Application of Wavelet Denoising and Time- Frequency Domain Feature Extraction on Data Processing of Modulated Signals

Abstract

Signal modulation is an essential part of the communication system. Researches on reducing the noise interference in the modulated signal and improving the signal-to- noise ratio can help recognize the signal. In this paper, it is proposed to apply the wavelet denoising and time-frequency domain feature extraction to the modulated signals. Combine the characteristics of the modulated signal and the theory of wavelet denoising and time-frequency domain feature extraction, remove the noise interference in the signal, extract the time-frequency domain feature, and input the processed data into the decision tree model for classification and recognition and evaluate the signal processing effect. In addition, the accuracy of decision trees before and after data processing under different signal-to- noise ratios is further studied. Experimental results show that wavelet denoising and time-frequency domain feature extraction obviously affect modulated signal processing and are generally applicable under different signal-to-noise ratios.