up of several
Page Number: 967-977 inexpensive, low-power, compact sensor nodes that are densely placed
Publication Issue: across the monitoring area. Wireless connection creates a multi-hop
Vol. 71 No. 3s (2022) wireless network system. WSN is particularly well suited for deployment in
harsh environments and remote monitoring locations that are not suitable
for personnel. It has significant advantages like a large coverage area and
broad application prospects in the fields of military, environmental
monitoring, industrial control, and urban transportation. IoT-enabled
networks, however, suffer a variety of difficulties because of the enormous
heterogeneous data generated by many sensing devices, including long
communication delays, limited throughput, and short network lifetimes. In
this study, a hybrid cluster-based routing protocol model is suggested that
makes use of the advantages of both butterfly and particle swarm
optimization methods. The suggested approach splits the network into many
clusters and chooses the best node as the cluster leader to exacerbate the
Article History network's premature demise. The typical cluster-based routing protocols
Article Received: 22 April 2022 PSO and BOA are assessed using simulation results in terms of the quantity
Revised: 10 May 2022 of alive nodes, throughput, and remaining energy of the nodes.
Accepted: 15 June 2022
Publication: 19 July 2022 Keywords-Butterfly Optimization, Particle Swarm Optimization, Energy
optimization, Cluster based IoT Routing, Hybrid Routing Protocols.

1. INTRODUCTION

In wireless sensor networks, the energy is consumed when the communication is

between nodes to nodes and between nodes and base stations, so efficient routing protocols
have a crucial impact on the performance of the entire network. Clustering routing protocols,

Vol. 71 No. 3s (2022) 967

http://philstat.org.ph
 Mathematical Statistician and Engineering Applications
ISSN: 2094-0343
2326-9865
a class of efficient wireless Compared with the plane routing protocol, the sensor network
routing protocol can better balance the energy consumption of the network, improve the
efficiency of energy utilization, and prolong the life cycle of the network. Because the Internet
of Things (IoT) demands a lot of power to sense, process, and send data, the application of IoT
to Wireless Sensor Networks (WSNs) presents a big barrier for network durability. As a result,
there are multiple known algorithms that combine diverse optimization methodologies to
increase the performance of WSN networks. Many energy-efficient clustering routing
protocols have been suggested in recent years as more and more academics have concentrated
on the study of wireless sensor network clustering routing protocols: Through the cluster head
node's rotational mechanism, Leach [1], a common low-power clustering routing protocol,
assures the nodes' energy efficiency. PEGASIS [2] is an energy-efficient protocol. The Leach
protocol's frequent cluster head replacements consume a lot of communication energy; Leach-
C [3] is an enhancement to the Leach protocol. The life cycle of the network is prolonged by
reducing the energy of the candidate cluster heads during the cluster head selection process.
protocol balances the network's energy usage by introducing sophisticated nodes, making full
advantage of the nodes' heterogeneity; PSO-C [4] To enhance the performance of the network,
the Leach protocol's cluster head selection procedure is optimised using the Particle Swarm
Optimization method (PSO) [5].

A routing protocol based on Particle Swarm Optimization (PSO) and the Butterfly
Optimization Algorithm is proposed in this study (BOA). The method for hybrid meta-heuristic
algorithms, which integrates chaos theory with the fundamental PSO and BOA, is also used to
improve the algorithm's capacity for the optimization problems we specified. Additionally, a
nonlinear control method is presented for adjusting the global and local search capabilities of
the improved algorithm.

The remainder of the paper is organised as follows. The earlier study on energy conservation
in IoT networks is described in Section 2. The suggested model is described in Section 3. The
suggested method's simulation results are discussed in Section 4, and the paper's mentioned
sources are listed in Section 5.

2. LITERATURE SURVEY

Wireless sensor networks are gradually becoming a new attractive communication

mode due to their low cost and rapid deployment [6]. In most sensor networks, reducing energy
consumption is the main challenge [7]. Battery charge must be minimized because sensor
networks are typically made to operate unattended for extended periods of time [8]. This will
increase the lifespan of each sensor and the network as a whole. Energy efficiency is one of the
primary factors in the design of routing protocols for wireless sensor networks, and it should
always be taken into account when a routing protocol is applied to a sensor network [9–10]. At
the moment, a growing number of WNS protocols are created with consideration for energy
efficiency, including scale-free topology models for energy-efficient wireless sensor networks
[12] and energy-efficient data collecting protocols for heterogeneous wireless sensor networks
[11].

Vol. 71 No. 3s (2022) 968

http://philstat.org.ph
 Mathematical Statistician and Engineering Applications
ISSN: 2094-0343
2326-9865
Energy-efficient routing methods for wireless sensor networks have been the subject of
extensive research. For wireless sensor networks with mobile sinks, Tunca et al. (2015) [13]
offer an energy-saving ring routing system that minimises the energy consumption and packet
delay caused by the position broadcast of mobile sinks to the network. For wireless sensor
networks, Kuila and Janap (2014) [14] suggest PSO-based routing and clustering. The
clustering algorithm balances the energy consumption of CH, while PSO routing provides a
trade-off between transmission distance and hop count. Su and Zhang (2016) [15] provide a
non-uniform clustering-based wireless sensor network routing algorithm that chooses cluster
heads based on many factors (such as amount of energy left, distance from nodes, etc.) and
delivers data using a combination of single-hop and multi-hop techniques. This approach can
lessen the issue of energy holes and increase network longevity.

In opportunistic routing, the distribution of nodes is sparse, and the network topology
is constantly changing. The data method is multi-hop, and its forwarding node selects the final
forwarding node from multiple candidate nodes. Higher transmission reliability and end-to-end
throughput are two of its benefits. quantity. Opportunistic routing enables forwarder list nodes
to take part in packet protection and make opportunistic selections depending on the link
quality circumstances to get routing pathways [16]. These nodes may listen in on transmissions
and are closer to the destination. Shi et al., (2017) [17] proposes an opportunistic routing
protocol with optimal cooperative delay between candidate nodes, realizes the mechanism
modeling of the cooperative forwarding process of candidate nodes, and obtains the expected
end-to-end cooperative delay, which is used as a routing measure to increase the network
throughput, reduces the average end-to-end delay. So an Byun (2014) [18] proposes
opportunistic routing for intra-network aggregation of asynchronous duty cycle wireless sensor
networks, which can increase the duty cycle of nodes that keep data packets and increase the
chance of intra-network aggregation, thereby reducing energy consumption and prolonging the
network life cycle.

In multi-hop sensor networks, Zhao (2015) [19] introduced an energy-aware

opportunistic routing protocol and a novel routing measure based on the idea of balancing link
quality and energy. Based on this criterion, a candidate set selection was created. Mechanisms
for node coordination and algorithms. Yao (2018) [20] proposes an opportunistic routing
optimization algorithm for WSN based on network coding, which reduces the number of
retransmissions of coded packets by calculating the failure probability of receiving coded
packets, so as to prolong the network lifetime and reduce the average energy consumption. A
queue-based swarm optimization approach was presented by Praveen and Joe (2021) [21] to
choose a better route for a future route based on numerous restrictions, which enhances the
route-discovering process. In terms of energy consumption, node lifespan, throughput, end-to-
end latency, packet delivery ratio, and packet overheads, the proposed ECRR approach is
implemented in the Network Simulator (NS-2) tool. The simulation results are then compared
with the current state-of-the-art techniques. In energy-efficient green-IoT networks with a
mobile sink, Yarinezhad & Azizi (2021) [22] presented a tree-based routing protocol that is
efficient in power consumption and decreases end-to-end latency. The suggested protocol adds
two new, distinct ways for controlling network routing. For IoT-enabled WSNs, Kaur et al.

Vol. 71 No. 3s (2022) 969

http://philstat.org.ph
 Mathematical Statistician and Engineering Applications
ISSN: 2094-0343
2326-9865
(2021) suggested an intelligent routing system based on Deep Reinforcement-Learning (DRL)
that greatly reduces latency and extends network lifetime. The suggested approach separates
the whole network into many uneven groups based on the data load that is now present in the
sensor node, considerably preventing the network from dying prematurely.

It is evident from the aforementioned studies conducted by various academics that

much work is being done on energy-efficient routing in various IoT network topologies. The
capacity of clustering to achieve energy efficiency has been demonstrated in a variety of
methods, which inspired the authors to conduct this research.

3. NETWORK MODEL

The network models utilised in this study are discussed in this section. The used system
model, together with its setup settings and presumptions, are briefly presented. This study takes
into account a two-dimensional network model with sensor nodes arranged in a clustered
topology. There was a single base station that gathered all of the IoT nodes coming from the
source. The energy consumption model for the transmission and reception of messages with
the same length, n bits, was a first order radio model.

3.1 GENERAL PARTICLE SWARM OPTIMIZATION FOR OPTIMIZED CLUSTER HEAD

SELECTION

The clustering routing algorithm based on PSO optimization needs to firstly divide the
network into clusters, then use the PSO algorithm to synthesize the state information of
neighbor nodes to optimize the selection of cluster heads, and finally publish the determined
cluster head information to the entire cluster.

(1) The initial stage of cluster formation.

Several clustering techniques, like the LEACH algorithm, can be utilised at this point to cluster
the wireless sensor network in an initial manner. The cluster heads and clusters are essentially
established after this stage, although the generated clusters are vulnerable to the blind nodes
issue. At this point, each cluster's constructed cluster head is referred to as the auxiliary cluster
head.

(2) The auxiliary cluster head gathers data about the cluster nodes' states. The position and
energy status data of each neighbour node in the cluster is sent to the auxiliary cluster head.
The auxiliary cluster head now stores the location information P (p1, p2,..., pn) and energy
information E of the neighbouring nodes. where n is the number of nodes in the cluster, pi is
the ith node's position value, and ei is the ith node's energy value

(3) Use the PSO technique to find the cluster head stage and to optimise.

The position updation formulas in the normal PSO method must be updated in order to
make the algorithm acceptable for this problem domain, and the corresponding fitness function
f(x) is provided at the same time. This step is the heart of the optimization algorithm. The
search is carried out on the network plane, and as a result, the velocity is a vector with
components in the x and y directions as well as a magnitude. so there are,

Vol. 71 No. 3s (2022) 970

http://philstat.org.ph
 Mathematical Statistician and Engineering Applications
ISSN: 2094-0343
2326-9865
Vxid=WVxid + c1rand() . (Pid-Xxid) + c2 rand() . (Pgd -Xxid) --- (1)

Vyid=WVyid + c1rand() . (Pid-Xyid) + c2 rand() . (Pgd -Xyid) --- (2)

Similarly, there are position components in the x and y directions, we have,

Xxid = Xxid + Vxid --- (3)

Xyid = Xyid + Vyid --- (4)

Since the distribution of nodes in the wireless sensor network is discrete, the values
calculated by the nodes according to equations (3) and (4) cannot be mapped to the
corresponding network nodes one by one. Xxid ∈ {px1, px2, …, pxn }, Xyid ∈ {py1, py2, …, pyn },
where pxi is the x component of the ith node in the cluster, and pyi is the ith node in the cluster y
component of i nodes.

Let Δpxj = ûXxid - pxjû, Δpyj = ûXyid - pyjû, Δpj = (Δpxj)2 + (Δpyj)2. where Δpxj represents
the relationship between Xxid and network node j in the cluster. The absolute value of the x-
component difference, Δpyj represents the absolute value of the y-component difference
between Xyid and the network node j in the cluster, and Δpk = min{Δp1, Δp2, …, Δpn }, it means
that the position of the kth node in the network is the closest to Xid, so the adjusted value is
Xxid≈ pxk , Xyid≈ pyk , that is, the search point is now located at the position of node k.

The properties of the issue domain are directly tied to how the fitness function is
determined. A node's fitness should take into account both the magnitude of its own energy as
well as the energy distribution of the nodes around it. The neighbouring node's energy should
increase with distance from the node. The distance from the node should instead be greater.
The fitness function is built using this feature, and the closer neighbour nodes should have less
energy.

f(k) = ηek+ λê ---(5)

where: η+ λ= 1 and η and λ ∈ [ 0, 1 ]; e is the equivalent average energy of other nodes; k is

the current network node number; η is the current node energy influence factor, λ is the
neighbor node energy influence By adjusting the factor, the contribution ratio of neighbor
nodes to the fitness value can be determined.

At this point, the auxiliary cluster head node broadcasts the optimised cluster head
information to the cluster nodes and enables the optimised cluster head to acquire the
information of the cluster nodes. The energy loss of the nodes in the cluster is balanced because
the building of the optimised cluster carefully takes into account the state information of the
neighbouring nodes, thereby preventing the frequent occurrence of blind nodes.

3.2 GENERAL BUTTERFLY OPTIMIZATION ALGORITHM

In the standard BOA algorithm, the butterfly population uses random initialization of
the butterfly population:

XNxd = lb + rand (N,d) x (ub – lb) ---(6)

Vol. 71 No. 3s (2022) 971

http://philstat.org.ph
 Mathematical Statistician and Engineering Applications
ISSN: 2094-0343
2326-9865
In the equation (6), N represents the butterfly population size, and d represents the
search space dimension.

The butterfly algorithm assumes that each individual in the search space can perceive
each other's fragrance; the butterfly can move randomly or move towards the best butterfly
according to the intensity of the fragrance concentration; the sensory mode of the butterfly is
affected or determined by the range of the objective function. The transition probability P is
utilised to govern the butterfly's search mode, which may be either local or global, while it is
looking for food sources because the magnitude of the fragrance is impacted and constrained
by elements like wind, rain, temperature, etc.

Butterfly fragrance calculation as in the equation (7):

f=cIa ---(7)

c=0.01 + (0.025/0.1Tmax) ---(8)

where f is the size of the scent, c is the intensity of the sensory modality, I is the intensity
of the stimulus, an is the power exponent that depends on the sensory modality, and Tmax is
the maximum number of repetitions. The butterfly uses both local and global search to update
its position. If a butterfly can smell other butterflies but can't because of distance or other
natural causes, it will randomly scan its immediate area; if it can smell other butterflies, it will
fly to the best butterfly based on the strength of the scent. The populace hunts for nearby food
sources as a result of location updates.

When r1≥p, perform a local search:

dis1 = r12 xjt − xit − − − (9)

xit+1 = xit + dis1 fi − − − (10)

Among them, dis1 is the distance formula between two butterflies in the local search,
xit xit
represents the position of the jth and ith butterflies in the tth iteration, and if represents the
fragrance size of the ith butterfly.

When r1<p, perform a global search:

dis2 = r22 g ∗ − xit − − − (11)

xit+1 = xit + dis2 fi − − − (12)

Among them, dis2 is the distance between the current butterfly and the best butterfly in
the global search, *g is the position of the best butterfly, and x it is the position of the ith butterfly
in the tth iteration. Equations (10) and (12) can simulate the feeding behavior of butterfly
populations.

Vol. 71 No. 3s (2022) 972

http://philstat.org.ph
 Mathematical Statistician and Engineering Applications
ISSN: 2094-0343
2326-9865
3.3 PROPOSED HYBRID CHAOTIC PARTICLE SWARM OPTIMIZATION AND BUTTERFLY
OPTIMIZATION ALGORITHM

In this part, a novel hybrid chaotic HCPSOBOA that combines different PSO and BOAs
is suggested. The primary distinction between PSO and BOA is how new members are
produced. The PSO approach has the drawback of being able to tackle high-dimensional
optimization problems only in a small area. To optimise the advantages of both methods, we
combine their capabilities rather than applying them sequentially. In other words, the two
algorithms' end results are heterogeneous because of the method employed to get them. Here,
a brand-new hybrid method is put forward, along with the initialization of BOA using a cubic
one-dimensional map and a nonlinear parameter management technique. Additionally, BOA
and PSO are combined to enhance the fundamental BOA for global optimization. The cubic
map for the initial population, the nonlinear parameter management technique of power
exponent a, the PSO algorithm, and BOA are then combined to form the unique Hybrid Chaotic
HCPSOBOA.

Algorithm 2. Proposed Algorithm (HCPSOBOA)

Step 1: Generate the initial population of butterflies using cubic map and initialize the
parameter r1, r2, c1, c2, switch probability, senser modality c and power exponent a.

Step 2: Evaluate the fitness of each butterflies.

Step 3: Update the fragrance of each search agent in a population by Equation fi=cIa

Step 4: Find the best fitness f.

Step 5: Based on the search criteria either move towards best position(r<p) by equation
(12) or random (r>=p) by equation (10).

Step 6: Update velocity using equation 11and calculate the new butterflies' fitness and
update the best fitness f.

Step 7: If newf < best f then updates the best f using equation Xt+1
i = Xti + Vt+1 .

Step 8: Evaluate the power exponent using equation

π t 2
a(t) = afirst − (afirst − afinal ). sin( ) ( ) )
μ Tmax

Step 9: Repeat the steps 3 to step 8 until termination criteria reached.

In the field, butterflies can search both locally and globally for food and a mate. A
switch probability p is given to transform the regular global search and the intense local search.
In order to decide whether to do a local or global search, the BOA generates a number at
random in the range [0,1] for each iteration. This number is then compared to the switch
probability p. It is clear from the suggested model that during the search phase, the chaotic map

Vol. 71 No. 3s (2022) 973

http://philstat.org.ph
 Mathematical Statistician and Engineering Applications
ISSN: 2094-0343
2326-9865
can disperse the butterfly population to a random value in the range (0, 1). The cubic map was
recommended to initialise the location of the algorithm, and the proposed approach sets the
cubic map's z(0) value to 0.315 to guarantee that the initialised interval is in the range (0, 1).

4. RESULT OF SIMULATION

Network Simulator 3 software is used to run the simulation. The proposed routing architecture
is first subjected to three types of jamming assaults, with activity limitations set between (0.1
to 0.9). It has been noted that the proposed strategy gradually increases the packet delivery
ratio. Standard cluster-based protocols BOA and PSO are compared in order to assess the
performance of the proposed model. In Table 1, simulation setups are displayed.

Table 1. Simulation Parameters

Parameter Value

Area for simulation 1000*1000 meter

Probability 0.3

Receiver energy 20*10-8

Nodes 100

Bandwidth of the channel 2mhz

IoT threshold 3db

Transmission Energy 0.1

Transmitter Energy 20*10-8

Maximum lifetime 3*10-8

Jamming Duration 3, 0.2, 3, 1.1, 4, 2.5, 1, 1.4 ,7, 0.1ms

The parameters such as network lifetime, average energy, harvested energy are used to
evaluate the performance. The network parameters are set up with different nodes between 50
to 200 nodes. The number of packets transmitted successfully gives the network throughput.
The stability period is chosen for estimating the average energy with the throughput of the
receiver nodes.

Table 2. Energy Comparison with Proposed HCPSOBOA

Proposed
Nodes BOA PSO
HCPSOBOA

50 80.7 77.2 76.8

Vol. 71 No. 3s (2022) 974

http://philstat.org.ph
 Mathematical Statistician and Engineering Applications
ISSN: 2094-0343
2326-9865
100 78.2 75.7 74.3

150 75.8 73.1 72.6

200 72.9 71.5 70.7

10
Throughput (Mbps)
5
0
10 20 30 40
Data Load (MB)

Proposed HCPSOBOA BOA PSO

Figure 1. Throughput Analysis

The throughput performance is shown in Figure 1, and it is evident that the suggested
HCPSOBOA has greatly outperformed both the BOA and PSO.

30
Alive nodes

20
10
0
1000 2000 3000 4000 5000 6000
Number of iterations

Proposed HCPSOBOA BOA PSO

Figure 2. Network Lifetime Estimation

Figure 2 shows the Network life time performance when it is compared to the node that
is still operational even after an assault, and it is evident that the suggested approach has been
greatly enhanced. The time it takes for each node in the network to reach the maximum amount
of time a node may stay alive in the network is used to calculate the average energy.

20
15
Energy(J)

10
5
0
1000 2000 3000 4000 5000 6000
Total Iterations

Proposed HCPSOBOA BOA PSO

Figure 3. Comparison of Residual Energy of Proposed Model with BOA and PSO

Vol. 71 No. 3s (2022) 975

http://philstat.org.ph
 Mathematical Statistician and Engineering Applications
ISSN: 2094-0343
2326-9865
Figure 3 shows the energy performance, and it is evident from the graph that the
proposed approach has significantly increased energy savings. When compared to BOA and
PSO based approaches, the suggested HCPSOBOA method retains the greatest amount of
residual energy during the simulation.

5. CONCLUSION

This paper proposes a unique hybrid chaotic particle swarm optimization technique
with butterfly optimization. In order to address the energy issues in cluster-based routing in
IoT-enabled wireless sensor networks, the suggested model makes use of the advantages both
optimization methods. When the IoT uses this hybrid approach, data is gathered through
clustering, and energy-conscious devices are used. It is adequately proved via the simulation
results and data analysis that the proposed routing protocol improves the balance of energy
consumption throughout the whole network and accomplishes the goal of prolonging the
network life cycle.

REFERENCES

1. Heinzelman WR, Chandrakasan A, Balakrishnan H. Energy-efficient communication

protocol for wireless microsensor networks. Annual Hawaii International Conference
on System Sciences. 2000.
2. Lindsey S, Raghavendra C, Sivalingam KM. Data gathering algorithms in sensor
networks using energy metrics. IEEE Trans. on Parallel and Distributed Systems, 2002,
13(9): 924-935.
3. Heinzelman WR, Chandrakasan AP, Balakrishnan H. An application-specific protocol
architecture for wireless microsensor networks. IEEE Trans. on Wireless
communications, 2002, 1(4): 660-670.
4. Latiff NA, Tsimenidis C, Sharif B. Energy-aware clustering for wireless sensor
networks using particle swarm optimization. Personal, Indoor and Mobile Radio
Communications, 2007: 1-5.
5. Kennedy J. Particle swarm optimization. Encyclopedia of Machine Learning, 2010.
760-766.
6. Li Wenfeng, Fu Xiuwen. Invulnerability of wireless sensor network [J]. Chinese
Journal of Computer, 2015, 38(3): 625-647.
7. Brar, G.S., Rani, S., Chopra, V., Malhotra, R., Song, H., & Ahmed, S.H. (2016). Energy
Efficient Direction-Based PDORP Routing Protocol for WSN. IEEE Access, 4, 3182-
3194.
8. P. Nayak and A. Devulapalli, "A Fuzzy Logic-Based Clustering Algorithm for WSN to
Extend the Network Lifetime," in IEEE Sensors Journal, vol. 16, no. 1, pp. 137-144,
Jan.1, 2016, doi: 10.1109/JSEN.2015.2472970.
9. Ang, Liul, Jiangj, Analysis software-efficiency city connected targeted coverage
algorithms for industrial Wireless sensor networks[J]. IEEE Transaction Industrial
Information, 2017, 13(1): 135-143.
10. Cheng Guoying, Wang Yan. Multi-mobile agent cooperative control data offloading
method in wireless sensor network [J]. Computer Applications, 2015, 35(4): 910-915.

Vol. 71 No. 3s (2022) 976

http://philstat.org.ph
 Mathematical Statistician and Engineering Applications
ISSN: 2094-0343
2326-9865
11. Yaoy, Caoq, Vasilakosav. Edal: An energy-efficient, delay-aware, and lifetime-
balancing data collection Protocol for heterogeneous wireless sensor networks[J].
IEEE/ACM Transaction on Networking, 2015, 23(3):810-823.
12. Liu Zhouzhou, Wang Fubao. Energy Efficient Wireless Sensor Networks Scale-Free
Topological model [J]. Journal of Beijing University of Posts and Telecommunications,
2015, 38(1): 87-91.
13. Deepak Mathur, N. K. V. . (2022). Analysis &amp; Prediction of Road Accident Data
for NH-19/44. International Journal on Recent Technologies in Mechanical and
Electrical Engineering, 9(2), 13–33. https://doi.org/10.17762/ijrmee.v9i2.366
14. Tunca C, Isik S, Donmezm Y, Ring routing: An energy-efficient routing protocol for
wireless sensor networks with a mobile sink[J]. IEEE Transactions on Mobile
Computing, 2015, 14(9): 1947-1960.
15. Kuila P, Janap K. Energy efficient clustering and routing algorithms for wireless sensor
networks: Section Swarm optimization approach [J]. Engineering Application of
Artificial Intelligence, 2014, 33:127-140.
16. Su Bing, Zhang Yujing. Wireless sensor network based on non-uniform clustering by
the agreement [J]. Computer Measurement and Control, 2016, 24(2): 325-327.
17. Jadhavp, Sataor. A survey on opportunistic routing Protocols for wireless sensor
networks [J]. Procedure Computer Science, 2016, 79: 603-609.
18. Shi Xiaodong, Yang Bo, Hao Jun, The optimal coordination delay between candidate
nodes Opportunistic routing [J]. Computer Engineering and Design, 2017, 38(3): 576-
580.
19. So J, Byun H. Opportunistic routing within-network aggregation for asynchronous duty
cycled wireless sensor networks[J]. Wireless networks, 2014, 20(5): 833-846.
20. Zhao Jie. Opportunistic Routing Algorithm Based on Energy Awareness in Multi-hop
Sensor Networks Law [J]. Journal of Jilin University (Science Edition), 2015, 53(6):
1241-1245.
21. Yao Yukun, Zhang Yi, Li Juan. Opportunistic Routing of WSN Based on Network
Coding Optimization algorithm [J]. Computer Engineering, 2018, 44(6): 68-73.
22. Praveen, K.V., Prathap, P.M.J. Energy Efficient Congestion Aware Resource
Allocation and Routing Protocol for IoT Network using Hybrid Optimization
Techniques. Wireless Pers Commun 117, 1187–1207 (2021).
https://doi.org/10.1007/s11277-020-07917-8.
23. Yarinezhad, R., & Azizi, S. (2021). An energy-efficient routing protocol for the Internet
of Things networks based on geographical location and link quality. Computer
Networks, 193, 108116.
24. Kaur, G., Chanak, P., & Bhattacharya, M. (2021). Energy efficient intelligent routing
scheme for IoT-enabled WSNs. IEEE Internet of Things Journal.
25. Nouby M. Ghazaly, A. H. H. . (2022). A Review of Using Natural Gas in Internal
Combustion Engines. International Journal on Recent Technologies in Mechanical and
Electrical Engineering, 9(2), 07–12. https://doi.org/10.17762/ijrmee.v9i2.365

Vol. 71 No. 3s (2022) 977

http://philstat.org.ph