针对基于SVM的雷达辐射源识别方法中SVM模型参数对识别性能影响较大的问题,提出一种新的雷达辐射源识别方法。该方法将智能优化算法应用于SVM,对SVM参数进行选择以提高识别准确率;为分析所提出新方法的性能,提出有效解的标准差、解的质量和精度时间比作为评估指标对所提方法进行性能评估。通过计算机仿真,验证了新方法的有效性,并分析了3种典型优化算法在新方法中的综合性能。
Aiming at the problem that support vector machine (SVM) model parameters have great impact on recognition result in the method of radar emitter recognition based on SVM, a new radar emitter recognition method is proposed. Three intelligent optimization algorithms are applied in SVM and selec- ting the SVM parameters to improve the recognition accuracy in this method. The standard deviation, quality and accuracy time ratio of effective solution as the evaluation indicators are used to analyze and e- valuate the performance of the new method. Computer simulation shows that the new algorithm is availa- ble and the comprehensive performances of three typical optimization algorithms used in the new algorithm are analyzed.