Airborne External Radiation Source Radar Target Detection Based On Subspace Methods

Airborne passive radar,also known as external radiation source radar,is a radar system that detects targets based on existing electromagnetic waves in space.The system has high concealment and anti-interference performance,and has become an indispensable means of detection on today’s battlefield.However,in this radar system,the clutter distribution is non-uniform,which makes it difficult to obtain enough independent and identically distributed training data.In order to effectively suppress clutter,the Space-Time Adaptive Processing(STAP)algorithm can be used,but it needs enough training samples,and the inverse calculation of the covariance matrix is very heavy.The CG-STAP algorithm based on Krylov subspace can achieve good performance with fewer training samples,and its computational load will also be reduced.Thesis studies the clutter suppression of airborne passive radar based on the Krylov subspace CG-STAP algorithm.The main work is as follows:1、Introduced the related theory of airborne platform signal detection,STAP algorithm,feature subspace MNE algorithm and CG algorithm of Krylov subspace.The geometric scene construction of the airborne passive radar system is completed,and the ECA-B algorithm is used to cancel the direct wave.For four typical motion scenarios,the clutter of airborne passive radar is modeled and simulated,and the characteristics of clutter in different scenarios are analyzed.2、The method of estimating the rank of clutter covariance matrix in the scene of airborne passive radar is studied.Firstly,the method of estimating the rank of clutter covariance matrix for monostatic radar proposed by Brennan is analyzed,and combined with the actual scene,the rank of covariance matrix is analyzed.Aiming at the characteristics of the clutter covariance matrix of airborne passive radar,a rank estimation method suitable for the radar is proposed.Further combining the obtained rank with the CG gradient algorithm,the CG-r-STAP algorithm is proposed,and the effectiveness and performance of the algorithm in the airborne passive radar scene are analyzed,and compared with the STAP algorithm and feature subspace MNE algorithm.3、The clutter suppression algorithm based on the subspace method is studied in the scene of airborne passive radar.When the covariance matrix changes,two Krylov subspace CG-STAP algorithms based on covariance matrix update and weight vector adaptive update are studied on the basis of the original covariance space-time adaptive processing.The detection results and effectiveness of the algorithm are analyzed in the two scenarios where the unit sliding window and the external radiation source are rotated at a certain angle.4、The clutter suppression algorithm based on the subspace method in the scene of airborne passive radar is completed on FPGA.First,complete the overall framework of the FPGA implementation of the system,and design each module in the framework,and finally implement the CG-STAP algorithm for adaptive update of the weight vector on the FPGA,the clutter is suppressed,and can be detected The moving target signal,and the correctness of the design is verified by comparing the matlab experimental data with the modelsim simulation data.