Throughput analysis of cognitive wireless networks with Poisson distributed nodes based on location information

Abstract

This paper provides a statistical characterization of the individual achievable rates in bits/s/Hz and the spatial throughput of bipolar Poisson wireless networks in bits/s/Hz/m2. We assume that all cognitive transmitters know the distance to their receiver’s closest interferers and use this side-information to autonomously tune their coding rates to avoid outage events for each spatial realization. Considering that the closest interferer approximates the aggregate interference of all transmitters treated as noise, we derive closed-form expressions for the probability density function of the achievable rates under two decoding rules: treating interference as noise, and jointly detecting the strongest interfering signals treating the others as noise. Based on these rules and the bipolar model, we approximate the expected maximum spatial throughput, showing the best performance of the latter decoding rule. These results are also compared to the reference scenario where the transmitters do not have cognitive ability, coding their messages at predetermined rates that are chosen to optimize the expected spatial throughput – regardless of particular realizations – which yields outages. We prove that, when the same decoding rule and network density are considered, the cognitive spatial throughput always outperforms the other option