Abstract:In the classical “Current” statistical model, maneuvering frequency is normally set as a typical value, and an acceleration limit has to be set. In order to break this limitation, this paper utilizes the recursive linear smoothed Newton method to predict acceleration. An analytical solution of target maneuvering frequency is then achieved, by which the need of setting maneuvering acceleration limit is avoided. Meanwhile, adaptive calculation of process noise covariance matrix is accomplished. Moreover, an adaptive update of measuring noise covariance matrix is established by using simplified Sage-Husa filter. The computer simulation demonstrates that the proposed algorithm is with a better range tracking performance.