Alpha-beta tracking filter combined with ellipsoidal prediction using the generalized Hough transform

In track-while-scan (TWS) systems using phased array antennas, an α-β filter has been employed. An α-β filter can estimate reliably the target position and velocity. However, when the target maneuvers, the quality of the position and velocity estimates provided by an α-β filter can degrade significantly. The target may be lost, since the filter uses only linear prediction. As the nonlinear tracking filter for a maneuvering target, the α-β tracking filter combined with maneuver-driven circular prediction was proposed. On the other hand, the generalized Hough transform has been used for detecting curves in image processing. In this paper, we take account of the fact that a maneuvering target trajectory can be regarded as an ellipsoidal arc and propose the use of an α-β tracking filter combined with ellipsoidal prediction using the generalized Hough transform to improve tracking quality over the α-β filter combined with circular prediction. Simulation results for two target profiles are included for comparison of the performance of our proposed scheme with that of a conventional α-β tracking filter, the Kalman filter, the two-stage Kalman filter, and the α-β filter combined with circular prediction. The effectiveness of our proposed scheme is examined by means of a computational simulation.