Intelligent Trust-Based Utility and Reusability Model: Enhanced Security Using Unmanned Aerial Vehicles on Sensor Nodes

Abstract

Due to its importance in prolonging the lifetime of battery-restricted wireless sensor networks, network longevity has garnered considerable research attention, with the rechargeable wireless sensor network emerging as a viable solution. In this research, the novel methodology of a trust-based mechanism for enhanced security integrated with an energy utility and re-usability model is proposed with software-defined networking (SDN) to maximize energy utilization. We proposed a novel framework with SDN for the service station in a wireless sensor network (WSN). The results showed that the life capacity of the network increases to a maximum of 290% when compared with no charging, with the charge increasing by 30% intervals. We also present how the network survives through this choice of sink. As there is variation in the network size while it increases, the proposed approach with the static method works well until the network size reaches 200. Furthermore, the proposed approach also uses the heuristic method to achieve the best performance.

1. Introduction

Wireless Sensor Networks continue to attract the attention of both academic and industrial researchers all around the world due to its distributed nature and are exploited in numerous applications [1] from environmental monitoring to military surveillance including human computer interaction. Privacy, security, and consumption of energy by the nodes happen to be a foremost anxiety in the growth of WSNs, as they are exposed to malicious attacks. If a single sensor node becomes compromised, it can target multiple nodes to become engaged in and to obstruct the services rendered for legitimate users over the network. On the contrary, the wireless sensor network security needs to be addressed significantly; specifically, there is a need for mission-critical tasks [2]. It is a critical task to have secure WSNs, and it is also critical in the plans of action and applications in military battlefields. For instance, if the possibility of security vulnerabilities is present, it is necessary for neighboring troops (nodes) to perform very well or even exceedingly well in the battlefield [3,4,5,6].

Another important issue that WSNs need to consider is the network lifetime as, on many occasions, outdoor sensor networks mostly rely on batteries for energy. Depending on the size of the deployments, proportionally larger batteries are required for larger deployments. If the network is located where regular recharging is not useful, crop growth under a bridge or in a field covered by sun and or affected by wind can significantly limit the battery life, relying solely on the effective network. We developed a system that allows for such situations in an energy feeding system using UAV by utilizing the energy of compromised nodes and external energy in the legitimate nodes to ensure 100% non-failure of the network system in the field, as shown in Figure 1.

In our technique, we define the lifespan of a network and how it can be extended with UAV-based wireless charging. The extended life authenticates the addition of a recharge circuit to the sensor node. If there is a reasonable increase, we investigate under which conditions it works and focus on the amount of nodes that the UAV needs to reload. It also describes which nodes need to act as sinks for network data and how you can change the sinks to maximize network life in response to recharging.

The benefits of UAV charging in terms of WSN and the optimal use of UAV charging through compromised nodes have not been investigated. However, a significant examination into the community of the sensor network has considered highly efficient modules based on the energy- and cost-effective hardware communication [3], improved protocols with the software communication [4,5], and various approaches with data routing [6,8,9,10]. A large subset is considering a sufficient mobilization strategy [11,12,13,14,15,16]. Using this early research, we developed a method for sensor arrays to maximizes the benefits of UAV charging.

In our research work, we proposed a new approach that reveals a technique to extend the life of a network using UAV-based wireless charging in three different cases.

• UAV recharges nodes from the entire network using the energy utilization and re-usability method from the detected compromised node by using the intelligent trust evaluation system and external energy.
• UAV recharges only the lowest-powered nodes in the network and utilizes the static method of choosing a sink, with the next sink consequently only utilizing the power from the energy utility model.
• UAV renews the sink node by recharging energy from the network with a different sink selection algorithm and provides the best sink node selection algorithm and performance measures of each algorithm with network scalability.

We performed some simulations introduced from a preliminary study [17] to investigate these issues that have shown that UAV charging can extend the network life by approximately three times. Then, we look at other issues and set up a list of guidelines dependent on the size of the network. According to the guidelines, we select the nodes (sink or low power nodes), the amount of UAVs required to enter that node, and what node to choose next. Interestingly, as the size of the network grows, it becomes more effective for the sink to remain intact, with up to 30% charged at every interval. However, when the network extends the number of nodes to 200, a change of sink is required.

This paper is further structured as follows: In Section 2, we review existing techniques used in detection systems with sensor nodes, in energy optimization models, in wireless charging models, and in energy maximization. Section 3 gives our complete research statement. Section 4 describes a range of reproducible sensor network problems in energy transfer through UAVs. Section 5 describes the energy utility and re-usability model, and Section 6 describes the UAV energy consumption and recycling model used to optimize energy distribution and shows the results of our analysis. Section 7 gives our conclusion and insights into future work.

2. Literature Review

WSNs exhibit low power interference, bandwidth transmission, memory size, and data storage. Following these WSN regulations (strong IT resources, power supplies, and temporary communication environments), a number of security measures (including infiltration detection methods) have been developed. However, these security measures do not work directly on the WSN environment for older wired/wireless networks. Despite all of the efforts made to find a way to infiltrate WSNs [9,10,11,12,13], methods of energy conservation (or energy recovery [14,15,16,18] are intended to generate energy in the environment, and their effectiveness for sensory networks is still measured in practice, although this may be due to nodes being malicious, which has however not been thoroughly researched, and there is no doubt that wireless networks are unreliable due to device checks and off-network connections. Security outbreaks against WSNs are divided into active and passive attacks. In unmanned attacks, the square-sized attackers are often hidden from view and can be a communications network that collects data or destroys active network components. Serious attacks can be categorized using visual inspection, node anomalies, and damage and traffic testing. In intense attacks, the opponent actually marks the performance of the network attacked. This result can also lead to a targeted attack. For example, network services can be compromised or interrupted with effect of the attacks. Acute attacks are categorized as DOS, jamming, hole attacks (black hole, worm, sinkhole, etc.), floods, and Sybil attacks. Readers interested in vulnerable security attacks against WSNs can visit [19,20,21] for more information.

The grounds of our research claim rely on a remarkable discovery of innovation on wireless power transmission by Kurs et al. [22], which was revealed in 2007 and has since gathered attention globally. In [22], Kurs et al. utilized constant frequency resonance for the transmission of energy efficiently, wasting comparatively less energy in extraneous objects.

This article further presents a novel approach for detecting compromised nodes and permits the sensor network to stay operational forever by victimizing and optimizing the energy utilized from the detected compromised node in the legitimate sensor nodes within the network through unmanned aerial vehicles.

3. Research Statement

We contemplate the set of WSN nodes denoted by N and forwarding nodes, which were distributed over a 2D area, as shown in Figure 2. Each sensor node sends sensing information periodically to monitor the events in military surveillance to the forwarding node, considered to be a trusted sensor node, which forwards the aggregation result $A_{result}$ calculated using the double weighted trust evaluation [23] to the base station. The sensor node’s battery capacity is denoted by $E_{max}$ initially with fully charged battery up to the maximum capacity. The minimum energy of a node is denoted as $E_{min}$ for WSN nodes (for it to be capably operational). The weight trust evaluation system updates the weight of the sensor node periodically after the aggregation result to detect whether nodes are faulty or malicious in the WSN to make correct conclusions in monitoring applications based on prior decisions.

Affected or compromised sensor nodes transmit information to the base station. Accurate and timely identification of malicious nodes is essential to provide reliable operation of networks. Network life limitations are widely considered to be the main performance barrier to providing single network security and permanently changing the network in a wireless sensor network.

Every sensor node produces sensitive information (alarm) to monitor specific military use targets. In the divergence of a malicious node, when a node with normal characteristics starts sending an alarm, the neighboring node also starts sending an alarm after a short interval. The FN collects the sensor measurements of the corresponding sensor nodes, where “1” indicates the alarm. It calculates the weighted values as $A_{result}$ = 0 or 1. At the end of the FN integration, all loads assigned to the member nodes are renewed [22,24,25].

Then, nodes with $W_i$ that are more or less equal to a certain $W_{min}$ converging value are identified as malicious. The $W_{min}$ value is expected to be 0 if the load from 0 to 1 is considered a malicious terminal and is not disabled from full processing.

Every sensor node uses power to send and receive data. $P_i$ is determined as the rate of consumption of energy by sensor node $i\in N$. In this research proposal, we utilize the certain energy dissipation [26,27]. Multi-hop data transfer is used to transfer data through nodes. The symbol $f_{ij}$ implies that the transmission flow from the sensor network node is represented between i to j, whereas $f_{iB}$ is the rate of flow transmission from network node i to the B, which represents the base station. The balancing constraint at every node i for the flow rate is shown by Equation (1).

$$P_i=\rho .\sum _{k\in N}^{k\ne i}{f}_{ki}+\sum _{j\in N}^{j\ne i}{f}_{ji}+C_{iB}{f}_{iB}\forall i\in N$$

(1)

where $C_{ij}/C_{iB}={\beta }_1+{\beta }_2{D}_{ij}^{\alpha }$.

We proposed a novel approach to recharging the battery at every legitimate node: an unmanned aerial vehicle (UAV) is engaged in the network, and it is assumed that UAV is completely charged to complete a full rotation until it reaches the charging station for the next cycle. As discussed earlier, the double-weighted trust-based evaluation of nodes belonging to the network yields the value of $W_i$ depending on the data sensed and on the aggregation result from the base station. The trusted weight for the individual node is calculated in the base station. Further aggregation result weighting is given accordingly. Faulty or malicious nodes are identified through the value of $W_i$. Here, we use the technology of wireless energy transfer [22] through which we can charge the sensor nodes via unmanned aerial vehicles [22,24,25]. The UAV begins from the service station (S), with its travel speed being V (in $m/s$). When it travels and stops at a legitimate node $(W_i⩽W_{min})$, ${\tau }_i$ is the interval required to charge the legitimate WSN node’s battery wirelessly via the wireless energy transmission method. In the case of a malicious node $(W_i⩽W_{min})$, ${\tau }_{ivac}$ time is taken for the UAV to be charged from the malicious node despite the node being isolated with the existing energy at the time when $W_i$ reaches $W_{min})$ for the malicious sensor node. At each renewable cycle, the energy of the UAV is maximized to be spent on the legitimate sensor nodes from the energy obtained from the detected malicious nodes [25,28].

After the UAV visits all nodes placed in the network (see Figure 3), it returns to its original starting point (the service station) to be serviced and prepared for successive rounds. We denote this as a break period or the vacation period, ${\tau }_{svac}$. Thanks to our proposed method, the energy acquisition ratio is maximized along the path of the cycle $\tau$.

4. Energy Transfer through UAV

Here, we present the renewable cycle framework. Let the process start with the service station; the travel of the UAV begin initially with the service station; and the travel take place over a path to reach all of the sensor nodes once in a cycle, coming back to the original position, the service station [25,28].

Path $P=({\pi }_0,{\pi }_1,{\pi }_2,\cdots ,{\pi }_n)$ is the path traversed by a UAV over one cycle of the trip, starting initially with the service station and traveling over the path to reach all sensor nodes in a cycle, coming back to the original position, the service station $\left({\pi }_0\right)$ [25,28].

$$a{\pi }_i=\tau +\sum _{k=0}^{i=1}{\displaystyle \frac{D{\pi }_k{\pi }_{k+1}}{V}}+\sum _{k=1}^{i-1}\tau {\pi }_k$$

(2)

The $i^{th}$ node reached by the UAV through path P is ${\pi }_i$, $1⩽i⩽\left|N\right|$. We specify $D_{{\pi }_0{\pi }_1}$ as service station as well as the initial node visited, with distance as P and $D{\pi }_k{\pi }_{k+1}$ representing the space between the $k^{th}$ and ${(k+1)}^{th}$ nodes. We set $a_i$ as the UAV reaching time at node i in the initial roundtrip, as shown in Figure 4. We denote $DP$ as the distance to path P and ${\tau }_P={\displaystyle \frac{DP}{V}}$ as the total time spent for the distance path $DP$. Remember that ${\tau }_{svac}$ is the vacation time spent by UAV at the service station. Additionally, the rotation period $\tau$ can be represented [24,25]

$$\tau ={\tau }_P+{\tau }_{vac}+\sum _{i\epsilon N}^{}{\tau }_i$$

(3)

where $\sum _{i\epsilon N}{\tau }_i$ is the time that the UAV devotes to all the sensor nodes for energy transfer from wireless energy transfer. The power intensity of node $i\epsilon N$ demonstrates a renewable energy cycle when it satisfies the criteria below.

• It begins and finish with the same level of energy with a time $\tau$ and
• It does not go below $E_{min}$

$$\tau .P_i=i\epsilon N$$

(4)

Consider the fact that the UAV reaches a node i at time period $a_i$ through a renewable energy cycle; it does not need to recharge the sensor nodes’ battery capacity to $E_{max}$.

Gain in Energy at Sensor Node in the First Cycle

$G_i$ symbolizes the initial energy state of sensor node i in a renewable cycle and $g_i\left(t\right)$ denotes the level of energy at time t (denoted with the sawtooth graph). During a roundtrip cycle $[\tau ,2\tau ]$, we note the level of energy with only two slopes: [28] (i) a slope of $p_i$ when the UAV is not at this node (i.e., non-charging period) and (ii) a slope of ($U{p}_i$) when the UAV is charged at a rate of U (i.e., charging period).

$$E_i=E_{max}+P_i(a_i+{\tau }_i)-U{\tau }_i$$

(5)

5. Energy Utility and Re-Usability Model through UAV

We propose the novel energy utility and re-usability model by considering the scenario in which the available energy from a detected malicious node is transferred to legitimate nodes of the network to increase the lifetime of the network nodes. Thus, it reduces the DOS that occurs due to the low energy nodes [25].

Figure 5 shows the unmanned aerial vehicle energy through which the sensor nodes charged wirelessly during the energy renewable cycle.

Considering the detected compromised nodes, the $P_{UAV}$ energy was maximized, through which the average network life time has also been maximized [25].

Malicious node energy=$\sum _{i=1}^N{P}_i\left({\tau }_i\right)(W_i<W_{min})$ Remaining Energy (RP) at time $a_i$ = Total Energy-Total Energy Consumed.

Total Energy Consumed (TPC) = $\sum _{i=1}^{N}U{\tau }_i$.

Remaining Power: After the duty cycle time, the detected compromised nodes are isolated and Tot nodes N = N-Tot node $W_i<W_{min}$

Figure 5 shows the evidence of the next cycle of renewable energy (represented with graph-Sawtooth), where $E_{max}$ is the capacity of a battery energy charge during a UAV visit. The lifetime of the detected compromised nodes is allocated to the legitimate sensor node to optimize the energy failure among nodes in the sensor networks with practical implementation, as shown in Figure 5.

Remaining Energy

Remaining Energy =$P_{UAV}+\sum _{i=1}^N{P}_i(W_i<W_{min})+\sum _{i=1}^{N}U{\tau }_i$ Here, we consider only the average energy gain obtained from the detected compromised nodes at time ${\tau }_{ivac}$. This energy gain of UAV in addition to the charged Energy $P_{UAV}$ at ${\tau }_{svac}$ has been analyzed in the Section 6.

$$AverageEnergyGain={\displaystyle \frac{\sum _{i=1}^N{P}_i(W_i<W_{min})}{N}}$$

(6)

Equation (6) shows that the average energy gain from the detected compromised nodes is the maximum energy obtained through the energy utility and re-usability model Figure 6.

6. Optimization of Energy Distribution

Theorem 1.

The distribution of energy attained using an unmanned aerial vehicle (UAV) from the compromised node detected using weighted trust evaluation raises an alarm. While the UAV travels on the path $P=({\pi }_0,{\pi }_1,{\pi }_2,.....,{\pi }_n,{\pi }_0)$, an alarm is raised from the node on the path ${\pi }_i$ at the arrival time $a_i$ and the time spent on the node i is ${\tau }_i$, taking energy $p_i\left({\tau }_i\right)$ of the compromised node.

Proof. 

Case 1: $(W_i<W_{min})$ $P_{UAV}\left(a_i\right)<P_{UAV}\left(a_i\right)+p_i\left({\tau }_i\right)$ (i.e)$P_{UAV}\left(a_i\right)<P_{UAV}(a_i+{\tau }_i)$ Here, $p_i\left({\tau }_i\right)$ is the energy transferred from node i to the UAV through wireless energy transfer in addition to the total charged energy of $P_{UAV}$ through the service station during vacation time, denoted as ${\tau }_{svac}$. Case 2: $(W_i>W_{min})$ $P_{UAV}\left(a_i\right)>P_{UAV}(a_i+{\tau }_i)$ $P_{UAV}\left(g_i\left(t\right)\right)=P_{UAV}-U{\tau }_{i}or{P}_{UAV}\left(e_i\left(t\right)\right)=P_{UAV}-U{\tau }_i$ Here, $U{\tau }_i$ is the energy charged from the UAV to the legitimate nodes during a cycle of the trip. In our proposed system, there is no loss of energy and the energy from a detected compromised node utilize. The energy is recycled to the legitimate node in the network, through which security has been enhanced, represented in Figure 7. □

6.1. Optimization Proof

We prove the optimization by presenting the feasibility of reformulating the error function with respect to the normal objective and by tuning it with the controlled error gap, as shown in Equation (7).

$$\begin{array}{c}{\eta }_{vac}^{\prime }-{\eta }_{vac}\le \widehat{\eta }vac-\eta vac\hfill \\ =\left[1-\sum _{s\in N}\widehat{\eta }s-\frac{u·\tau tsp}{E_{max}-E_{min}}·\widehat{\eta }i·\left(1-\widehat{\eta }i\right)\right]-\hfill \\ \left[1-\sum _{s\in N}\widehat{\eta }s-\frac{u·\tau tsp}{E_{max}-E_{min}}·{\eta }_{max}·\left(1-{\eta }_{max}\right)\right]\hfill \\ =\frac{u·{\tau }_{tsp}}{E_{max}-E_{min}}\left({\gamma }_{max}-{\eta }_{max}^2\right)\hfill \\ \le \frac{u·{\tau }_{tsp}}{4\left(E_{max}-E_{min}\right)}·\frac{1}{m^2}\hfill \end{array}$$

(7)

6.2. Gain in Energy $P_{UAV}$

Thus, Figure 5 shows that the sum of energy gain at some points denotes $p_i\left({\tau }_i\right)$ from the trips $(\tau ,2\tau ...)$ of an energy renewable cycle.

$$P_{UAV}\left(e_i\left(t\right)\right)=P_{UAV}+\sum _{i=1}^N{p}_i\left({\tau }_i\right)-\sum _{i=1}^{N}U{\tau }_i$$

(8)

$$P_{UAV}\left(g_i\left(t\right)\right)=P_{UAV}+\sum _{i=1}^N{p}_i\left({\tau }_i\right)-\sum _{i=1}^{N}U{\tau }_i$$

(9)

The first renewable cycle $\tau \sum _{i=1}^N{p}_i\left({\tau }_i\right)=0$ in $g_i\left(t\right)$ becomes

$$P_{UAV}\left(g_i\left(t\right)\right)=P_{UAV}-\sum _{i=1}^{N}U{\tau }_i$$

(10)

so $P_{UAV}\left(e_i\left(t\right)\right)\ge P_{UAV}\left(g_i\left(t\right)\right].$ In the another renewable cycle, $2\tau \sum _{i=1}^N{p}_i\left({\tau }_i\right)>0$

$$P_{UAV}\left(e_i\left(t\right)\right)=P_{UAV}+\sum _{i=1}^N{p}_i\left(\tau i\right)-\sum _{i=1}^{N}U{\tau }_i\ge P_{UAV}\left(g_i\left(t\right)\right)$$

(11)

If $\sum _{i=1}^{N}U{\tau }_i>\sum _{i=1}^N{p}_i\left({\tau }_i\right)$, $P_{UAV}\left(e_i\left(t\right)\right)<P_{UAV}\left(e_i\left(a_i\right)\right)$. If $\sum _{i=1}^N{p}_i\left({\tau }_i\right)>\sum _{i=1}^{N}U{\tau }_i$, check whether $N_{com}<{\displaystyle \frac{N}{2}}$ mostly satisfies $P_{UAV}\left(e_i\left(t\right)\right)>P_{UAV}\left(e_i\left(a_i\right)\right)$. Furthermore, $\sum _{i=1}^N{p}_i\left({\tau }_i\right)$ is considered ${\tau }_{ivac}.$ Node and UAV energy: The solution ${\phi }^{*}.Full$ achieves same maximum ratio of vacation time to the cycle time similar to the solution ${\phi }^{*}$ [25]. Under $\phi$, it is proven that $e_i(a_i+{\tau }_i)=E_{max}$. Thus, it is a feasible cycle for node energy renewal. The assumption ${\phi }^{*}$ is the optimal solution for further increasing the objective value, as shown in Figure 5.

Lemma 1.

At this condition, ${\phi }^{\beta }$ satisfies $P_{UAV}(e_i\left(t\right)$ at UAV and $e_i\left(t\right)$ at the nodes, both greater than $P_{UAV}\left(e_i\left(a_i\right)\right)$ and $e_i\left(a_i\right)$, respectively, at the arrival time in our proposed solution. It satisfies the condition that both energy renewable sources and energy receiving sources co-ordinate with each other to utilize the waste energy of compromised nodes, recycled to increase the overall network lifetime.

Total energy gain at ${\tau }_{ivac}$: $N_{com}$ * $p_i\left({\tau }_i\right)$ when $(W_i<W_{min})$. This is the energy transferred to $P_{UAV}$ in addition to the energy charged in the energy station at ${\tau }_{vac}$ time at cycles $(\tau ,2\tau ...)$ until $N_{com}>{\displaystyle \frac{N}{2}}$, where the total number of nodes detected as compromised node in the network is denoted as $N_{com}.$ $P_{UAV}$ is maximum at some points at each cycle $(\tau ,2\tau ...)$ when $W_i<W_{min}$ nodes transfer energy at ${\tau }_i$ is $p_i\left({\tau }_i\right)$ to the sensor nodes by isolating the malicious nodes through the calculated weights denoted with $W_i$.

6.3. Optimal Travel Path

The UAV must move along the shortest Hamiltonian cycle, and the shortest travel path can be found by solving the problem called the Traveling Salesman Problem (TSP) [28]. The symbol $D_{TSP}$ represents the distance traveled through the shortest Hamiltonian cycle, and let ${\tau }_{TSP}=dfrac{D}_{TSP}V$. The optimal travel path is

$${\tau }_{TSP}+{\tau }_{svac}+{\tau }_{ivac}+\sum _i^N{\tau }_i=\tau$$

(12)

$${\tau }_{vac}={\tau }_{svac}+{\tau }_{ivac}$$

(13)

By our proposed system, the min ${\displaystyle \frac{{\tau }_{vac}}{\tau }}$ was obtained, ${\tau }_{svac}$ is the time spent on the service station to be charged by a UAV, ${\tau }_{ivac}$ is the charging period to maximize the energy from a UAV at some points along the physical path, and ${\tau }_{vac}$ is period of time as mentioned in Section 6 case(1) such that max ${\displaystyle \frac{{\tau }_{vac}}{\tau }}$ has been satisfied in OPT by the near optimal solution procedure in our proposed solution.

Here, in an optimized research solution with the maximum ${\displaystyle \frac{{\tau }_{vac}}{\tau }}$, the UAV must have traveled through the shortest Hamiltonian cycle that links all of the sensor nodes and the initial point (service station).

OPT max ${\displaystyle \frac{{\tau }_{vac}}{\tau }}$ s.t

$$\sum _{j\epsilon N}^{j\ne i}{f}_{ij}+f_{iB}\left(i\epsilon N\right)-\sum _{k\epsilon N}^{k\ne i}{f}_{ki}=R_i\left(i\epsilon N\right)$$

(14)

$$\rho .\sum _{k\epsilon N}^{k\ne i}{f}_{ki}+\sum _{j\epsilon N}^{j\ne i}{C}_{ij}{f}_{ij}+C_{iB}{f}_{iB}-P_i=0\left(i\epsilon N\right)$$

(15)

$$\tau -\sum _{i\epsilon N}{\tau }_i-{\tau }_{vac}={\tau }_{TSP}$$

(16)

$$\tau p_i-U{\tau }_i=0\left(i\epsilon N\right)$$

(17)

$$\tau -{\tau }_i{p}_i⩽E_{max}-E_{min},i\epsilon N$$

(18)

$f_{ij},f_{iB},{\tau }_i,\tau ,{\tau }_{vac},p_i\ge 0(i,j\epsilon N,i\ne j)$.

In this research problem, the rate of flow $f_{ij}$ and $f_{iB}$; time periods $\tau$, ${\tau }_i$, and ${\tau }_{vac}$; and energy utilization $p_i$ are the variables used to denote the optimization, and $R_i,\rho ,C_{ij},C_{iB},U,E_{max}$, $E_{min},and {\tau }_{TSP}$ are constants. This research issue has both a non-linear term ${\displaystyle \frac{{\tau }_{vac}}{\tau }}$ and non-linear terms $(p_i$ and ${\tau }_i{p}_i)$. Of note, there are two possible outcomes for the optimization problem OPT: either the solution is optimal or infeasible, denoted by OPTS. There are various scenarios where the outcomes may occur later, e.g., (i) the UAV energy charged rate is varied, and it may be very low or small or may be large; (ii) the interval between UAVs is high. Due to the problem constraints, OPT could not be held. These are some of restrictions for a UAV to regain some network energy with a maximized lifetime in a sensor network [28].

7. Results and Discussion

In the results, the proposed framework is shown to regain energy for a WSN by maximizing the energy of UAVs as well as the sensor node, with strong properties in such a network.

7.1. Simulation Settings

We consider the WSNs to consist of 50 nodes. The sensor nodes were deployed over a square area of 1 km × 1 km. The data rate (i.e., $R_i$, i belongs to N) from each node is random.

The data flow rate $R_i$ and location for each node in a 50-node network is generated within [1, 10] kb/s. The energy consumption co-efficient ${\beta }_1={\displaystyle \frac{50nJ}{b}}$, ${\beta }_2$=${\displaystyle \frac{0.0013pJ}{b.m4}}$, $\alpha =4$, and $\rho ={\displaystyle \frac{50nJ}{b}}$. The travel speed of the UAV is V = 5 m/s, and $P_{UAV}$ =100 KJ at the start of the vacation time.

7.2. Results

In this research problem, the rates of flow $f_{ij}$ and $f_{iB}$; time periods $\tau$, ${\tau }_i$, and ${\tau }_{vac}$; and energy utilization $p_i$ are the variables used to denote the optimization, and $R_i,\rho ,C_{ij},C_{iB},U,E_{max},$$E_{min}$, and ${\tau }_{TSP}$ are constants. This research issue has both a non-linear term ${\displaystyle \frac{{\tau }_{vac}}{\tau }}$ and non-linear terms $(p_i$ and ${\tau }_i{p}_i)$. Of note, there are two possible results for the optimization issue OPT: either the result is optimal or infeasible, denoted by OPTS. There are various scenarios where the outcomes may occur later, e.g., (i) the UAV energy charge rate is varied, and it may be very low or small or may be large; (ii) the interval between UAVs is high. Due to the problem constraints, OPT could not be held. These are some of the restrictions for a UAV to regain some energy in a WSN network [22,25,29,30,31,32,33,34,35,36,37,38,39,40,41].

The good results estimated by the energy utility model proposed with UAV are shown in Figure 8, Figure 9, Figure 10 and Figure 11.

8. Conclusions

We found that there are energy blocking nodes in the network, with their energy dipping to a minimum. While considering a maximized $P_{UAVs}$ through which security increased, we also optimized this type of node in the network by looking for the ratio of energy expended by each node in the appropriate travel cycle. Using the concept of SDN in WSNs, we easily tackled complex issues, such as energy utilization and maximization with network management and obtained a simplified solution. We also proposed an exclusive framework for software-defined WSNs; using that framework, we prevented this type of energy from falling to the network nodes by supplying energy with a suitable UAV. Scalability must also be achieved as the number of sensor nodes in the network escalates. Here, Kurs et al. [22] stated that transferring energy wirelessly can be achieved from a single resource node to several energy receiving nodes simultaneously. This recommends that a drone populates numerous nodes simultaneously in its travel path and therefore has the latent ability to operate within a densely deployed network of sensor nodes with security. We will discover these improvements in our future research, and future work will focus on comparing existing harvesting techniques such as backscattering based WSN, ambient-backscattering, and remappable WSN, which provide broad applications of energy utilization, and will give remarkable results within the wireless network field.