论文部分内容阅读
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。
A novel modulation recognition method based on support vector machine (SVM) is proposed to extract the classification features by using wavelet transform. Multiple SVM classifiers are used to construct a multi-class classifier for modulation recognition of multi-class signals. In the IF processing, do not need synchronization information; support vector machine has better promotion ability, you can use less training samples. Computer simulation results show that the proposed method can work well in the communication environment with wide range of signal-to-noise ratio.