Study of eigenvalues detection based multiantenna blind spectrum sensing algorithm

In the multiantenna sensing scenarios, the sensing performance of the classical ED method can be degraded drastically because both the noise uncertainty and the correlation between the signal samples may be present simultaneously. Using the correlation characteristics of the multiple antenna received signal, a blind algorithm based on all the eigenvalues of the sample covariance matrix (SCM) is proposed. The new method can execute spectrum sensing without information about the noise variance, the primary signal and the wireless channel. Compared with the ED method, the sensing performance of the proposed method is robust to noise uncertainty because it does not need noise variance to help the sensing node to make a right decision. The multivariate statistical theory and the random matrix theory (RMT) are used to obtain the theoretical decision threshold. Simulation results show that the proposed algorithm has better false alarm performance and more reliable detection performance than the ED method when there exists noise uncertainty.