Identification of PSK signals

PSK signals may be identified using decision theoretic techniques. The paper compares the performance of the optimum, the statistical moments, the DFT and the maximum likelihood DFT classifiers. The robustness of each classifier is examined for the effects of symbol imbalance due to a finite signal time frame, error in the SNR estimate, channel filtering and phase error. Simulation results are presented in terms of the SNR at which 1% misclassification probability occurs, in order to provide a comparison between the techniques.