Optimal Clustering With Weighted Least Square Method for RSSI-Based Localization in Wireless Sensor Networks

Accurate node localization is crucial in wireless sensor networks (WSNs) for tasks such as environmental monitoring, target tracking, and surveillance. Existing methods face challenges in achieving optimal convergence rates, reliable clustering, and accurate position estimation, motivating the development of a novel approach. This study addresses the critical need for accurate node position identification in WSNs. The proposed methodology integrates the improved coati optimization (ImCo) clustering algorithm, which incorporates an adaptive strategy for enhanced convergence, with the average backward RSSI error (ABRE) method for anchor node selection and the weighted least square (WLS) approach for accurate target node position estimation. Comparative analysis demonstrates the superiority of the ImCo + ABRE + WLS method over existing techniques, showcasing its remarkable achievements. The minimal root mean square error (RMSE), localization error, energy consumption, and processing time achieved by ImCo + ABRE + WLS are 0.6470, 1.2372, 0.8553, and 4664.70, respectively. The proposed method significantly outperforms fuzzy C-means, neighborhood grid clustering (NGCGAL), and elite oppositional farmland fertility optimization (EOFFO) methods, highlighting the effectiveness of the proposed methodology in addressing key challenges and achieving superior performance in WSN node position identification.