Automatic channel selection for cognitive radio systems

In this paper, we propose a novel channel selection algorithm for a cognitive radio system that coexists with legacy wireless LAN systems operating in the ISM bands. The cognitive system is capable of scanning its environment to measure ISM channel occupancy in terms of WLAN channel utilization. The system takes the opportunity of spectrum availability to facilitate its communication to the cognitive radio users. Different users experience different levels of WLAN activities in their surroundings which also vary across channels. Based on measured utilization, the proposed algorithm selects a channel considering cognitive radio users' QoS requirements. The algorithm is flexible to achieve different degrees of trade-off between the system throughput and users' fairness. Simulation results show that the proposed scheme can achieve up to 30% higher throughput while ensuring a higher level of fairness compared to a scheme where channel selection is random.