Wavelet packet entropy based spectrum sensing in cognitive radio

Cognitive radio (CR) is viewed as a promising approach to improving the utilization of the radio spectrum. It is a wireless communication technology, which has the capability of sensing and utilizing the underutilized spectrum resources. Recognizing idle spectrum efficiently is a key function for cognitive radio (CR) and requires high precision and fast signal processing technique. Sensitivity to noise uncertainty is limitation of current spectrum sensing strategies in CR networks. In this paper, a wavelet packet entropy based algorithm for detection of primary users in cognitive radio networks is proposed. In proposed algorithm, the presence of primary user can be examined by comparing the estimated wavelet packet entropy of signal to a threshold. Simulation results show that the proposed algorithm is robust against noise uncertainty; the normalized wavelet packet energy vector which carries important information on the frequency locations of subbands can estimate spectrum holes in the signal spectrum with simple structure and low computational complexity comparing to the conventional schemes. In addition, the proposed method does not need any prior knowledge of primary signal and noise energy level.