Spectrum sensing based on the singular value of Gaussian random matrix and performance analysis

In this paper, a novel spectrum sensing algorithm called Gaussian singular value (GSV) algorithm is proposed. The method uses the ratio of the largest and the smallest singular values of the received signal matrix to detect the presence or absence of the primary signal. More importantly, GSV algorithm is improved to satisfy the requirements of MIMO situation. In addition, a measure called the credibility of detection (COD) is put forward to evaluate an algorithm's performance comprehensively, and both the probability of detection and false alarm are considered. Simulation results show that the GSV algorithm overcomes the noise uncertainty difficulty while keeping the advantages of the energy detection. The computational results of COD verify its credibility and effectiveness.