Study On Cognitive Radar Theory And Waveform Selection Algorithm

Cognitive radar is the next generation of intelligent radar, it can form a closed loop dynamic system which includes the target environment, adjust the transmitted waveform radar transmitter adaptively according to the target environment, and continually get information and interact with the target environment, thereby improving the overall radar performance.How to adaptively choose the transmitted waveform is one of the research hotspots of cognitive radar, current waveform adaptive selection algorithm is mainly based on the one-step state prediction and optimization method, the system is "greedy" and only focus on the next time performance and failed to consider in the future. And in the target tracking problem, we are more concerned about the long-term tracking system performance, namely tracking stability. Based on multi-step target state prediction waveform optimization has become one of the issues of cognitive radar. However, due to the prediction accuracy and computational complexity, processing multi-step waveform optimization problem in real-time is facing many difficultiesThe thesis firstly reviews the formation of the concept of cognitive radar and studies the current research status of the related technologies, emphases on the basic principles and structural properties of the cognitive radar.On this basis, for lack of the tracking performance of cognitive radar target tracking which uses traditional one-step prediction and optimization method. We propose waveform rolling optimization algorithm that based on multi-step state prediction. Provide a recursive process of multi-step state prediction based on Kalman filtering theory, and use dynamic programming method to solve the multi-step waveform optimization problem.Finally, the above method is applied to reentry trajectory tracking problem, computer simulation results illustrate that the feasibility and effectiveness of this method, and also to verify the performance of waveform scroll optimization algorithm that based on the multi-step prediction achieved to enhance tracking stability purpose.