Wavelet Mutation with Aquila Optimization-Based Routing Protocol for Energy-Aware Wireless Communication

Wireless sensor networks (WSNs) have been developed recently to support several applications, including environmental monitoring, traffic control, smart battlefield, home automation, etc. WSNs include numerous sensors that can be dispersed around a specific node to achieve the computing process. In WSNs, routing becomes a very significant task that should be managed prudently. The main purpose of a routing algorithm is to send data between sensor nodes (SNs) and base stations (BS) to accomplish communication. A good routing protocol should be adaptive and scalable to the variations in network topologies. Therefore, a scalable protocol has to execute well when the workload increases or the network grows larger. Many complexities in routing involve security, energy consumption, scalability, connectivity, node deployment, and coverage. This article introduces a wavelet mutation with Aquila optimization-based routing (WMAO-EAR) protocol for wireless communication. The presented WMAO-EAR technique aims to accomplish an energy-aware routing process in WSNs. To do this, the WMAO-EAR technique initially derives the WMAO algorithm for the integration of wavelet mutation with the Aquila optimization (AO) algorithm. A fitness function is derived using distinct constraints, such as delay, energy, distance, and security. By setting a mutation probability P, every individual next to the exploitation and exploration phase process has the probability of mutation using the wavelet mutation process. For demonstrating the enhanced performance of the WMAO-EAR technique, a comprehensive simulation analysis is made. The experimental outcomes establish the betterment of the WMAO-EAR method over other recent approaches.


Introduction
With the rapid advancement in Future Internet technologies, the mobile Internet, the Internet of Things (IoT), the physical world, and the sensor cloud are regularly getting more connected and moving faster toward the always-connected model [1]. In this model, easily configurable, wirelessly connected, and cheap sensors can sense and gather environmental easily configurable, wirelessly connected, and cheap sensors can sense and gather environmental information anywhere and anytime [2]. Unfortunately, the cheapness and higher availability of sensing techniques could only soften the barrier to truly supporting the level of resilience, scalability, and reliability needed for these models to be global. Efficient software methods and a dearth of accurate simulation models specifically intended for wireless sensor networks (WSNs) result in a lack of reliable algorithms, whose accessibility is crucial for setting up future deployment in better conditions [3]. For instance, systematic consideration of the impacts of network dynamicity, channel interference, and node mobility are key points not sufficiently taken into account yet, preventing or weakening the appropriate operation of the resultant network [4]. Therefore, to smash those barriers, we need fast, accurate, and efficient techniques, particularly to handle communication and judiciously manage the tiny computation, memory, and battery resources of the sensor nodes (SNs) in the network [5]. Figure 1 depicts the overview of a routing protocol for WSNs. In WSNs, numerous types of research have involved routing strategy, corporeal design, procedure for sensing capability, power management, and security issues of SNs. The lifetime of SNs is the major problem for WSNs, since SNs have constrained energy resources. The routing algorithm has played a crucial role in the SN's lifetime [6]. Routing in WSNs is different from other wireless networks due to unique properties of SNs, such as processing accomplishment, energy constraints, communication of gathered data from more than one node to a single base station, random deployment of SNs, the improbability In WSNs, numerous types of research have involved routing strategy, corporeal design, procedure for sensing capability, power management, and security issues of SNs. The lifetime of SNs is the major problem for WSNs, since SNs have constrained energy resources. The routing algorithm has played a crucial role in the SN's lifetime [6]. Routing in WSNs is different from other wireless networks due to unique properties of SNs, such as processing accomplishment, energy constraints, communication of gathered data from more than one node to a single base station, random deployment of SNs, the improbability of global addresses, and so on [7]. To accommodate this, distinct kinds of routing approaches have been designed. The final objective is to maximize the overall network lifetime and accomplish energy efficiency. The lifetime of SNs is based on the battery life that gives power to SNs, where the lifespan of the entire network is dependent. A power resource supplies energy to the transceiver, memory unit, and sensing unit [8]. The memory unit is applied for the storage of application-related data and information on device identification, the sensing unit has a sensor for capturing information from the environment, and the transceiver takes the responsibility for the reception and transmission of information. Fast energy dissipation in SNs making them lifeless is a crucial issue in the arena of WSNs [9]. It is noted that ineffective routing protocol causes fast dissipation of batteries. Therefore, it is a fundamental prerequisite to use and design energy-effective routing algorithms in WSNs that would rise their [10].
This article introduces a wavelet mutation with Aquila optimization-based routing (WMAO-EAR) protocol for wireless communication. The presented technique aims to accomplish energy-aware routing in WSNs. To do this, it initially derives the WMAO algorithm for the integration of wavelet mutation with the Aquila optimization (AO) algorithm. A fitness function (FF) is derived using distinct constraints, such as delay, energy, distance, and security. Once the mutation probability P has been set, every individual next to the exploitation and exploration process has the probability of mutation using the wavelet mutation process. For establishing the enhanced performance of the WMAO-EAR technique, a comprehensive simulation analysis is made.

Literature Review
Jagadeesh and Muthulakshmi [11] examined a hybrid metaheuristic (MH) algorithmbased clustering with multihop routing (MHR) (HMA-CMHR) protocol for WSNs. The projected method integrated data transmission, node initialization, clustering, and routing. Initially, the HMA-CMHR algorithm utilized a quantum harmony search algorithm (QHSA)-based clustering procedure for choosing a better subset of CHs. Secondarily, the improved cuckoo search (ICS) technique-based route approach was utilized for a better optimum selection of routes. Gupta and Saha [12] presented a hybrid MH approach where optimum features of ABC and differential evolution (DE) are integrated to measure an optimum group of load-balancing (LB) CHs. Regarding energy efficiency and LB clustering, a new objective function has been developed based on average energy, intracluster distance, and delay parameter. An ABC-based MH technique was presented to the dynamic relocalized mobile sink in a cluster-based network structure.
Al-Otaibi et al. [13] developed a hybridization of the MH cluster-based routing (HM-BCR) approach for WSNs. The HMBCR approach primarily contained a brainstorm optimization with levy distribution (BSO-LD)-based clustering employing an FF integrating four parameters: network load, distance to neighbors, energy, and distance to BS. The WWO-HC-based routing method was implemented for optimum route selection. Subramani et al. [14] concentrated on MH-based clustering with routing protocol for UWSNs-MCR-UWSNs. The purpose of the aforementioned system is to choose an effectual group of CHs and route to target. Lakshmanna et al. [15] presented an improved MH-driven energy-aware cluster-based routing (IMD-EACBR) technique for IoT-supported WSNs. The presented IMD-EACBR approach aimed to achieve maximal energy consumption and network lifespan. To attain this, the IMD-EACBR technique used an improved Archimedes optimized algorithm-based clustering (IAOAC) system for CH selective and cluster organization. The TLBO-based MHR (TLBO-MHR) approach has been applied to optimal selective routes to targets.
Srikanth et al. [16] established an MH optimized enabled unequal clustering with MHR protocol (MOUC-MRP) for WSNs. The MOUC-MRP system's purpose is for selecting CHs and optimum routes to target from WSNs. The harmony search (HS) route-selective manner was developed to an optimum route solution. Mann and Singh [17] examined an improved ABC (iABC) MH with an enhanced searching formula to improve their exploitation abilities and increase the global convergence of the presented MH. An enhanced population-sampling approach was established with Student's-t distribution that needs only one control parameter for computing and storing, then enhances the efficacy of presented MH.

The Proposed Model
In this study, a new WMAO-EAR algorithm is proposed for effectual and energyaware wireless communication. The presented WMAO-EAR technique aims to accomplish energy-aware routing in WSNs.

System Model
Approaches utilized to carry out the presented routing protocol were the network model, energy model, and energy-dissipation model [18]. In the following section, a detailed description of all the models is given.

Energy Model
WSNs involve link heterogeneity, energy heterogeneity, and computational heterogeneity. Energy heterogeneity is regarded as the most important to ensure optimal network performance. In the study, we discussed three levels of energy heterogeneity: advanced, normal, and intermediate nodes. Intermediate node initial energy is between advanced and normal node initial energy, given that m and b represent the percentage of advanced nodes and intermediate nodes correspondingly. Advanced node energy is α times more than that of the intermediate and normal node energy is β times more than normal node energy. Now, the β is connected with α via β = α/2. With E N , E I , and E A signifying the initial energy of normal, intermediate, and advanced nodes, then energy relationships are formulated, using:

Energy-Dissipation Model
Reception and data transmission are two fundamental functions in WSNs. Usually, the data-communication method expends lots of energy when compared to data reception. Now, energy cost during k-bit data transmission over distance d is formulated as follows.
In Equation (4), d 0 indicates the reference distance and d 0 ε f s /ε mp . Elec shows the per bit energy cost for running the receiver or transmitter circuits, and ε f s and ε mp indicate amplification parameters of the transmitter for free space and multipath fading, correspondingly.
Furthermore, energy costs during k-bit data reception are formulated as follows:

Network Model
Consider that the n sensor is positioned in the region of M size and the sensor is static. Every sensor is regarded as being aware of the identification and location of other SNs. Also, consider that the advanced node's location is predetermined, while the normal and intermediate nodes are placed randomly. All the nodes transmit information to neighboring CH for data aggregation. The distance between the BS and nodes is d s ≥ d r s , where d indicates a predetermined range (d r s = 10 m).

Design of WMAO Algorithm
Abualigah et al. [19] introduced the fundamental formula of the Aquila optimizer (AO). Usually, the AO algorithm mimics the natural behavior of Aquila for catching the prey. Like other metaheuristic (MH) models, AO was a population-based optimization algorithm that takes place by creating an early population X that has N agents. In Equation (6), the mathematical modeling of the algorithm is given.
UB j and LB j indicate parameters of the searching space, r 1 ∈ [0, 1] denotes a random number, and Dim shows the dimension of the agent. Next, exploration and exploitation are performed until the optimal solution is found. To perform exploration and exploitation, two strategies are employed. The initial strategy performs the exploration based on the average of agents (X M ) and the finest agent X b , as follows: where 1 − t T controls the search in the exploration stage, T indicates the maximal amount of iterations, and rand shows a random number within [0, 1]. Figure 2 showcases the flowchart of the AO method. neighboring CH for data aggregation. The distance between the BS and nodes is ≥ , where d indicates a predetermined range ( = 10 ).

Design of WMAO Algorithm
Abualigah et al. [19] introduced the fundamental formula of the Aquila optimizer (AO). Usually, the AO algorithm mimics the natural behavior of Aquila for catching the prey. Like other metaheuristic (MH) models, AO was a population-based optimization algorithm that takes place by creating an early population that has agents. In Equation (6), the mathematical modeling of the algorithm is given.
and indicate parameters of the searching space, ∈ [0, 1] denotes a random number, and shows the dimension of the agent. Next, exploration and exploitation are performed until the optimal solution is found. To perform exploration and exploitation, two strategies are employed. The initial strategy performs the exploration based on the average of agents ( ) and the finest agent , as follows: where (1 − ) controls the search in the exploration stage, indicates the maximal amount of iterations, and shows a random number within [0, 1]. Figure 2 showcases the flowchart of the AO method. The next strategy used to perform the Levy flight ( ( )) distribution and for updating the exploration capability of the solution is: The next strategy used to perform the Levy flight (Leνy(D)) distribution and X b for updating the exploration capability of the solution is: where s = 0.01 and β = 1.5, u and v denote random numbers, and Γ indicates a constant number. In Equation (9), X R indicates an arbitrarily selected agent. Furthermore, y and x imitate the spiral shape, as given below: where ω = 0.005, U = 0.00565, r 2 ∈ [0, 20] represents a random number, and D l indicates integer numbers from 1 to the length of the search domain. An initial strategy is used to upgrade agents inside the exploitation stage based on X b and X M , like exploration as follows: where α and δ signify the exploitation adjustment parameter and rand ∈ [0, 1] denotes a random number. In the next strategy of exploitation, the agent gets upgraded based on X b , Levy, and the quality function QF: where rand () denotes a function that produces a random number and G 1 specifies distinct motions that are used to track the best individual solution, as follows: where rand shows a random number. Given that, G 2 specifies reducing value from 2 to 0, and it is evaluated by Equation (17): Algorithm 1 illustrates the fundamental steps of the AO.
Upgrade the X i based on Equation (7).
Upgrade the X j based on Equation (13).
In the WMAO algorithm, the mutation was considered a significant technique to support the technique's jump from local optimum [20]. In this phase, the wavelet mutating approach for enhancing the execution of the AO method is presented. Once the mutation probability P is set, every individual after the exploitation and exploration phase of the method will be getting a mutation chance via wavelet mutation strategy. Where rand < P, an individual executes Morlet wavelet mutation. The formula of mutation is: where X i (t)(i = 1, 2, · · · , N) denotes the i-th individual location in t-th generation, and LB and UB are the lower and upper bounds of the present search space. Correspondingly, σ represents the wavelet mutating coefficient. Its formula is [21]: where ψ(ϕ/α) = e −(ϕ/α) 2 /2 · cos (5ϕ/α) is the Morlet wavelet function, and 99% of its energy can be concentrated within −2.5 and 2.5, so ϕ signifies an arbitrary number within −2.5α and 2.5α. The a is the scaling parameter and its expression is [22]: This procedure assures that an individual with superior fitness will be entering the next iteration, thereby enhancing the convergence speed and optimization capability of the algorithm.

Process Involved in the WMAO-EAR Technique
In the presented WMAO-EAR technique, an FF is derived using distinct constraints, such as delay, energy, distance, and security. The main concept of the proposed algorithm is to decrease the communication distance between the selected nodes and CH [23]. It also focuses on decreasing the delay to transmit the information from one node to another. In contrast, network energy has to be higher, i.e., during data transmission, it needs to employ a small amount of energy. Finally, the node must tolerate the risk attained in the network. The objective function of the adapted CH demonstrated in Equation (22), while the value of η needs to depend on 0 < η < 1. Now, v m and v n indicate the operation, as demonstrated in Equations (23) and (24). The constraints on delay, energy, distance, and security are indicated as σ 1 , σ 2 , σ 3, and σ 4 . The condition of this constraint is indicated as σ 1 + σ 2 + σ 3 + σ 4 = 1. In Equation (24), Y Z − S s denotes the distance between the normal and sink nodes.
Equation (25) illustrates the FF for distance, where di J (m) is related to packet transmission from the normal nodes to CH and CH to BS. Generally, v i dis lies amongst [0, 1] and the value goes high when the distance between the normal node and CH is higher. Equations (26) and (27) demonstrate v dis (m) and v dis (n), where Y z indicates the normal node in zth cluster, G z represents the CH of Zth cluster, the distance between the BS and CH is stated as G Z − S s , G Z − Y y symbolizes the distance between the normal node and CH, Y z − Y y indicates the distance between two normal nodes, and M z and M y ∨ represent the node amount that eliminates the Zth and yth clusters.
Equation (28) illustrates the FF of energy. The value v i ene is greater than 1 and the whole CH cumulative v ene (m) and v ene (n) concern the maximum energy value and the greater amount of CH.

Conclusions
In this study, a new WMAO-EAR approach was devised for effective and energyaware wireless communication. The presented WMAO-EAR technique aims to accomplish an energy-aware routing process in WSNs. To do this, the WMAO-EAR technique initially derives the WMAO algorithm for the integration of wavelet mutation with the AO algorithm. An FF is derived using distinct constraints, such as delay, energy, distance, and security. Once the mutation probability P is set, every individual next to the exploitation and exploration stage process has a probability of mutation using the wavelet mutation process. To demonstrate the enhanced performance of the WMAO-EAR technique, a comprehensive simulation analysis has been presented. The experimental results establish the superiority of the WMAO-EAR method over other recent approaches. In the future, lightweight cryptographic solutions can be applied to boost secure communication.

Conclusions
In this study, a new WMAO-EAR approach was devised for effective and energyaware wireless communication. The presented WMAO-EAR technique aims to accomplish an energy-aware routing process in WSNs. To do this, the WMAO-EAR technique initially derives the WMAO algorithm for the integration of wavelet mutation with the AO algorithm. An FF is derived using distinct constraints, such as delay, energy, distance, and security. Once the mutation probability P is set, every individual next to the exploitation and exploration stage process has a probability of mutation using the wavelet mutation process. To demonstrate the enhanced performance of the WMAO-EAR technique, a comprehensive simulation analysis has been presented. The experimental results establish the superiority of the WMAO-EAR method over other recent approaches. In the future, lightweight cryptographic solutions can be applied to boost secure communication.