Enhanced Hybrid American Zebra Particle Swarm Optimization for Improving Energy Efficiency in Wireless Sensor Networks

The vast array of applications for Wireless Sensor Networks (WSNs) has made them a popular study topic in recent years. One major issue that WSNs must deal with is energy consumption, which directly affects the network's operational lifetime. To address this issue, clustering has emerged as an effective solution for minimizing energy usage; however, traditional algorithms like LEACH suffer from limited scalability and suboptimal cluster head selection. In clustering, SNs are constructed into multiple clusters while formed clusters are assigned a cluster head (CH) for centralizing communication and reduce energy consumption. This paper introduces a Hybrid American Zebra Particle Swarm Optimization (HAZPSO) algorithm, achieving up to 35\% reduction in energy consumption compared to previous algorithms and increases the number of alive SNs by up to 40\% compared to existing methods. HAZPSO algorithm combines the particle swarm optimization algorithm with American Zebra to achieve suitable CH. The algorithm works in two primary stages: the first stage selects the best CHs, and the second stage forms clusters using the Improved K-mean Clustering Algorithm (IKCA) based on the centroids identified in the first stage. Comparisons with eight other algorithms are used to assess the suggested HAZPSO algorithm's performance. The findings demonstrate that HAZPSO and IKCA considerably lowers energy usage while raising the nodes' leftover energy. In addition, compared to previous algorithms, it keeps a greater number of nodes alive, extending the network's operational lifetime. The results show that HAZPSO reduces energy consumption by 20\% compared to GWO and 15\% compared to MPA, while increasing the residual energy of the nodes. A future study might involve applying the suggested HAZPSO algorithm to real-time data operators and evaluating it in dynamic scenarios for the BS.