Enhanced Cyclostationary Detector Complexity Based on Haar Wavelet and Signed Correlator

The spectrum sensing function plays a significant role in the performance of cognitive radio (CR).Spectrum sensing specifies if free channels exist and identifies free channels for secondary users, actively helping in the improvement of spectrum usage and recognizing available channels in CR systems.Cyclostationary feature detection (CFD) is a spectrum sensing method that detects signals depending on different characteristics such as carrier frequency, modulation types, cyclic frequency, and symbol rates with an extremely low signal-to-noise ratio.At low SNR, CFD achieves a detection process with a high computation complexity.This paper designs Enhanced Cyclostationary Detector complexity with improved detection speed performances.For the sake of minimizing system complexity, utilizing the advantages of the Haar wavelet transform and signed correlator method for estimating the cyclic spectra of a detected signal.The proposed method performance was evaluated over Rayleigh flat fading and AWGN channel that had low SNR values.The acquired simulation results indicated the efficiency of the proposed method in terms of reduction 70% in complexity, 60% in time, and 7% in memory storage, with improved detection performance that is about 8% compared to conventional method at low SNR values reach to -30dB.