Abstract：Aimingat the problem that the traditional method(chi-square) has poor ability to identify false targets when the angle difference between the radars relative to the target is small, an adaptive anti-rangedeception jamming method combining chi-square test and support vector machine is proposed. Firstly, determine whether further screening is required according to the number of combinations that meet the chi-square test; If further screeningis required, use the Mahalanobis distance(chi-square test variable), Euclidean distance, and Randolph distance between tracks as the multi-feature input of SVM. Then, train the SVM model. Finally, use the trained SVM to identify true and false tracks. The simulation results show that the true target track recognition rate of this method is 95.5%, which is better than the chi-square test and the improved chi-square test method significantly. In addition, the training time of this method is short; the recognition time is equivalent to the traditional method; the real-time performance is good, and the engineering application is strong.
施裕升, 王晓科, 周宇泰, 蒋国韬, 徐天洋. 基于卡方检验与SVM的多雷达抗欺骗干扰方法[J]. 空天防御, 2022, 5(1): 108-114.
SHI Yusheng, WANG Xiaoke, ZHOU Yutai, JIANG Guotao, XU Tianyang. Multi-Radar Anti-Deception Jamming Method Based on Chi-Square Test and SVM. Air & Space Defense, 2022, 5(1): 108-114.
