Abstract
Digital signal modulation recognition technology serves as the foundation and basis for signal demodulation, playing a crucial role in communication signal reconnaissance and holding significant research significance. This paper conducts research on digital signal modulation recognition technology from the perspectives of signal preprocessing and feature extraction. Seven modulation signals, namely 2ASK, 4ASK, 2FSK, 4FSK, 2PSK, 4PSK, and OFDM, are selected as recognition targets. The paper compares the effects of four different endpoint detection algorithms on modulation signal recognition. The results indicate that, for these seven modulation signals, the short-time energy entropy ratio algorithm performs the best, achieving a correct endpoint detection rate of over 93% in a Gaussian channel with a signal-to-noise(SNR) ratio of 0dB. Based on this, three different denoising algorithms are introduced to further enhance the performance of the short-time energy entropy ratio algorithm. The results show that the wavelet denoising algorithm achieves the greatest improvement in the performance of the short-time energy entropy ratio algorithm, with a short processing time. In a Gaussian channel with a SNR ratio greater than − 10dB, the endpoint detection accuracy of this algorithm can be maintained at over 95%. Finally, for the accurate identification and differentiation of 2FSK and 4FSK, this paper optimized the relevant algorithm in the cyclic spectrum. The kurtosis coefficient value Kur of the cyclic spectrum parameter matrix at the cyclic frequency \(\alpha =0\) is utilized to distinguish between these two signals. The results show that, at a SNR ratio of 4dB, the modulation recognition algorithm proposed in this paper can effectively distinguish between these two signals, achieving a recognition accuracy of over 99%.