Research Of Spectrum Allocation Algorithm In Cognitive Radio Network

The Research are supported by the National Natural Science Fund (NO. 60772110),Huawei Technology Fund and The level of field Fund of Beijing University of Posts and Telecommunications.With the rapid evolution of the wireless communications in the broadband, seamless, intelligent direction, we have to face one of the bottlenecks is the lack of spectrum resources. At present specific communications business fixed allocation for spectrum, the spectrum will often uneven distribution of resources, and even a waste of circumstances, this widespread concern with the current shortage of spectrum resources contradictory. Cognitive radio technology as a smart spectrum-sharing technologies to be effective in alleviating these contradictions, through sensing frequency domain, time domain and space domain such as space environment, automatically search and occupy free licensed spectrum, to address a new way on how to improve spectrum utilization of the wireless communications problems created in the limited spectrum resource conditions.This paper analysed research projects in the background at first, explained the definition and the function of cognitive radio briefly, explained cognitive radio key technology and typical applications more detailed. Next in the third, fourth and fifth chapters of this paper discussed in detail the major work completed. Three aspects of spectrum allocation in cognitive radio Network were studied, one is about spectrum allocation algorithm based on graph coloring of cognitive radio, one is about spectrum allocation algorithm based on rate requirement of cognitive radio and another is about spectrum allocation algorithm based on protecting licensed link of cognitive radio.At chapter 3, spectrum allocation algorithms applied to the actual network were researched according to channel weight problem.A model of spectrum allocation based on graph coloring was presented. Weighted Distributed Greedy Algorithm, Weighted Distributed Fairness Algorithm and Weighted Distributed Random Algorithm considering channel throughput weight were proposed. The simulation analysis shows that the three algorithms achieve good performance of throughput, fairness and complexity respectively.At chapter 4, the optimization of unoccupied spectrum allocation according to the needs of CR users was researched. A model of spectrum allocation based on congestion game was presented and a fast convergent spectrum allocation algorithm based on transmission rate requirement is proposed. The simulation analysis shows that the algorithm achieves fast convergence speed and optimization of spectrum allocation according to transmission rate requirement of Cognitive Radio subscriber.At chapter 5, spectrum allocation algorithm based on protecting licensed link in Cognitive Radio Network was researched. A spectrum allocation model based on protecting licensed link was proposed. The simulation analysis shows that the algorithm based the model achieves spectrum allocation optimization of cognitive radio link based on protecting licensed link; the simulation also presents the relation between interference of licensed link and SINR of cognitive radio link and proportional factor which provides the parameter to cognitive radio network with different performance requirement for model application.