Use of pseudo measurement to real-time target tracking

In this paper, a pseudo measurement model is adopted to deal with the Kalman filter problem with color measurement noise. It is assumed that the measurement noise is color, and is described by an AR(n) model. The state equation is set up according to Singer´s model. The measurement equation is set up in polar coordinate and the range rate measurement is combined with ordinary data (i.e., there are four measurement data: range r, azimuth θ_a, elevation θ_e and range rate r˙). By means of proper coordinate transformation, the matrix of Kalman gain and the matrix of filter error covariance can be decoupled. Thus the computing requirement of the algorithm is reduced and can be performed in real time