【摘要】 提出了一种应用小波变换提取分类特征的基于支撑矢量机的通信信号调制识别方法。使用多个支撑矢量机分类器构成一个多类分类器用于多类信号的调制识别。在中频进行处理,不需要同步信息;利用支撑矢量机具有更好的推广能力,可以使用较少的训练样本。计算机仿真结果表明该方法可以很好地工作于信噪比变化范围较大的通信环境。更多还原

【Abstract】 A new method of modulation identification for communication signals, using wavelet transform to extract characteristics and based on support vector machines(SVM), is proposed. Several SVM classifiers are used to make a multi-classes classifier for modulation identification. The signals to be identified are processed on frequency band, so the synchronization information is unnecessary; Because of the better generalization performance of SVM, less training samples are needed. The good performance of the proposed identification scheme is proved through computer simulation.更多还原