Global sensor selection for maneuvering target tracking in clutter

This paper studies the problem of sensor selection for maneuvering target tracking in the cluttered environment. By modeling the target dynamics as jump Markov linear systems, a decentralized tracking algorithm is developed by applying the extended Kalman filter and the probabilistic data association technique. A cost function that minimizes the expected filtered mean square position error is utilized and a sensor selection scheme is proposed. A numerical example is provided to illustrate the effectiveness of the proposed approach.