Adaptive Neuro‐Fuzzy Inference System‐Particle swarm optimization‐based clustering approach and hybrid Moth‐flame cuttlefish optimization algorithm for efficient routing in wireless sensor network

Abstract

SummaryThe selection of good rendezvous points (RPs) is a significant role for M2obile Sink (MS) in the wireless sensor network (WSN). For the mobile sink, the selection of RP is one of the major problems in WSN. The rendezvous points are only selected based on the local information, so the possibility of selecting an optimal sensor node as RP will be extremely low. The next problem is to find the mobile sink path which visits all the RPs. The above problem is comprehended by utilizing an optimization algorithm. In this paper, Adaptive Neuro‐Fuzzy Inference System‐Particle swarm optimization based clustering approach and hybrid Moth‐flame cuttlefish optimization (MFO‐CFO) algorithm for efficient routing in WSN is proposed. Initially, the number of clusters is formed due to the Fuzzy C‐means (FCM) based Ant lion optimization (ALO) clustering approach. An Adaptive Neuro‐Fuzzy Inference System (ANFIS) based Particle swarm optimization (PSO) technique is presented for best cluster head (CH) selection. Here, residual energy (RE), node degree, and histories are considered as an input parameter to find the optimal cluster head. At last, we use hybrid MFO‐CFO techniques to find the minimum RPs which decreases energy consumption. The performance metrics of end to end (E2E) delay, energy consumption, channel load, throughput, bit error rate (BER), Network Life Time (NLT), packet delivery ratio (PDR), latency, jitter, and packet loss are computed using several nodes. The performance is compared with some existing algorithms such as Genetic Approach (GA), Fuzzy, PSO (Particle Swarm Optimization), Multiple Access Data Gathering using Mobile Sink for Path Constrained Environment (MADGAMSPCE).