A new nonlinear technique for spectrum sensing in the presence of impulsive noise

In this paper a new spectrum sensing technique for multiple antenna cognitive radio network is proposed. While a large majority of the existing literature assume that the signal is corrupted by additive white Gaussian noise, this assumption is not always true in practice. In the present work, slow Rayleigh fading channel in the presence of impulsive noise is considered and BPSK modulation is employed. We use the enhanced energy detector in each antenna branch and exploit the combining with order statistics (COS) technique for detection. Simulation results show that our proposed technique has better performance than generalized likelihood ratio test (GLRT) detector.