Indoor Mobile Node Localization Algorithm based on RSSI and Improved MCL

Abstract

With the characteristics of mobile sensor network for dynamic change of indoor network structure and node position, in order to overcome the shortage of localization accuracy and sampling efficiency of Monte Carlo localization algorithm in wireless sensor networks (WSN), a location algorithm based on RSSI and improved Monte Carlo localization is proposed to locate indoor mobile nodes. First, the paper describes the classical MCL algorithm and the received signal intensity RSSI model, and then an improved MCL algorithm is designed. The algorithm through introducing the received signal strength indicator model of range prediction, using distance information filtering samples, and finally using the filtered samples of the weighted average to estimate the location of nodes and reducing the sampling area, improves the sampling efficiency and positioning accuracy. The simulation results show that the improved MCL algorithm based on RSSI improves the convergence speed compared with the traditional algorithm, and uses the distance information to filter samples to reduce the computational overhead and improve the positioning accuracy. Under the same conditions, the improved MCL algorithm based on RSSI reduces the positioning error by about 45% compared with the traditional Monte Carlo algorithm.