The effects of channel knowledge on cooperative spectrum sensing in Nakagami-n/q fading channels

Abstract

Cooperative spectrum sensing (CSS) is an efficient method to detect vacant spectrum of a primary user (PU) by combining sensing information of multiple cognitive radios (CRs) in presence of fading. In this paper, effects of perfect and imperfect channel state information (CSI) on CSS network is evaluated. The proposed network is operated over Nakagami-q (Hoyt) and Nakagami-n (Rician) fading affecting both reporting (R) and sensing (S) channels. The CRs which employ energy detectors (EDs) are selected on the basis of CSI between them and a fusion center (FC). The knowledge on the quality of R-channels is estimated at FC using a channel estimator. The estimated CSI is either perfect when there is no error in the estimator, or imperfect when errors are present. Accordingly, CRs are selected under both perfect CSI and imperfect CSI cases. All CRs use the decision statistics obtained by EDs and make one-bit binary decisions about the availability of a PU. Selected CRs only transmit decision information to the FC. The miss detection probability and error rate of network under two cases of selection are evaluated by operating majority and maximal ratio combining rules at the FC. Performance is analyzed for different channel and network parameters and comparison between fusion rules for different fading channels is also highlighted.