Spectrum sensing based on MUB and goodness of fit test

Spectrum sensing is one of the most important technologies in cognitive radio. Based on the mutational unbiased bases (MUB) matrices and goodness of fit test, a novel spectrum sensing method based on the traditional Kolmogorov-Smirnov (KS) test is proposed in this paper. Specifically, by using MUB matrices to process the received samples, new samples with low correlation are generated, and thus, the number of samples is enlarged, and its sensing performance is improved. Simulation results show that when the number of received samples is limited and the noise is subject to Gaussian distribution, detection probability of the proposed method is higher than the traditional KS test method.