Modulation identification by the wavelet transform

There is a need, for example in surveillance of the radio spectrum, to determine the modulation type of a received signal. By knowing the modulation type, one can demodulate the incoming signal for useful information. This paper proposes the use of wavelet transform for the identification problem. Application of wavelet transform on a digital modulation signal results in distinctive patterns for different types, which enables simple processing for identification. Three identifiers for classifying PSK and FSK, M-ary PSK, and M-ary FSK are considered and the relevant statistics for optimum decision making are included. Simulations are included for performance evaluation