Performance analysis of cognitive MAC under saturation condition using Statistical Channel Allocation

An efficient cognitive MAC protocol should be able to predict future spectrum usage to enable dynamic spectrum access (DSA) to any continuous part of the spectrum without (or with minimal) interference to primary users. To fulfill such a goal, we present a new system model to evaluate the performance of cognitive networks in this paper. Based on this model, we conducted simulations using NS2 to show the performance including throughput, delay and loss-rate of C-MAC under saturation conditions, followed by the theoretical analysis. In this research, the model is designed using the CSMA/CA based algorithm and Statistical Channel Allocation algorithm (SCA-MAC) to do real-time opportunistic access to the spectrum.