采用主分量分析的非合作水声通信信号调制识别

由于信道传输特性、信噪比低等因素的影响，非合作水声通信信号的调制识别极具挑战性。对信号功率谱、平方谱进行主分量分析，提取代表不同类型调制信号特有信息的主分量作为特征参数，从而降低特征参数维度、抑制噪声影响，并在此基础上设计一种基于人工神经网络的水声通信信号调制方式分类器。海上实录信号数据的识别实验结果表明了该方法的有效性。

Modulation Recognition Method of Non-cooperation Underwater Acoustic Communication Signals Using Principal ComponentAnalysis

The modulation classification of the non-cooperation underwater acoustic communication signals is extremely challenging due to channel transmission characteristics and low signal-to-noise ratio. The principal component analysis (PCA) is used to analyze the power spectra and square spectrum features of signals, which is capable of extracting the principal components associated with different modulated signals as input vector, thus reducing the feature dimension and suppressing the influence of noise. An artificial neural network (ANN) classifier is proposed for modulation recognition. The experimental modulation classification results obtained from field signals in 4 different underwater acoustic channels show that the proposed modulation recognition method has good classification performance.