雷达信号处理是现代雷达系统的核心内容之一，其直接影响着雷达系统的适用范围和工作性能等。而对雷达信号的有效识别是对未知雷达信号进行预判的重要组成部分。基于支持向量机（SVM ）对四种不同的雷达信号智能辨识，选取径向基核函数（R BF）作为支持向量的非线性映射函数，经过理论推导得出惩罚因子c和核函数参数g是影响其分类性能的重要因素。利用粒子群（PSO）优化SVM 的两个重要参数。结果表明，在没有进行参数优化的SVM 的分类性能极其不稳定，识别准确率在79.6992% ~90.2256% 之间，而经过PSO 优化的SVM 分类准确率高达100% ，有效证明了优化方法的有效性，实现了基于PSO 优化的SVM雷达信号的准确识别。

Radar signal processing is one of the core contents of modern radar systems, which directly affects the scope and perfor- mance of radar systems. The effective identification of radar signals is an important part of predicting unknown radar signals. This paper uses the support vector machine (SVM) to intelligently identify four different radar signals, and selects the radial basis func- tion (RBF) as the support vector’s non-linear mapping function. The theoretical derivation shows that the penalty factor c and the kernel function parameter g affect its classification performance. This paper uses particle swarm optimization (PSO) to optimize two important parameters of support vector machines. The results show that the classification performance of SVM without parameter optimization is unstable and the recognition accuracy is between 79.6992% and 90.2256%, while the accuracy of SVM optimized by particle swarm optimization algorithm is as high as 100%. The effectiveness of the optimization method is effectively proved, and the accurate recognition of radar signals of SVM based on particle swarm optimization algorithm is realized