Communication Modulation Recognition Technology Based on Wavelet Entropy and Decision Tree
Pengfei An, Yiwen Sun, Xinyi Li · 2021 2nd International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT) · 2021
The signal modulation recognizer is a very critical system component in both military and civilian communication systems. In this research, the characteristics of time domain, frequency domain, and wavelet entropy are extracted from different modulation signals. The wavelet entropy includes wavelet packet energy entropy, scale entropy, and singular entropy. The feature extraction parameters are used as the input of the feature layer of the decision tree model for training and verification. Experiments show that the time-frequency domain, and wavelet entropy feature extraction of the signal can more completely represent the characteristics of the signal, while reducing the number of training calculations, improving training efficiency and recognition accuracy.