The Disquisition Of Targets Tracking In The Bistatic Forward Scattering Radar

Bisktic forward scattering radars (BFSR) provide effective detection and tracking of very small radar cross section (RCS) targets due to bistatic RCS rises (up to 20-40 dB relative to monostatic RCS) in narrow region close to the base line between transmit and receive positions. It can be used to detect, recognize and track the target with small detection probability in the low altitude sky, on the land or on the sea. Then the key issue of imaging and detecting is the precise estimation on parameters of moving targets.The application of three algorithms of Extended Kalman Filter (EKF), Unscented Kalman Filter (UKF), Particle Filter (PF) in BFSR's tracking targets of very small radar cross section (RCS) is investigated. Two target models are established based on the knowledge of Doppler shift and DOA obtained from the BFSR's observed datum. The first one is composed of three parameters: x-coordinate, y-coordinate and velocity. The second one is composed of four parameters: x-coordinate, y-coordinate, velocity and acceleration. Then four target models are established based on the movement status of targets. Those target models are constant-velocity moving model,constant-acceleration moving models and two kinds of maneuvering targets models: Singer model and current statistical model, respectively. The parameters of these target models are simulated and estimated using the three algorithms. And the efficiency and accuracy of the estimation has been greatly approved by using the approximation and Gauss-Newton iterance.