A Two-Stage Detector for Cognitive Radio Networks with Correlated Multiple Antennas over Dynamic Channel Environment

In this paper, we consider a cognitive radio network with correlated multiple antennas, where the Rayleigh fading channel between PU and SU can be either slow or fast over one detection period. Under such a dynamic channel, most of the existing multi-antenna-based detectors cannot offer a reliable detection performance. To tackle this issue, a two-stage detector combined the eigenvalue moment ratio (EMR) and weighted-covariance-based detection methods is proposed. Numerical results demonstrate that our proposed two-stage method can effectively guarantee the performance when the channel environment is changed. Moreover, the computational complexity of the proposed two-stage detector is still kept comparable with EMR detector.