The Research Of Spectrum Allocation Strategy In Cognitive Radio System

With the rapid development of radio communication technology, the scarceness of spectrum resource has hampered the continual development of wireless communication. However, recent researches show that the shortage of wireless spectrum is a relative term, since some spectrum has been implemented quite a lot of service, while other parts of spectrum have fewer efficiency of utility.Cognitive radio system can percept the external environment by terminal equipment, and look for "secondary use" of spectrum resources. It can communicate on a reliable basis at anytime and anywhere by changing real time operation parameters, such as transmit power, carrier frequency to keep the interference to the licensed user under a certain level. So how to allocate spectrum efficient and equitable among cognitive radios is a research focus.Firstly, the dissertation introduces an overview of cognitive radio technology, as well as the sharing model and principle in cognitive radio spectrum briefly.Based on this, the dissertation mainly proposed a new user satisfaction function, which takes transmission rate, transmission power and perceived contribution into account. And use it as the utility function, joint channel and power allocation algorithms are designed by matching game between cognitive user and spectrum.Secondly, we proposed an improved matching game algorithm based on POMDP and reinforcement learning model due to the cognitive radio spectrum allocation algorithm based on matching game is a static way in each time slot, without considering the time-varying characteristics of channels and users. Through the observation of historical information statistics, cognitive users can maximize the system revenue by adjusting the game strategy. Simulation results show that the algorithm can improve system performance of spectrum allocation significantly.