Maneuvering target tracking in the presence of complicated measurement noises

Abstract

In modern radar tracking systems, measurement noise is significantly correlated when the measurement frequency is high enough. The problem of maneuvering target tracking in the presence of complicated measurement noise is considered in this paper. The measurement noise is modeled as the sum of a high‐order autoregressive process and a white process. This noise can be decorrelated by including the noise‐correlation variables in the target state. The theoretical analysis of tracking performance is derived. If some of the parameters (including the noise‐correlation parameters) are unknown, these unknown parameters can be estimated adaptively using a modified innovation correlation method. This parameter estimation method is very useful when the measurement noise can be modeled as the sum of a first‐order Markov process and a white process.