Bandwidth Utilization Optimization on Agent Based CR Networks

Cognitive radio (CR) increases spectrum efficiency by allowing secondary users (SUs) to dynamically use the idle spectrum owned by primary users (PUs). This paper studies bandwidth allocation of SUs for throughput efficiency. Taking on consideration a scenario of five nodes on which two of them are Primary Users (PUs) and the other three are secondary users (SUs). What amount of bandwidth should be leased to the secondary users? In this paper, we study the optimal bandwidth allocation for multiple SUs. Our approach addresses the transmission parameters both in Primary Users (PUs) and Secondary Users (SUs). We first analyze the amount of bandwidth used by the Primary Users (Pus) and design an algorithm that will distribute the remaining bandwidth and allocate the optimal bandwidth an SU should use to maximize the per-SU throughput in the long term. We adjust the Secondary users TCP parameters so as it should not use more than the allocated Bandwidth to improve SUs' performance and also propose the algorithm for bandwidth sharing among the Pus and SUs more efficient while taking on account for the dynamics of the SUs on the idle bandwidth exploitation. We use the NS2 cognitive radio simulations to validate our results. The work contribute and steer the initiatives of the researches on finding the adequate bandwidth sharing algorithm in cognitive radio networks.