A Particle Filter for Objects Tracking in Cognitive MIMO System

In this paper, we concern with a Cognitive radar operating in an environment with multiple interested targets. It is an intelligent parameter estimation system that the transmitted signal depends on the input signal. The mean-square error of the estimated parameters is derived in the Cramer-Rao Lower Bound that represent the performance of the system. To improve the ability to find a new target when the number of targets is changing, we degenerate the prior information about the targets. A Bayesian approach is used to estimate state space of the system or tracking target parameters. For non-linear system and non-Gaussian distribution, a sampling method called particle filter is introduced. We also consider the effect of the Importance Sampling in the particle filter to the performance of the system. Finally, the simulation result is provided to demonstrate the capability of the cognitive radar system to track the targets.