Wavelet transform for spectrum sensing in Cognitive Radio networks

Spectrum sensing is a critical component for Cognitive Radio technologies. In traditional spectrum sensing algorithms, certain degree of the prior knowledge such as the level of the white noises is assumed. In practice, however, such information is dynamic and its maintenance and distribution can be costly. In order to conduct the efficient and accurate spectrum sensing without any prior knowledge, in this paper we propose a novel wavelet transform based spectrum sensing algorithm called WATRAB. The basic idea is that the signals of Primary Users (PU) signals only carry a limited amount of information, while the noise can be considered as having many. By carefully selecting a wavelet transform method, such a difference can be exploited to generate very different transform results. To demonstrate the effectiveness, comprehensive simulation experiments are conducted. Experimental results show that WATRAB can significantly improve the sensing accuracy by up to 14 db with the same accuracy compared with the simple energy detection, and the compared with the cyclostationary method, the complexity reduced from O(n²) to O(n).