Investigation Of Channel Selection And User Cooperation In Cognitive Radio Networks

Since the inefficiency and inflexibility of the current spectrum allocation strategies, the industry and academic are researching on dynamic spectrum access based on cognitive radio for improving the spectrum efficiency. Now, the cognition user based on cognitive radio technology accesses the vacancy spectrum dynamically via periodic detection on licensed spectrum. But due to the randomness of the appearance of idle resource, the spectrum sensing will have a low efficiency and the transmission quality can not be guaranteed if the cognition user selects the target frequency bands in a random manner, especially for the cognition user who has requirements on data rate and delay. Moreover, frequent spectrum detection and handover will consume the effective transmission time, which will limit the improvement on spectrum efficiency. Therefore, how to introduce machine learning into the procedure of spectrum sensing, accessing and handover of cognition user for achieving intelligent channel selection, which can improve the spectrum sensing efficiency and guanrantee the transmission quality, is a hot research topic in cognitive radio networks now.This paper researches on the channel selection extensively based on traffic statistic learning and prediction and multi-user cooperation separately. This paper proposes the traffic prediction method and channel selection criterion for the channel selection scheme based on traffic statistic and learning as well as the cooperation method and information processing algorithm for the channel selction scheme based on multi-user cooperation. At the end of this paper, the research on channel selection based on user cooperation is expanded to the user cooperation transmission. The corresponding cooperative transmission protocol, user stability region and throughput models are provided.Firstly, this paper gives a generic function framework supporting traffic cognition for the selective spectrum sensing and accessing method based on traffic statistic learning and prediction. The theoretical analysis and performance evaluation models of the channel selection scheme based on traffic statistic learning and prediction are provided in terms of packet loss ration and throughput. These models can suppor the performance evaluation of diverse traffic prediction algorithms. Meanwhile, the precognition scenario is considered in the research and the corresponding mathematical models are proposed for providing the performance upper bound and performance comparison between different traffic methods.Secondly, this paper proposes a user cooperation channel selection method and the corresponding Bayesian information fusion algorithm for interence of the target channel features by learning from the user cooperative information based on the research on the user traffic statistic learning and prediction assisted channel selection. The mathematical description of the updating methods of the target values and uncertainty are provided. In the performance evaluation of channel selection methods, this paper gives the caluation model of delay mathematically and the effects of the cooperation information distribution and updating probability factor on the performance of user cooperation channel selection are evaluated via simulation.Finally, this paper expands user channel selection to user cooperative transmission and proposes the user opportunistic cooperation relaying protocol. This protocol utilizes the user idle time-slot to relay partner's data opportunistically for improving the transmission reliability and providing diversity gain. Meanwhile, this protocol can support the cooperation between cognition users and that between primary user and cognition user as well. Different with the existing cooperative diversity research foucusing on physical layer gain, this paper analyzes the performance vairat of higher network layers in terms of queue stability regin and throughput for two user coopeation pair under fixed and arbitrary time-slot allocation ratio when user cooperation is being applied.