Wireless Sensor Networks Based on Multi-Criteria Clustering and Optimal Bio-Inspired Algorithm for Energy-Efficient Routing

Abstract

Wireless sensor networks (WSNs) are used for recording the information from the physical surroundings and transmitting the gathered records to a principal location via extensively disbursed sensor nodes. The proliferation of sensor devices and advances in size, deployment costs, and user-friendly interfaces have spawned numerous WSN applications. The WSN should use a routing protocol to send information to the sink over a low-cost link. One of the foremost vital problems is the restricted energy of the sensing element and, therefore, the high energy is consumed throughout the time. An energy-efficient routing may increase the lifetime by consuming less energy. Taking this into consideration, this paper provides a multi-criteria clustering and optimal bio-inspired routing algorithmic rule to reinforce network lifetime, to increase the operational time of WSN-based applications and make robust clusters. Clustering is a good methodology of information aggregation that increases the lifetime by group formation. Multi-criteria clustering is used to select the optimal cluster head (CH). After proper selection of the CH, moth flame and salp swarm optimization algorithms are combined to analyze the quality route for transmitting information from the CH to the sink and expand the steadiness of the network. The proposed method is analyzed and contrasted with previous techniques, with parameters such as energy consumption, throughput, end-to-end delay, latency, lifetime, and packet delivery rate. Consumption of energy is minimized by up to 18.6% and network life is increased up to 6% longer compared to other routing protocols.

1. Introduction

The WSN consists of many tiny, low-priced sensor devices that are used in many recent applications [1]. It can be easily and efficiently connected wirelessly with multiple sensor devices. The WSNs are very flexible and easily modified due to their self-organizing capabilities. WSNs are preferred since it is used in many applications, such as multimedia information transmission, multimedia device tracking, populace tracking, improvising commercial process, fitness tracking, and so on [2]. These networks are maintained to detect event information and transmit it to the BS for further processing.

While the WSN has many challenges to overcome, energy conservation is a foremost challenge. A tiny battery that is connected to a device provides the power needed to carry out multiple functions. These batteries are rapidly depleted when they carry out several functions, such as data processing and transmission. Such energy consumption will result in the network failing or reducing the lifetime [3]. Therefore, during the data transmission, energy efficiency is important to extend the life of the WSN.

For regulating the power needs of WSNs, several techniques have been presented by researchers. Heinzelman’s clustering methods are used to achieve effective energy savings [4]. Routing protocols are categorized according to their logical structure in order to increase their resilience. In the first type of planar routing, each node in the network plays the same duty and has no special nodes. The resilience of this kind of protocol is a benefit. Clustering is one of the oldest hierarchical routing schemes, where the cluster serves as the infrastructure and the nodes perform various functions. This technique includes creating clusters, or small groups of nodes, within the network. This network structure consists of a two-level hierarchy.

In clustering, the WSN is arranged into groups known as clusters [5] and they are controlled by cluster heads (CHs) [6]. The contiguous cluster heads are the top-stage nodes, and the remaining are the next-level nodes. The CH on their respective cluster is to perform processes for instance data collection and aggregation. The CH in each cluster collects information from the neighbor SN and transmits it to the sink with the help of neighbor nodes. All CHs in the cluster devour extra resources compared to child nodes of the cluster. LEACH (low-energy adaptive clustering hierarchy) is constructed on the pinnacle with the idea that the strength of all nodes is identical in the polling even as HEED(hybrid energy efficient distributed clustering) considers the change in strength within the nodes to optimize the existence of the network. This can cause uneven traffic distribution throughout distinctive center node clusters.

LEACH [7] and HEED [8] are two different typical models and they differ in CH selection. The main assumption of LEACH is all nodes having equal energy in polling while HEED assumes energy change within the node to optimize the network lifetime.

Kumar et al. suggested a clustering protocol for a heterogeneous model called the energy-efficient clustering (EEC) scheme [9]. Based on the initial energy, SNs are categorized into regular nodes (RN), advanced nodes (AN), and super nodes (SN). The RN have the smallest amount of energy, the AN have more energy than the RN, and the SN have the uppermost energy. Energy aware evolutionary routing protocol [10], evolution based clustering routing protocol, stable aware and energy efficient routing protocols with stable sensitivity thresholds use DE, HSA [11] and SMO [12].

Kuila and Jana recommended a differential evolution (DE)-based protocol [13] to extend the lifetime considerably by how to balance the lifespan of the CH. Nodes with the minimum fitness value are selected as the CH [14,15]. Moth flame optimization (MF) is inspired by nature’s recent algorithm, which is driven by the horizontal point of a moth reference in the wild [16]. Transverse navigation represents the moth’s progress towards the moon on a straight line. In this work, the MF algorithm is demoralized to maximum potential to answer the upper load balancing problem in the clustering algorithm. The CH selection is a challenging non deterministic task. To address the CH issue, a variety of evolutionary approaches have been proposed, including genetic algorithms, ant colony optimization, and particle swarm optimization [17,18]. For the best outcome in WSNs, these evolutionary strategies seek to resolve energy-related restrictions [19,20]. El Alami and Najid [21] implemented hierarchy clustering techniques for impoverished data transmission in a WSN by using the high-density node.

In this paper, multi-criteria clustering and bio-inspired routings such as moth flame and salp swarm optimizer are combined to diminish the energy consumption, which will get a better lifetime. The clusters are organized in grids for finding the optimal CH based on the decision-making approach. Once the CH is chosen, the combined routing methods are implemented to find the shortest path for transmission. The flow diagram of the proposed approach is portrayed in Figure 1. Immediately after the generation of population through the hybrid optimization algorithm; the transmission rate can be estimated.

2. Methodology

2.1. Energy Model

It is the radio energy model [22] which determines the amount of energy consumed during the transmission of data in the WSN. This model sends a one-bit message from the sender ‘a’ to the nearby receiver ‘b’ to estimate the energy consumption $e\left(a,b\right)$ with a distance $dis\left(a,b\right)$ and it is determined by

$$e\left(a,b\right)=Z_1+Z_2\ast dis\left(a,b\right)\ast \gamma$$

(1)

where $Z_1$ and $Z_2$ are constants with path loss of ‘$\mathsf{\gamma }$’. Z₁ = C_s + C_r, Z₂ is a constant depending on the given wireless devices and application environment. For a given network, C_s and C_r are two constants depending on the characteristics of the wireless devices. We assume that each sensor node has the same $Z_1$ and $Z_2$.

The total area is partitioned into grids. The grid formation process is shown in Figure 2 and is provided below:

Step 1: Total region is divided as ‘s’ number of sub-regions which will be similar to the ‘y’ axis with p and q as width and length, respectively. The sub-regions are numbered from left to right (1 to s). The obtained regions are arranged into the ‘s’ overlapping cell to form a grid.

Step 2: Each grid is represented by the location and distance measurement among the sink nodes and it is provided with a unique identity.

Step 3: It is important to cover all the sensors which are present inside the boundaries. Thus, the grid structure is considered with the parameters of cell density/distance. The node position inside grid boundaries is provided by Equation (2).

$$\{\begin{array}{c}\left(col-1\right)\times p\times b\le col\times p;\\ {\displaystyle {\displaystyle \sum }_{l=0}^{x-1}}heigh{t}_{col.l}<q\le {\displaystyle {\displaystyle \sum }_{l=o}^x}heigh{t}_{col.l}; col=1,2,\dots ,s, x=1,2,\dots ,s\end{array}$$

(2)

Step 4: For achieving optimized cluster information, Steps 2 and 3 could be repeated. The grid ID for the corresponding grid is provided in

Table 1. Grid ID representation in a sensor network area.

(5,1)	(5,2)	(5,3)	(5,4)	(5,5)
(4,1)	(4,2)	(4,3)	(4,4)	(4,5)
(3,1)	(3,2)	(3,3)	(3,4)	(3,5)
(2,1)	(2,2)	(2,3)	(2,4)	(2,5)
(1,1)	(1,2)	(1,3)	(1,4)	(1,5)

.

2.2. Multi-Criteria Clustering

Multi-criteria clustering is considered to be an important advancement in the field of complex decision-making problems with many objectives. The six nominal things are taken for optimal CH election: node energy, node degree, cluster formation, neighborhood node distance, cluster distance, and residual energy. These parameters are utilized as an energy efficient clustering approach. The multi-criteria clustering is aimed to minimize cell impact in the outer boundaries with the aid of grid structure.

(i) Node energy

It is important to conserve energy at the time of information transmission. This model was proposed by Logambigai et al. [23] for minimizing the energy consumption.

$$NE=\frac{Optimal No. of clusters}{Size of CH set}$$

(3)

(ii) Node degree

The node degree provides the summation of approachable cluster members and their balance with loads of CH.

$$ND={\displaystyle \sum }_{i=1}^{m}C{M}_i$$

(4)

where $C{M}_i$ indicates the ith member of CH.

(iii) Cluster formation

Cluster formation is used to increase the quality of links among clusters. It provides information about the cost required for transmitting the information between the sink nodes (SN) to the CH.

$$CI={\displaystyle \sum }_{i=1}^R\frac{dis\left(C{H}_i,N_m\right)}{Min\left[dis\left(C{H}_i,N_m\right)\right]}$$

(5)

where $C{H}_i$ represents CH at ith member and $N_m$ is number of nodes, R is a number of clusters, and $dis\left(C{H}_j,N_m\right)$ provides the Euclidean distance.

(iv) Neighborhood node distance

Neighborhood node distance provides the ratio between the Euclidean distance and half of the total area. This parameter reduces the distance between the CH and BS.

$$NN=\frac{1}{R}{\displaystyle \sum }_{i=1}^R\frac{dis\left(C{H}_i,BS\right)}{\frac{A^2}{2}}$$

(6)

where $A^2$ is the total area covered by BS.

(v) Intra-Cluster distance

It provides the distance among CH and SN of its members. It aids the improvisation of cluster quality. The value ‘1’ is assigned for a specific node when it is connected to CH; otherwise, the value ‘0’ is assigned.

$$I_{n}C={\displaystyle \sum }_{i=1}^R\left[{\displaystyle \sum }_{i=1}^{\left|C{M}_i\right|}\frac{dis\left(C{H}_i,C{M}_i\right)}{\left|C{M}_i\right|}\right]$$

(7)

where the term $dis\left(C{H}_i,C{M}_i\right)$ represents the Euclidian distance among the ith node of and CH and ith cluster member (CM).

(vi) Residual Energy (ReE)

It provides the details about the available energy for each node after nominating the CH.

$$R_{e}E={\displaystyle \sum }_{i=1}^n\frac{1}{E_{C{H}_i}}$$

(8)

where n is number of CH, $E_{C{H}_i}$ is remaining ith node energy of CH.

2.3. Optimal Bio-Inspired Algorithm

The CH selection is obtained through the multi-criteria decision-making grid-based method. The optimal path is estimated using the optimal bio-inspired optimization algorithm. This methodology combines the MF algorithm [24,25], and the SS algorithm [26] is utilized to obtain optimal routing path. The main phases of the optimal bio-inspired algorithms are initialization, fitness evaluation, updation of solution, and selection of optimal [27,28].

(i) Initialization

The path is routed between the CH and sink nodes as an initial approach to optimization. At first, only one CH is selected and the other CH is treated as a sink or intermediate path. Thus, all the possible paths are represented in Equation (9),

$$K=K_i, i=1,2,\dots ,p$$

(9)

where, ‘$K$’ provides the initial population set as provided in Equation (10), ‘$K_i$’ describes the ith path among CH and destination, and ‘$p$’ gives number of paths.

$$K=\left\{R_{e}E,ND\right\}$$

(10)

where ‘$R_{e}E$’ represents the residual energy, and ‘$ND$’ is distance among line.

The σR_eE is used to estimate characteristics of load distribution among the available sensor nodes and it equated as

$$\sigma R_{e}E=\sqrt{\frac{1}{m}{\displaystyle \sum }_{i=1}^m{\left[\frac{1}{m}{\displaystyle \sum }_{i=1}^m{E}_{\left(nod{e}_i\right)}-e\left(a,b\right)\right]}^2}$$

(11)

where ‘$m$’ gives number of nodes present in path and ‘$E_{\left(nod{e}_i\right)}$’ represents residual energy of ith node.

Euclidean distance measures distance between sender CH to destination and it is represented by Equation (12)

$$dis={\displaystyle \sum }_{i=1}^{m-1}\sqrt{\left[C{H}_i\left(x\right)+C{H}_i\left(y\right)-C{H}_{i+1}\left(x\right)-C{H}_{i+1}\left(y\right)\right]}$$

(12)

where ‘$C{H}_i\left(x\right)$’ and ‘$C{H}_i\left(y\right)$’ give the x and y coordinate of ith path, respectively. Therefore, the energy efficiency is provided by formulating residual energy and minimization of distance amongst the path is provided by Euclidean distance.

(ii) Evaluation of fitness

The main reason for estimating the fitness is to obtain the optimal path with minimal energy and distance requirements. Hence, the objective function is based on the energy and distance on each path.

$$F_i=min\left\{R_{e}E\times ND\right\}$$

(13)

where ‘F_i’ gives the fitness at the ith population, ‘$R_{e}E$’ is the residual energy, and ‘$ND$’ is the total distance.

(iii) Updation of Solution and selection of optimal

In this level, a path is either chosen or taken to revise the solution based on the fitness value. Here, the optimization is carried out by combining the moth search and SSO. A contemporary optimization technique, the MFO algorithm, was introduced in 2015. Since then, it has found extensive usage in both research and a variety of industrial fields. The moths are similar to butterflies, which maintain fixed angle concern to the moon. Getting the solution for the optimization problem depends on the position of the moth.

The next optimization approach is based on the salps method. The salp algorithm navigates as two groups’ heads and followers by using a jellyfish movement concept. These groups are worked together in swarms and search for food. The head of the salp will lead the group and the follower will follow the salp leader. Thus, this salp search algorithm utilizes the low-level nodes to get a high-level network.

Once fitness is estimated, the population is ordered in ascending order, and an equal half of the total d population is updated through arranged levy flight as per moth search optimization and salp swarm rule optimization.

The accelerating factor (AF) is defined as the ratio between high temperatures and lower temperatures. In this work, the AF value is chosen as 1.7 experimentally with an assumption, and assumed that the maximum step size should be greater than zero.

The updation of the CH based on salp swarm optimization is equated as,

$$Z_T^{i+1}=\{\begin{array}{c}{Z}_T^i+\beta levy\left(s\right)i<\frac{m}{2}\\ \frac{1}{2}\left(Z_T^{i+1}+Z_T^i\right)otherwise\end{array}$$

(14)

where ‘$Z_T^i$’ represents the actual cluster head position and ‘$Z_T^{i+1}$’ provides the new CH position with the ‘$T$’ population dimension.

Levy flight of the salp swarm optimization provides the steps obtained so far and it is equated as

$$levy\left(s\right)=\frac{\left[\left(AF-1\right)\sin \left(\pi \left(AF-1\right)\right)\right]}{\pi \left(AF\right)}$$

(15)

The parameter β is represented as the maximum step size, which usually changes its position from one CH to another CH.

$$\beta =\frac{ste{p}_{max}}{T^2}$$

(16)

The SN transmits the information such as the actual position of the node to the nearby network through grid protocol. Once the information is received by the neighboring node, the list will be updated. The drawbacks of the cross-level protocols are resolved with the aid of the moth flame algorithm, which could provide minimal transmission distance to the SN (as shown in Algorithm 1). The CH should be selected with the residual value greater than 0.5 joules.

Algorithm 1. Pseudo code of the proposed Optimal Bio-inspired algorithm

1: Start 2:  Initialization of the new path Ki 3:  Generation of maximum path MG 4:  Maximum step size (Step max), β and acceleration factor (AF) 5:    Iteration = 1, 6:    Fitness evaluation using equation 7:  While iteration < MG 8:  Then 9:  Compute the average of all the population/path 10: If i < M/2 11: Then 12: Update the CH position based on moth flame algorithm using Equation (13) 13: Else 14: Update the CH position based on salp swarm algorithm using Equation (14) 15: End if 16: Newly updated position iteration = iteration +1 17: End While 18: Output optimal path 19: End

3. Results and Discussion

The performance of proposed methodology is compared and analyzed with other methods such as MFO, SSO, Fuzzy-PSO, and LEACH. The parameters such as energy consumption, throughput, end-to-end delay, latency, network lifetime, and PDR are estimated with some simulated parameters tabulated in

Table 2. Simulation parameters.

Parameters	Values
Simulation Tool	Matlab 2018 a
MAC protocol	IEEE 802.11 p
Deployment area	450 m × 450 m
BS Location	(225,225)
Transmission energy per node	0.8 × 10−3 J
Receiving energy per node	2.9 × 10−3 J
Initial node energy Depend	0.7 J
Initial network energy	Varying based on the number of nodes
Bandwidth	9 Mbps
Number of nodes	100–500
Packet Size	512 Kb
Number of rounds	2400
Type of Antenna	Omni directional
Bit Error Rate	2%
Transmission Delay	3.1 s

.

(i) Energy consumption

It is a parameter which is used to estimate the energy consumed quantity by sink. Energy demoralized by CH is nominated as $C{H}_i\left(m\right)$ and energy used by cluster memberis $C{M}_i\left(m\right)$.

$$Energy consumption=\left\{{\displaystyle \sum }_{m=1}C{H}_i\left(m\right)+{\displaystyle \sum }_{m=1}C{M}_i\left(m\right)\right\}$$

(17)

Figure 3 provides the outcome obtained for estimating the energy consumed for various nodes utilizing various techniques.

Table 3. Comparison of energy consumption with various routing protocols.

Number of Nodes	Proposed Approach	MFO	SSO	Fuzzy-PSO	LEACH
100	0.1621	0.2369	0.4985	0.5856	0.8567
200	0.2323	0.3065	0.5500	0.7528	0.9851
300	0.3078	0.3967	0.5748	0.7589	0.9763
400	0.3356	0.4135	0.6012	0.8139	1.1331
500	0.3981	0.4689	0.6779	0.8320	1.0356

represents a proposed work comparison with other methods for energy use calculation. The proposed methodology consumes less than 0.4 mJ for the maximum considered nodes. Thus, the energy consumption is reduced to approximately 0.6 mJ compared to LEACH. It is also noticed that node density rises between 100 and 500 nodes, consumption of energy increases from 0.1 mJ to 0.3 mJ.

(ii) Throughput (Mbps)

Throughput measurement is computed to estimate the amount of data transferred to the base station via various nodes and is given by Equation (18).

$$Throughput=\frac{Number of Packet sent\ast packet size}{Time taken}$$

(18)

The proposed approach indicates that throughput of about 0.97 Mbps is achieved, which is the highest among the considered protocols even when the number of nodes increases. The MFO obtains a slightly similar range of throughput for all the nodes. Thus, the hybrid approach confirms that the data transmission can be achieved with less energy and high throughput. The obtained throughput for the proposed work and its comparison with other protocols is tabulated in

Table 4. Comparison of throughput with various routing protocols.

Number of Nodes	Proposed Method	MFO	SSO	Fuzzy-PSO	LEACH
100	0.975921	0.962033	0.927321	0.901537	0.879736
200	0.967780	0.947245	0.906462	0.879131	0.859863
300	0.961233	0.931253	0.891598	0.867812	0.839068
400	0.955117	0.927517	0.871758	0.845910	0.826147
500	0.945241	0.916425	0.856928	0.833919	0.819231

for different nodes, and it is figured out in Figure 4.

(iii) End-to-end delay

The end-to-end delay is calculated to estimate the time duration utilized for data transmission from the sensor node to the sink node and it is represented as

$$\mathrm{End}\mathrm{to}\mathrm{end}\mathrm{delay}=\mathrm{Time}\mathrm{required}\mathrm{for}\mathrm{data}\mathrm{transmission}+\mathrm{time}\mathrm{required}\mathrm{for}\mathrm{data}\mathrm{processing}$$

(19)

The proposed work yields a good result with minimum delay when compared with previous similar protocols. When the node size increases, the delay value increases due to the larger transmission distance.

Table 5. End-to-end delay comparison with different routing protocols.

Number of Nodes	Proposed	MFO	SSO	Fuzzy-PSO	LEACH
100	3.091469	4.356343	6.106201	8.563210	8.936313
200	5.738117	6.126811	7.011156	7.723164	9.311644
300	6.363435	6.589253	9.193434	9.381968	10.14887
400	6.812625	7.027501	6.914689	8.016534	9.521367
500	8.615246	9.150356	10.11083	13.08119	13.99697

and Figure 5 provide the outcomes with the details that the delay ranges from 3 s to 8 s for the proposed algorithm, which is the least value for all particular node when compared with other methods. The SSO and Fuzzy-PSO show nearly the range of observation for 200 and 300 nodes. The highest end to-end delay was observed for LEACH.

(iv) Latency

Latency is time required to transmit data in a network between the source and SN. It is also used to estimate round-trip time and the comparison of the obtained latency value is plotted in Figure 6 and tabulated in

Table 6. Comparison of latency with various routing protocols.

Number of Nodes	Proposed	MFO	SSO	Fuzzy-PSO	LEACH
100	3.107003	3.812061	5.333651	7.018231	8.320626
200	2.812463	4.714863	6.519877	8.010662	9.125211
300	3.305389	5.051717	7.012548	8.921786	10.11163
400	4.108043	6.190821	8.018762	9.702966	10.09091
500	5.060925	7.081879	8.312151	10.01457	12.09015

. Figure 6 shows that latency value ranges between 3 to 5 s, which is found to be the lowest when compared with other protocols. The LEACH method shows the highest latency value even for a fewer number of nodes. The Fuzzy-PSO shows some marginal differences. Here, the proposed routing approach achieved latency of 0.712 s, 2.23 s, 3.91 s, and 5.19 s in comparison to MFO, SSO, Fuzzy-PSO, and LEACH, respectively for 100 nodes. The highest latency difference of about 7.03 s is obtained for 500 nodes.

(v) Packet delivery ratio

It is defined as the ratio between numbers of a packet transmitted successfully with a diverse number of SNs to BS and it is equated as

$$Packet delivery ratio=\frac{Total no. of packets received}{Total no. of packets sent}\times 100\%$$

(20)

The ratios of packet delivery for different nodes are represented in Figure 7 and

Table 7. Comparison of packet delivery ratio with various routing protocols.

Number of Nodes	Proposed	MFO	SSO	Fuzzy-PSO	LEACH
100	97.7	96.51	93.35	90.06	88.01
200	97.01	95.01	91.39	88.01	86.02
300	96.51	96.23	89.38	87.06	85.11
400	96.13	93.13	87.56	84.56	82.41
500	96.12	93.31	87.88	84.66	82.45

. The hybrid achieved up to 97.7% and 96.12%of PDR with 100 and 500 nodes, respectively. The highest packet delivery ratio is achieved by the proposed approach.

(vi) Network lifetime

It is used to estimate the amount of time required for transmission of a packet from sender to BS in the WSN. This will provide the approximate computation of SNs that die in the network. It is specified as

$$Network lifetime=Mi{n}_{ij}\left(\frac{{{\displaystyle \sum }}_{i=1}C{M}_{ij}\ast L_i}{N_m}\right)$$

(21)

Here, $C{M}_{ij}$ provides the total matrix coverage, $L_i$ gives the life of the sink node, and $N_m$ provides the total number of nodes.The lifetime of a network of this approach is compared with previous routing techniques and is plotted in Figure 8. It is evident from the graph that the hybrid method attains a good life for all SNs than other compared approaches. More than 2400 rounds with 500 nodes are used for analysis. The increased lifetime of this SN can be achieved due to the multi-criteria clustering approach, which reduces the network over-lapping.

Table 8. Network lifetime comparison with different routing protocols.

Number of Nodes	Proposed	MFO	SSO	Fuzzy-PSO	LEACH
100	1550	1313	1295	1312	1112
200	1708	1401	1429	1388	1376
300	1981	1698	1696	1630	1253
400	2298	1792	1765	1891	1547
500	2391	2173	2001	2121	1920

shows the proposed routing protocol achieves an improvement in a network lifetime of more than 8% in comparison with the LEACH approach of 500 nodes.

4. Conclusions

The transmission of data requires more energy consumption; thus, it is important to have an efficient energy routing algorithm for WSN. Thus, optimal routing methodology, namely the moth flame algorithm and salp swarm algorithm are used in this work. Initially, the SNs which are deployed are organized in a lattice and the multi-criteria clustering is utilized for selecting the appropriate CH. The proposed approach is compared with existing MFO, SSO, Fuzzy PSO, and LEACH techniques, in terms of energy consumption, throughput, end-to-end delay, latency, and packet delivery ratio, and this approach performed well. Subsequently, the combined bio-inspired algorithm would be a better substitute for energy-efficient routing for the WSN. As a result, the hybrid routing method obtained minimal energy consumption with 0.3 mJ, a throughput of 0.97 Mbps, a delay ranges between 3 s to 8 s, a minimal latency value of 3.1 s, and a maximum packet delivery ratio of 97.7%. These results provide the highest energy efficiency of the proposed system. The future scope of this work may be mobile nodes that would be taken into consideration for WSN-based mobile and IoT.