Research On Target Detecting And Tracking Methods For Cognitive Radar

Cognitive radar has the capabilities of perceiving and understanding environment,meanwhile,making an appropriate policy by introducing closed-loop feedback from the receiver to the transmitter,so that it can accomplish radar tasks in complex and changeable electromagnetic environment.Cognitive radar has two key technologies:waveform optimization technology and adaptive mechanism.This paper focuses on the application of these two key technologies in the detection and tracking of two typical radar missions.The main works are listed as follow:(1)Waveform optimization method of cognitive radar target detection and identification is discussed,based on the theory of statistical hypothesis testing,the target response model and the transceiver model in clutter environment are established.Then the objective function of optimal waveform is constructed based on Neyman-Pearson criterion and the optimal waveform is solved.The optimal transmit waveform selection algorithm for single target tracking is studied in this paper,it illustrates the relationship between the transmit waveform parameters and Kalman filtering,meanwhile,expound the minimum mean square error criterion and the maximum mutual information criterion for the selection of transmit waveform parameters in tracking.(2)This paper presents two adaptive mechanisms,which are the closed loop feedback of radar transmitter-receiver and the closed loop feedback of detector-tracker in receiver,establishes a cognitive probability data association for single target tracking in clutter environment.In the closed loop feedback of radar transmitter-receiver,the waveform selection model is added to the traditional probability data association algorithm to improve tracking precision.In the closed loop feedback of detector-tracker in receiver,the detection and tracking are processed synthetically,and the joint probability of amplitude and position is introduced to modify the correlation probability.Compared with the traditional algorithm,the target detection performance and tracking precision are improved.(3)The waveform selection model is added to the traditional joint probability data association algorithm in the single-beam transmit model,establishes a cognitive joint probability data association for multi-target tracking in clutter environment.Multi-target with cross-track is simulated and analyzed.Results show that the tracking accuracy is improved compared with the traditional algorithm.Then,the intelligent limited power allocation method based on multi-beam for multi-target tracking is studied,it improves the tracking accuracy of worst target.(4)An adaptive maneuvering target tracking algorithm is proposed,an adaptive gate adjustment method is adopted to deal with the suddenly increase of the tracking error caused by the target maneuver,and the waveform selection module is introduced to make the tracking error approximate to the lower bound of the Crame-Rao,the robustness of the waveform selection algorithm is also verified.