M-ary Pulse Ampplitude Modulation Recognition Using Discrete Meyer Wavelet and Reverse Biorthogonal Wavelet

Automatic modulation recognition (AMR) is a fundamental task in communication systems. Feature extraction (FE) is an essential part in the recognition system,the proper selection of FE will enhance the recognition accuracy, and reduce the complexity of the system. In this paper, Reverse Biorthogonal wavelet (RBW), andDiscrete Meyer Wavelet (DMW), followed by standard deviation are used for FE. They are used to reduce the FE sets, and complexity of the recognition system.Adaptive Neuro Fuzzy Inference system is used as a classifier, to classify the,M-ary Pulse Amplitude Modulation (PAM) signals (i.e.4PAM, 8PAM, 16PAM, 32PAM, 64PAM, and128PAM), in a wide range of signal to noise ratio (SNR). MATLAB programs were used to fulfill all the requested tasks.The results show that the recognition system of M-ary (PAM) signals exhibits a satisfactory level under low SNR, and the system can achieve success rates over 98% in SNR ( from -2 to 12) dB.