Ameliorate the performance using soft computing approaches in wireless networks

Wireless sensor networks are an innovative and rapidly advanced network occupying the broad spectrum of wireless networks. It works on the principle of “use with less expense, effort and with more comfort.” In these networks, routing provides efficient and effective data transmission between different sources to access points using the clustering technique. This work addresses the low-energy adaptive clustering hierarchy (LEACH) protocol’s main backdrop of choosing head nodes based on a random value. In this, the soft computing methods are used, namely the fuzzy approach, to overcome this barrier in LEACH. Our approach’s primary goal is to extend the network lifetime with efficient energy consumption and by choosing the appropriate head node in each cluster based on the fuzzy parameters. The proposed clustering algorithm focused on two fuzzy inference structures, namely Mamdani and Sugeno fuzzy logic models in two scenarios, respectively. We compared our approach with four existing works, the conventional LEACH, LEACH using the fuzzy method, multicriteria cluster head delegation, and fuzzy-based energy efficient clustering approach (FEECA) in wireless sensor network. The proposed scenario based fuzzy LEACH protocol approaches are better than the four existing methods regarding stability, network survivability, and energy consumption.


INTRODUCTION
Microcontrollers, networking technology, microelectromechanical systems, and nanotechnology have developed novel sensing and communication systems, referred to as wireless sensor networks. These new technologies have also contributed to the advancement of nanotechnology (WSNs). Figure 1 presents a schematic representation of a wireless sensor network.
They are skilled with opportunity providing technology which pings its role in a wide range of applications that bare minimum human intervention and play a critical function in making the future smarter. It uses spatially distributed self-governing nodes that use sensors to track any circumstance, either physical or environmental. WSN are continuing to be the most effective aid to sensing and monitoring tasks. The ability to work in uncertain environments, easy implementation, and high performances are among the many reasons for the popularity. Their primary benefit is the ability to be carried out to any discipline and in any surroundings. Unlike popular networks that, for their application, require extensively stringent conditions. In this mechanism, node energy, in conjunction with balancing glide, is the primary trouble in which sources are scarcely to claim approximately a way to reduce strength intake and prolong the network lifetime for the structure of WSNs using existing routing protocols. In terms of productivity in WSNs, routing protocols [1], [2] play a vital role. Routing is one of the most challenging techniques on WSN. It is necessary to find the best path between the source node and the target node. This method is the mathematical portrayal of human concepts. The main motto of any network is the efficient and effective transmission of information between the source and the final destination. This routing is one of the best solutions to provide an efficient and best path for wireless sensor networks. In routing, in this work the clustering approach is chosen [3]-[9] for selecting the best way for intra-clustering and inter-clustering. The classification of routing protocols [10]- [12] relies upon two critical elements, i.e., first off, primarily based on network structure and secondly based on the protocol operation. The chore of discovering and retaining routes in WSNs is not minor, as energy restrictions and sudden changes in a node. WSN has attracted tremendous attention from academics and industry worldwide in the past, present, and future. The wide range of applications [13] of WSNs is flora and fauna, military, industrial, urban, environmental, health, education, entertainment [14]- [20]. WSN involve millions of nodes that operate together, detecting and transmitting ambient knowledge to the base station [21]. Apart from those have worked on multiple sink nodes in a WSN to increase the scalability and lifetime of the network [22]. Many cluster heads choose a specific sink node and try to send their data to the sink simultaneously. Sujith et al. [23] suggested an energy-efficient zone-based clustering algorithm for WSN. In their approach, they considered zones as clusters and zone monitor as cluster head. Bagga et al. [24] has initiated an fuzzy logicbased clustering routing (FLBCR) protocol as a routing scheme that applies fluctuating logic to determine the likelihood of selecting a node as a cluster head (CH) with a variable setting existence a network. For instance, an improved low-energy adaptive clustering hierarchy-mobile fuzzy (LEACH MF) protocol has been addressed to prolong lifetime of wireless sensor networks and they show in their results that the proposed modified parameter LEACH MF is better in performance and energy consumption [25]. This compared type1 and type2 fuzzy logic approach in choosing cluster head selection to increase the wireless sensor networks.

METHOD
The main steps of our proposed fuzzy based LEACH approach are discussed in this section. In proposed algorithm, optimum number of head nodes are chosen using Kopt from (1).
where b is the number of nodes, Th is the user defined threshold value, θ0 is the computed ratio of energy dissipated in the free space (µ) by the energy dissipated in multi path (ϑ), M is the product of area of the network and sink coordinates, and dBS 2 is the first form of Euclidean distance of respective node to the sink node. The current heads of nodes calculate their W i (output fuzzy parameter) value using the fuzzy method and broadcast the current head node (HN) message to all nodes coming under the communication range. The member nodes choose the nearest HN and join it to form clusters, respectively using dBS (2): where and are the length and width of the area of the network, respectively, and and are the sink coordinates. The residual energy γRe is calculated based on node coordinates Bbij and user defined threshold value ψ: where i,j are the coordinating positions of the node deployed in the network, b is the total number of nodes in the area, H is x coordinate of the node, and C is the y coordinate of the node. The calculations of Dist and d are shown in (4).
Initially, the cluster range will be considered as a value. Based on the calculated distance of the nodes with respect to the mote is checked. Whether the distance is less than the range of the cluster ℧ Clusterrange in the network is calculated as (5): where and are the length and width of the area of the network, respectively, and and are the sink coordinates.
where Thn is the tentative head node, d is the minimum distance. The CH generates a time division multiple access (TDMA) schedule and sends it to its members until the CH and its members are fixed. The method of choosing a head node by the use of fuzzy procedure is done in 4 steps for Scenario 1. a. Input and fuzzification: We fetch the three input values to the fuzzy inference system using (3), (2) and (7). Those input fuzzy parameters determine the cost of each enter based on the respective membership feature µ. b. Fuzzy inferences: We offer the membership values acquired to our if then rules to acquire to our new fuzzy set output. Our fuzzy if then rules have more than one input and the bushy and operator which selects the minimal of our 3 membership values. c. Aggregation: The aggregation is a unity of all the results from the implementation of all rules received.
We use an OR fuzzy logic operator when we aggregate all our rules in our FIS as shown in Table 1. To produce a new aggregate fuzzy set which will be used in the defuzzification stage, this operator chooses the limit of the rule evaluation values. d. Defuzzification: With a purpose to calculate the chance of each node, we mixture the effects of every rule the usage of Mamdani technique. This manner is known as defuzzification which unearths the threat price of each node to choose head node. In our proposed we use centroid method has been used for defuzzification.
The following components constitute the type-2 fuzzy logic model is done in 6 steps for Scenario 2 used in this work. a. Fuzzification module: This module maps the crisp input to a type-2 fuzzy set using the gaussian2 membership function. b. Inference engine: This module evaluates the rules in the knowledge base against the type-2 fuzzy set gotten from the fuzzification module to produce another type-2 fuzzy set. c. Type reducer: Type reducer uses Karnik-Mendel algorithm to reduce an interval type-2 fuzzy set to type-1 fuzzy set. d. Defuzzification module: It maps the fuzzy set produced by the type of reducer to a crisp output using the center of gravity defuzzification method. e. Fuzzy knowledge base: This is a database of rules to be used by the inference engine. f. Membership function: This mathematical equation helps the fuzzification module convert the crisp input into a fuzzy set. Figure 2 represents pseudocode for the proposed algorithm.

RESULTS AND DISCUSSION
All the following results in this paper are executed using MATLAB R2018a, which system configuration is Intel i5 processor (2.7 GHz) with 16 GB memory running on Windows 10 operating system. During the simulation process of data transmission between nodes from the source, through intermediate nodes and to the final destination nodes, each node uses its limited power, causing depletion. Any node which has reached a specific limit value of user choice is considered dead. The simulation parameters for performing our experiment are shown in Table 2. Here we compared three protocols fuzzy LEACH, multi-criteria cluster head delegation based on fuzzy logic (MUCH) and the proposed fuzzy logic LEACH. Simulation was done in MATLAB simulator and initially a total of 100 nodes are randomly scattered in a 100 m-by-100 m square area. Depicts the number of sensor nodes alive in the network. Figure 3 illustrates the point in time at which the first node is removed from the graph (FND). Based on the findings of the study, we are able to conclude that the FND can be found using the existing approaches even at the early rounds when compared with the methods that we have provided.  Figure 4 illustrates the round at which a quarter of the nodes are removed from the graph (QND). Based on the findings of the investigation, we can conclude that, in comparison to our suggested technique, the existing methods produce at earlier rounds. Figure 5 illustrates the round at which the half nodes are considered dead (HND). The findings of the study allow us to conclude that, in comparison to the way that we have presented, the existing methods produce HNDs at earlier rounds.  Figure 6 shows the total number of nodes alive after completion of a certain number of simulation rounds. It is observed that in initial fuzzy LEACH and one of the improved fuzzy LEACHs termed MUCH, has 99% of nodes die after completing of 5000 rounds but in our proposed methods only 81% and 93% of nodes die these results better energy efficiency of the nodes as well as improved network lifetime. The comparison is made based on residual energy, lifetime of sensor nodes, and throughput. These three parameters are used as comparison parameters. Figures 7 to 9 show the comparative analysis in the following subsections, respectively.

Throughput
The term "universal performance" or "throughput" refers to the amount of packets that are received by the sink using its available resource. As can be seen in Figure 7, our suggested protocol has a higher throughput in comparison to LEACH, LEACH with fuzzy, MUCH, fuzzy-based energy efficient clustering approach (FEECA), and both of our proposed methods, which are Scenario 1 and Scenario 2, respectively. Figure 7. Performance of the node based on data transmission among nodes to mote

Lifetime of sensor nodes
The average network lifetime of the wireless network has been calculated using three mentioned routing protocols. Here, the forwarder node is regarded in each round to be the node with the most considerable rest energy. As Figure 8 shows, the average network life will be extended when the information is transferred from nodes to sinks through our suggested protocol routing. In other words, in terms of average network life, our proposed plan outperformed existing schemes.

Residual energy
In this experiment, we have considered the average residual energy of all the nodes as a measure of performance and compared our protocol to three others. Higher residual energy is essential to extend the network life. From Figure 9, we can see that this residual average power is above the three mentioned protocols. Figure 9. Total energy dissipated among the nodes during each round

CONCLUSION
Wireless sensor networks are becoming relevant in a broad range of emerging technologies. One of the challenges to WSNs is to reduce the use of resources and increase the networks' existence, for which routing may be a remedy. As the propagation energy is proportional to the distance between the sender and the receiver, the clustering mechanism minimizes energy use in routing. Using fuzzy reasoning, WSN can resist complicated mathematical models and provide considerable stability in the networks life to deal with uncertainty and interpretation. These suggested strategies described in both scenarios are a revision of the option of LEACH to select the optimal number of head node selection and select the best head node selection in any round during each cluster. The simulation findings indicate that our proposed solution delivers more robust results than four other existing state-of-the-art algorithms and proves to be more scalable after completing rounds in FND, QND, HND, and the number of nodes alive.