Research On Adaptive Dynamic Spectrum Allocation In Cognitive Radio Networks

The inefficient usage of the limited spectrum resources urges the investigators toreview the spectrum policy and start to seek for a novel communication technologywhich can exploit the wireless spectrum in a more intelligent and flexible way recently.The concept of cognitive radio has been receiving an increasing attention. It is proposedthat the existing available spectrum of primary users can be allocated to the secondusers who need it to communication. It can improve the spectrum efficiency andalleviate the shortage of spectrum resources. The system performance of the cognitiveradio network is depend on the effective spectrum resourse allocation. Some cognitiveradio spectrum allocation model and algorithm are proposed at home and abroad. Butthey do not consider that whether the spectrum allocation can ensure the QoS of usersand adapt the change of the communication environment. So to solve these problems,the following work can be considering:To solve the QoS optimization problem of the spectrum allocation in cognitiveradio network, a spectrum selection algorithm based on queueing game is proposed.Spectrum allocation is studied via potential game and queueing theory. The datatransmission problem is modeled as a potential game of spectrum selection. Consideringthe influence of delay caused by primary user, the utility function is built based onqueueing game. Each second user autonomously adjusts the spectrum choose strategy toget the spectrum allocation and cross-layer optimizes the transmission delay and theaverage packet loss rate. Simulations and analysis show that this proposed algorithmcan reduce average end-to-end delay and average packet loss rate effectively, convergeto strategy equilibrium point more quickly with smaller complexity.In order to avoid the waste created by the change of network environment afterstatic spectrum allocation, spectrum allocation based on trade mechanism is studied.Considering the change of the spectrum market capacity and the spectrum value forsecondary users, this algorithm modeled the spectrum trading as a dynamic Cournotgame. Depending on the demand and supply for spectrum trading, users adjust theleasing price and profit to maximize the utility and complete the spectrum allocation.The method is analyzed and the performance is compared with the algorithm aboutstatic market capacity model. Simulation results illustrate that the proposed algorithm has higher spectral efficiency and avoids the "secondary waste" of the spectrumallocation.