Study Of Design And Simulation In Cognitive Tracking Radar System

With the constant improvement of the modern information, the development of military technology is increasingly high demanding of modern radar. The variety of targets, environmental complexity and diversification of tasks, promote the development of radar technology towards intelligentizing. Cognitive radar as a new kind of intelligent system has been studied extensively, which uses the perceptive target, external environment information and other prior knowledge to choose emission signals intelligently.In this dissertation, cognitive tracking radar system is established using the Kalman filter applied more widely instead of the Bayesian filter. Firstly, the overall architecture, mathematical model, tracking principle and system design details are studied for cognitive tracking radar based on the cognitive radar. Secondly, selection and simulation of the transmitted signal for cognitive tracking radar system, two signal models of better performance are choosed: chirp signal and ultra-wideband chaotic signal. Chirp signal using pulse compression technology and ultra-wideband chaotic signal owning large time-bandwidth product characteristic with sine wave embedded, both can well solve the contradiction between the detection accuracy and resolution when they serve as transmitted signals. Furthermore, the echo signal filter processing methods at receiver of cognitive tracking radar are studied and simulated. Iterative cubature Kalman filtering algorithm is established after analyzing the Kalman filtering, extended Kalman filtering, unscented Kalman filtering and cubature Kalman filtering algorithm. Applying this algorithm, the estimation error of the target’s initial state and the propagation error introduced by linearized measurement equation are reduced, and then the performance of cognitive tracking radar is achieved. Finally, several filtering algorithms are simulated with MATLAB. The results show that the proposed algorithm achieves significantly higher tracking accuracy, better stability and higher fault tolerance to initial error compared to traditional filtering algorithms.