Optimizing Distributed Cooperative Sensing Performance Using Swarm Intelligence

Cognitive radio (CR) technology offers attractive solution to overcome spectrum scarcity problem and improve spectrum utilization. Opportunistic spectrum access concept adopted by a CR network allows unlicensed or cognitive users (CUs) to access a frequency band that is not in use by the licensed or primary users (PUs). In enabling such spectrum-aware communication protocol, spectrum sensing becomes the fundamental requirement to exploit the unused spectrum and effectively protect the quality of service (QoS) of PUs. An increase in sensing time results in a higher detection probability (

P

d

) and lower false alarm probability (

P

f

), which leads to better protection of the primary users from harmful interference in addition to improved utilization of the available unused spectrum. However, this will reduce the amount of time for data transmission and hence, affects the achievable throughput of CUs. In this paper, this sensing-throughput tradeoff is addressed in terms of distributed approach of cooperative channel sensing. In order to find the optimal sensing time that maximizes the throughput of an OSA based CR network, the decision fusion scheme of OR rule are subjected to particle swarm optimization (PSO). The results show that a lower sensing time is achieved under PSO based cooperative sensing. The maximum achievable normalized throughput also increases with the increase of sensing terminals under optimal fusion scheme compared to normal fusion technique.