Performance Improvement of DE Algorithm for Indoor Positioning in Wireless Sensor Networks

Target positioning in wireless sensor networks (WSNs) are necessary for real applications. In this research paper, we employ the Differential Evolution (DE) algorithm to enhance the effectiveness of K Nearest Neighbor (KNN) algorithms integrated with receive signal strength indicator (RSSI) for indoor positioning. We examine the performance of random and fixed sensor deployment strategies in both simple and complex environments in relation to target positioning. Specifically, we investigate the impact of reference point numbers, where K = 4 for simple environments and K = 5 for complex environments. The simulation results demonstrate that setting the K value to 4 yields the highest average correct rates in simple and complex environments, reaching 99.54% and 99.7% respectively. Moreover, the performance improvement between using 5 and 4 reference points is less than 1% in all cases, except for a 2.02% increase observed in the complex environment with random deployment.