Improving Signal Modulation Recognition Using Principal Component Analysis And Compressive Sensing

In order to solve the problem of low recognition rate of communication signals under low signal to noise ratio(SNR), a new dimensionality reduction method is proposed, which is based on compressive sensing and principal component analysis. Compared with the traditional dimensionality reduction methods, such as principal component analysis and compressive sensing, the proposed method has greater noise reduction function. Using random forest as a classifier, the recognition rate of several communication signals reached 90% at SNR>-10dB.