Blind Spectrum Sensing in Cognitive Radio

Tao Cui,Jia Tang,Feifei Gao,Chintha Tellambura

DOI: https://doi.org/10.1109/wcnc.2010.5506471

2010-01-01

Abstract:In this paper, we consider an interesting and practical scenario for spectrum sensing in cognitive radio network, where both the signal power of the primary user and the noise variance are treated as unknowns before the detection. Knowing accurate noise variance and signal power is crucial in most sensing algorithms, e. g., energy detection. By exploiting the received signal structure, we propose blind spectrum sensing methods in the sense that both the signal power of the primary user and the noise variance are estimated, which is a non-trivial task before knowing the status of the primary user. Three different algorithms, direct estimator, approximate maximum likelihood (ML) estimator and pseudo linear minimum mean square error (MMSE) estimator, are proposed based on the moments of received signals. Simulation results confirm that the proposed algorithms can estimate the noise variance and the primary user's signal power with high accuracy.