Joint Signal Detection and Classification Based on Cyclostationarity for Cognitive Radios

Joint detection and automatic modulation classification (AMC) of very low SNR signals with relaxed a priori information has been of significant importance for cognitive radios as it enables the cognitive radio (CR) to react and adapt to the changes in its radio environment. This paper propose an algorithm based on first-order and second-order cyclostationarity for joint detection and classification of frequency shift keying (FSK), minimum shift keying (MSK), and pulse amplitude modulation (PAM) signals. The proposed algorithm has the advantage that it avoids the need for timing and frequency recovery, and estimation of signal and noise powers. Simulations are given to verify the performance of the algorithm.