In this paper, we propose a Swarm Intelligence based Resource Allocation Algorithm (SIRAA) for optimum allocation of available spectrum holes to cognitive radio users. Usage statistics reveal that large portion of the spectrum are under utilized due to stringent licensing policy. Spectrum utilization ranges from only 15% to 85% with a high variance in both temporal and spatial domain. Due to the inefficiency in the spectrum usage and increase in the access to the available limited spectrum in recent years, there is a need to access the existing wireless spectrum opportunistically. This new networking paradigm is referred as Cognitive radio networks. We used existing bidding strategy to allocate channels to cognitive radio users. A cognitive radio sends bid values (the bid value is a function of bandwidth requirement, bit error rate, possible interference, user type etc.) for every available spectrum holes to the base station. The base station allocates the spectrum band to the cognitive user by maximizing the totald bid value received from all cognitive users. We consider four different cases of allocation such that each cognitive user (i) will get exactly one channel (ii) will get channels in a proportionate manner (iii) will get at least k number of channels and (iv) will get at least k1 and at most k2 number of channels. We proposed a binary particle swarm optimization (PSO) technique to find an optimum allocation of channels to all cognitive users for all four cases. We compared our simulation results with traditional Hungarian method wherever the later is used to verify in some feasible cases.
展开▼
