No abstract

TV white spaces (TVWS) can be utilized by Secondary Users (SUs) equipped with cognitive radio functionality on the condition that they do not cause harmful interference to Primary Users (PUs). Optimization of power allocation is necessary when there is a high density of secondary users in a network in order to reduce the level of interference among SUs and to protect PUs against harmful interference. Grey Wolf Optimizer (GWO) is relatively recent population based metaheuristic algorithm that has shown superior performance compared to other population based metaheuristic algorithms. Recent trend has been to hybridize population based metaheuristic algorithms in order to avoid the problem of getting trapped in a local optimum. This paper presents the design and analysis of performance of a hybrid grey wolf optimizer and Firefly Algorithm (FA) with Particle Swarm Optimization operators for optimization of power allocation in TVWS network power allocation as a continuous optimization problem. Matlab was used for simulation. The hybrid of GWO, FA and PSO (HFAGWOPSO) reduces sum power by 81.42% compared to GWO and improves sum throughput by 16.41% when compared to GWO. Simulation results also show that the algorithm has better convergence rate.

TV white spaces (TVWS) can be used by Secondary Users (SUs) through Dynamic Spectrum Access (DSA) as long as they do not cause harmful interference to Primary Users (PUs). Due to spectrum scarcity, there is increasing demand for DSA. When there is a high density of SUs in a TVWS network such as cellular access to TVWS, problem of interference among SUs will arise. Possibility of harmful interference to PUs may also arise. Power and spectrum allocation optimization is therefore necessary to reduce the level of interference among SUs and to protect PUs against harmful interference. In this paper different hybrid firefly algorithms with particle swarm optimization and genetic algorithm for optimization of spectrum allocation in a TVWS network as a discrete optimization problem and that of power allocation as a continuous optimization problem are compared. Simulation was done using Matlab. Simulation results show that hybrid firefly algorithm with genetic algorithm outperforms other hybrid firefly algorithms for spectrum allocation. On the other hand, hybrid firefly algorithm with genetic algorithm and particle swarm optimization outperforms all other algorithms for power allocation as continuous optimization problem

No abstract

No abstract