Distributed learning in mobile sensor networks using cross validation

Mobile sensor networks can increase sensing coverage both in space and time and robustness against dynamic changes in the environment, compared to stationary wireless sensor networks. For operations in a dynamic or unknown environment, mobile sensors need the capability of learning a suitable model during its operations. However, due to the limited communication bandwidth, it is prohibited to share all measurements with other mobile sensors. In this paper, we propose an efficient distributed learning algorithm based on cross validation for mobile sensor networks, which takes the advantage of a multi-agent system and minimizes the communication overhead while achieving excellent performance, and demonstrate its performance in simulation.