Optimal Performance of Cognitive Random Access Networks With Multi-Packet Reception

Abstract

This letter considers the performance of secondary networks using an Aloha-type random access protocol with multi-packet reception (MPR) for interference-limited communication. To simplify the analysis, it is usually assumed a specific backoff scheme, geometric retransmission for instance, for cognitive access in previous studies. Consequently, how to achieve the optimal performance of cognitive networks with a general backoff scheme has not been fully explored. The difficulty lies in a proper modeling of the transmission behavior of each secondary node with the interference constraint imposed by the primary user. In this letter, a markov chain is established to characterize all the states of each secondary transmitter. The markov chain enables a unified analysis of both unsaturated and saturated secondary networks, and derivations of explicit expressions of the throughput and the mean access delay. The effect of various factors, such as the interference constraint and the MPR capability of the secondary networks, is further characterized on the optimal network performance.