For the presence of interference signals,a modulation recognition algorithm is proposed based on independent component analysis( ICA) and support vector machine( SVM) . The multiple-antenna-recep-tion and the ICA method are used to extract the independent component from the sample vector. The multi-ple-valued classifier based on two-valued SVM is designed. The proposed algorithm has faster training speed and better recognition performance. The extensive simulation results show that the correct ratio of the recognition algorithm is greater than 0. 9 as SIR≥10 dB and speed less than 12 m/s. The corresponding studies are meaningful for engineering application.%针对干扰环境提出了一种基于独立成分分析( ICA)和支持矢量机( SVM)的通信信号调制识别算法。算法利用多天线接收技术，采用独立成分分析方法寻求观测样本矢量的统计独立分量，设计了以二值支持矢量机( SVM)为基础的多值分类器，该算法具有较高的训练速度和较好的分类性能。仿真表明，当信干比大于10 dB、移动速度小于12 m/s时，算法的总体正确识别率超过0.9。干扰环境下的信号调制识别研究具有较重要的实际工程应用意义。
展开▼
