A Monte Carlo Box Localization Algorithm Based on RSSI

There are some common problems, such as low location accuracy and low sampling efficiency, existing in the present node localization algorithms that are based on Monte Carlo Localization (MCL) in mobile wireless sensor networks. To improve these issues, a Monte Carlo box localization algorithm based on RSSI(MCBBR) is proposed in this paper. In the algorithm, sampling box was constructed through RSSI ranging as the optimal space for location estimation, sample number was adaptive according to the size of sampling box, and genetic algorithm method was referenced to optimize samples. Finally the mean value of all samples was the optimal location estimation. Simulation results show that the proposed algorithm can enhance the location accuracy by 30% comparing to MCB algorithm, and 10% comparing to Range-Based MCL algorithm. Furthermore, the results also show that the algorithm can achieve a higher sampling efficiency. Thus, MCBBR can be applied in the circumstance where the high location accuracy and sampling efficiency are required.