Improving Monte Carlo Localization algorithm using genetic algorithm in mobile WSNs

In wireless sensor networks, location information is essential for the monitoring activities. Accessing the locations of events or determining the locations of mobile nodes is one of basic functions of wireless sensor networks. Except for normal information, sensor nodes should also provide position information of sensor nodes. So it's necessary to have a reliable algorithm for localization. Using GPS (Global Position System) technology is a good way to fix position in many fields, and high precision and performance could be obtained in outdoor environment. However, high energy consumption and device volume make it not proper for the low cost self-organizing sensor networks. Some researchers used Monte-Carlo Localization (MCL) algorithm in mobile nodes localization, and revealed that better localization effects could be obtained. However, current MCL-based approaches need to acquire a large number of samples to calculate to achieve good precision. The energy of one node is limited and can't last for a long time. In this paper, a new method has been suggested to apply genetic algorithm to improve MCL in MSNs for localization. Experimental results illustrate that our methodology has a better performance in comparison with Monte Carlo localization algorithm.