Spectrum sensing algorithm based on sample variance in multi-antenna cognitive radio systems

The most significant feature that cognitive radio systems must include in order to operate properly is spectrum sensing. In this paper, we investigate the problem of designing accurate and efficient spectrum sensing algorithms in multi-antenna cognitive radio systems. Existing algorithms require an excessive number of sampling points to achieve the desired detection performance, and we address this issue by proposing a spectrum sensing algorithm based on sample variance that requires significantly fewer sampling points in multiple input multiple output (MIMO) scenarios. The proposed method uses the row vector of the sampling covariance matrix as the sample, and then uses the sample variance to construct the detection statistics. The judgment threshold is derived according to the false alarm probability. This algorithm makes full use of the relational structure of the signal, which allows it to reduce the number of sampling points while simultaneously enhancing the detection ability. Compared with existing algorithms, the proposed algorithm is more accurate and efficient. The superior performance of the proposed algorithm is demonstrated by Monte Carlo simulations in both Rayleigh and additive white Gaussian noise (AWGN) channels. These simulation results also show that the proposed algorithm has wider applicability than existing algorithms.