Speaker: Bhasker Dappuri (EE10D015)
Cognitive Radio Networks (CRN) are introduced to solve the spectrum scarcity problem and to improve the utilization of licensed channel. In this talk we propose a mechanism by which IEEE 802.11 DCF can be adapted by unlicensed secondary user (SU) in a Cognitive WLAN to opportunistically access and transmit data on the licensed channel owned by the primary user (PU). We modify the Markov chain model proposed by Bianchi for the cognitive WLAN. Our model handles both the cases of channel becoming busy due to the licensed user and also the collisions faced by the packets of SU due to the transmission by PU. By solving our model we derive saturated throughput, delay, channel busy probability, conditional collision probability of SU and interference to PU. We analyze the energy efficiency of the cognitive WLAN MAC protocol. The talk further dwells into the effect of number of SUs, spectrum availability and packet length of SUs on the above mentioned metrics. We validate our model by comparing the analytical results with simulation. The talk concludes with the derivation of the optimal channel sensing interval of the SUs and the corresponding sensing energy.