A Hybrid PSO and K-Means Clustering Approach for Enhanced Sensor Localization in WSNs

In Wireless Sensor Networks (WSNs), correct sensor location is crucial for optimal data collection and transmission. This work introduces an improved methodology for sensor placement that integrates Particle Swarm Optimisation (PSO) with K-Means clustering to boost the efficacy of RSSI-based localisation. To measure the distance between sensors, researchers commonly employ the Received Signal Strength Indicator (RSSI); nevertheless, errors may be introduced by outside variables including interference and signal attenuation. To tackle these problems, we present a hybrid approach whereby Particle Swarm Optimisation (PSO) is employed for the global optimisation of sensor placements, while K-Means clustering enhances the sensor aggregation based on proximity and Received Signal Strength Indicator (RSSI) signal intensity. The suggested technique reduces localisation errors by optimising sensor placement and classifying nodes into suitable clusters. Experimental findings indicate that the hybrid PSO-K-Means method markedly improves positioning accuracy relative to conventional localisation strategies. The results indicate that combining PSO with K-Means clustering is an efficient approach for enhancing sensor localisation for the considered performance criteria.