Performance evaluation of spectrum sensing based on wavelet entropy for cognitive radio networks

As wireless communication services grow quickly, the seriousness of spectrum scarcity has been on the rise gradually. Cognitive radio (CR) is viewed as a novel approach for improving the utilization of the radio electromagnetic spectrum. Among its fundamental functions, the most important function is spectrum sensing. This paper develops spectrum sensing based on wavelet entropy(WE) in cognitive radio networks. Then we compared with Spectrum Sensing based on wavelet packet entropy(WPE). According to the wavelet transform theory, we proposed spectrum sensing approach which has two advantages: low computational complexity and less response time. When the detected signal is concentrated in the low frequency band, our detection method is cost-effective. Simulation results show that the presented spectrum sensing algorithm is robust against noise uncertainty. Furthermore, the simulation results also indicate that the detection probability of the two spectrum sensing approaches is almost same to be close to 1, when the SNR is greater than -6dB.