Employing correlation coefficient and goodness of fit for blind spectrum sensing

he performance of the existing spectrum sensing algorithm based on goodness of fit (GOF) testing is only excellent when the primary signal is static during sensing interval, which is a very rare case in cognitive radio. For the use in the real Multi-antenna cognitive radio scenarios, a new spectrum sensing based goodness of fit in Multiantenna cognitive radio is proposed in this paper. Firstly, the correlation coefficients between each antenna are calculated; secondly, the author makes appropriate non-linear transformation of every correlation coefficient so as to obey Student’s t-distribution; finally, the Righttail Anderson-Darling (RAD) rule is adopted. Both analysis and numerical results show that the proposed method does not need any prior information and outperforms the AndersonDarling sensing method, the Ratio of Maximum eigenvalue to Minimum eigenvalue (MME) method, the Covariance absolute value (CAV) method and ED method with noise variance known in the same simulation environment.