Analytical Evaluation of Saturation Throughput of a Cognitive WLAN Overlaid on a Time-Scheduled OFDMA Network

In this paper, we analyze the saturation throughput of a cognitive WLAN overlaid on a primary OFDMA TDD network (e.g., LTE or WiMAX). In this scenario, after the contention among the secondary nodes, the winner node transmits its data packet in the empty resource blocks (RBs) of downlink and uplink subframes of the primary network. Regarding the OFDMA structure as well as time-scheduled resources in the primary network, the time duration of opportunities for the secondary network does not follow simple exponential on-off pattern. To model the dynamic behavior of opportunities for secondary nodes as well as contentions to exploit the opportunities, we propose an analytical model comprised of a discrete-time Markov chain and two inter-related open multi-class queueing networks (QNs). The effects of the random number of empty resource blocks at different frames as the result of random amount of download and upload data, random packet transmission time at WLAN, the dependency of the number of empty RBs at consecutive frames, and the details of 802.11 MAC protocol are included in our analytical approach. We include different resource allocations in the primary network in our analysis. Simulation results confirm the accuracy of our analytical approach in different conditions.