CAJ | 학술논문

为了准确地识别通信信号的调制方式，运用小波变换与支持向量机(SVM)对调制信号类型进行识别。采用小波分解重构方法对常用3种模拟信号和6种数字信号提取特征值，将提取的小波特征参数送到 SVM判决器，对信号调制类别进行训练与测试，得到平均识别率。实验结果表明，在信噪比不低于5 dB时，识别率达到了100%，具有良好的分类性能和抗噪能力。
위료준학지식별통신신호적조제방식，운용소파변환여지지향량궤(SVM)대조제신호류형진행식별。채용소파분해중구방법대상용3충모의신호화6충수자신호제취특정치，장제취적소파특정삼수송도 SVM판결기，대신호조제유별진행훈련여측시，득도평균식별솔。실험결과표명，재신조비불저우5 dB시，식별솔체도료100%，구유량호적분류성능화항조능력。
In order to identify accurately the modulation scheme of communication signals,the modulated signal types are i-dentified by combining the wavelet analysis with support vector machine (SVM).The eigenvalues of the three kinds of com-mon analog signals and the six kinds of digital signals are extracted by using wavelet decomposition and reconstruction meth-od.Then they are sent to the SVM j udgment,and the average recognition rate is obtained through training and testing the signal modulation types.The experimental results show that the recognition rate of communication signal modulation types is up to 100% when the signal-to-noise ratio is not less than 5 dB.The method has good classification performance and noise immunity.