Adaptive waveform and Cartesian estimate selection for multistatic target tracking

Abstract

This paper considers the problem of target tracking by a multistatic radar system. In order to use linear Kalman filtering for tracking, the time delay, Doppler shift and arrival angle measurements from multiple receivers are transformed into target position and velocity estimates in Cartesian coordinates before being processed by the Kalman filter. A new multistatic tracking algorithm is proposed with the novel feature of joint adaptive selection of transmitted waveform and Cartesian estimate to minimize the mean-square error of the tracking estimate. Cramér–Rao lower bounds (CRLBs) of Cartesian estimates are derived and exploited in the development of the joint adaptive selection scheme. Three main performance advantages of the proposed algorithm are (i) reduction of bandwidth and power consumption in transmitter–receiver communication links by utilizing a single Cartesian estimate in tracking process, (ii) optimization of the tracking performance by jointly selecting waveform and Cartesian estimate, and (iii) the inherent benefits of linear estimation from use of the linear Kalman filter including stability. Simulation examples demonstrate the dependence of the CRLBs of the Cartesian estimates on the transmitted waveform, target position and velocity, radar geometry, and particular elliptic-to-Cartesian transformation. Simulations also highlight the superior performance of our proposed algorithm over existing techniques.