A multiobjective optimization based sensor selection method for target tracking in Wireless Sensor Networks

In this paper, we propose a sensor selection strategy for target tracking in Wireless Sensor Networks by formulating it as a multiobjective optimization problem (MOP). At each time step of tracking, we obtain tradeoff solutions between two conflicting objectives: minimization of the number of selected sensors and minimization of the information gap between the Fisher Information when all the sensors transmit measurements and the Fisher Information when only the selected sensors transmit their measurements based on the sensor selection strategy. Simulation results show that the sensor selection strategy which is closest to the utopia point of the MOP adaptively decides on the required number of selected sensors at each time step of tracking and achieves estimation performance that is near the estimation performance when all the sensors transmit.