AM-signal detection in cognitive radios using first-order cyclostationarity

Cognitive radio is regarded as a novel approach for improving the utilization of precious radio spectrum resource. The detection of very low signal-to-noise ratio (SNR) signals with relaxed a priori information on the signal parameters is of high importance to such radios. This paper proposes a detection algorithm based on the first-order cyclostationarity for amplitude modulated (AM) signals that only requires rough information on the signal bandwidth and carrier frequency. A theoretical asymptotic analysis is performed. Simulation results show that the proposed algorithm performs well at low SNRs.