A service-oriented spectrum allocation algorithm using enhanced PSO for cognitive wireless networks

Because of the increasing growth of spectrum utilization in wireless networks, dynamic spectrum allocation (DSA) using cognitive radio has recently attracted attention. DSA provides a promising communication paradigm to relieve the bottleneck that arises in the spectrum utilization of wireless networks. In a cognitive wireless network, deficient matching between user demands and spectrum resources can result in communication deterioration; however, superfluous matching can lead to inefficient spectrum utilization. In this study, a service-oriented, multi-attribute normalization model is developed that considers user demands and spectrum characteristics. Based on this model, a service-oriented DSA, which applies the graph coloring problem (GCP) and an enhanced particle swarm optimization (PSO) method, is proposed to efficiently satisfy user service demands and achieve better utilization and fairness rewards. The simulation results demonstrate that the algorithm satisfies the requirements for accurate spectrum allocations in cognitive wireless networks and performs better with respect to iteration efficiency and utilization rewards than the existing algorithms.