Performance analysis of coexisting secondary users in heterogeneous cognitive radio network

In this paper, we propose an analysis framework for the performance of heterogeneous cognitive radio networks with multiple different secondary users (SU) coexisting and competing for spectrum access. In order to model the operation of different SUs in a feasible manner, we use a bank of Markov models, where each SU is modeled by a separate Markov model. The individual Markov models are connected implicitly in their transitional probabilities. Expressions for the transitional probabilities are derived by analyzing the mutual interference among the SUs. The throughput of each SU and the overall CRN can thus be calculated. In addition, to set up a comparison basis for evaluating the optimality of the CRN throughput, we optimize the SU channel access by solving a sum-of-ratios linear fractional programming. Simulations are conducted to verify the proposed techniques and to compare the throughput of the CRN under random channel access over the throughput under the optimized channel access.