Outage analysis of buffer-aided underlay cognitive relay networks with outdated CSI

This work investigates the impact of outdated channel state information (CSI) on the outage performance of decode-and-forward (DF) buffer-aided underlay cognitive relay networks (CRNs). By modeling the transition of the buffer states with a theoretical framework of Markov chain, we derive a closed-form expression of the outage probability of CRNs with finite-size buffers. In addition, we also provide the asymptotic outage analysis in the high signal-to-noise ratio (SNR) regime, based on which the coding gain and diversity gain of the system with both outdated and perfect CSI are investigated, respectively. We find that they are determined by the number of the relays (M), the size of the buffers (L), and the feedback delay (ρ). Specifically, due to the feedback delay, the diversity gain of the system degenerates from M into 1 with small-size buffers, and degenerates from 2M into 2 with large-size buffers. Meanwhile, the coding gain of the system also differs in the case of small-size and large-size buffers. Numerical results are provided for the validity of the theoretical analysis.