GLRT-based cooperative sensing in cognitive radio networks with partial CSI

This paper studies cooperative spectrum sensing in cognitive radio networks. Both the sensing and the reporting channels are assumed to be either slow or fast fading channels. In a practical system, due to the lack of cooperation between the primary and secondary users, the power of the primary signal, the channel state information of the sensing channels as well as the noise power level are unknown. We first provide an analytically tractable signal model using a Gaussian approximation and then propose generalized likelihood ratio test (GLRT) methods for the design of the corresponding detectors. The basic idea lies in the fact that the unknown parameters can be estimated by exploiting hidden information in the sample covariance matrix of the received signals. The effectiveness of the proposed GLRT-based sensing methods are validated through numerical results.