Research On Spectrum Sensing Scheme And Spectrum Resources Allocation In Cognitive Radio Network

The contradiction between the rapidly developing requirements of wireless communication technologies and the increasing scarcity of spectrum resources is bound to make cognitive radio become the key technologies for the future wireless communication networks. Therefore, it has very broad research and application prospects. However, cognitive radio development in the past ten years, as opposed to the more mature of the spectrum sensing algorithm, the spectrum scheme are rare. As cognitive radio environments are unique, relying simply on spectrum sensing algorithm in the physical layer can not achieve the objective of improving spectrum utilization, there must be a corresponding spectrum sensing scheme to effectively guide the user on the spectrum sensing. Also after the spectrum holes were found, how to assign the resources to users reasonably and reliably is the key issue for cognitive radio technology whether to realize the goals of improving the spectrum efficiency and avoiding interference on the primary users.In the context of Cognitive Radio Network (CRN), this dissertation investigated Channel Usage Strategy (CUS) in spectrum sensing scheme and spectrum resources allocation issues. In an aspect of CUS, the impact to channel selection of cognitive users from the uncertainty of real spectrum environment and channel fading stats is mainly consideration; while as to spectrum resources allocation, the selfishness of users including primary users and secondary users in CRN is entirely taken into account to ensure that the allocation is rational, reliability and fairness. The main contributions are expressed as follows:1. The uncertainty of detection of channel status in cognitive network environments is studied. As the cognitive users can not obtain the accurate channel state information by spectrum sensing algorithms, they may interfere with primary users while accessing the channels according to the false information and not complete their transmission tasks in the selected channel. To address this issue, a novel multi-radio multi-channel channel state prediction algorithm is proposed based on the theory of Partially Observable Markov Decision Process (POMDP) model. In this algorithm, the uncertainty of cognitive environment is modeled by POMDP and the Multi-Radio technique to avoid interference with the primary users is combined to design the spectrum sensing process for cognitive users. Hence, cognitive users can deduce the maximum possibility of channel state in next time slot and re-sort the channel index for cognitive users in terms of the result.2. The limited power of cognitive users is considered. Due to scan spectrum possibly in the larger scope and limitation of hardware condition, mobility and other factors make the cognitive users must consider the energy problem. To improve the sensing efficiency to save energy, based on the first part of the study and the combination of Finite State Markov Channel (FSMC) model, a POMDP-based multi-radio channel search scheme is further proposed. In the scheme, POMDP model is applied to deduce the channel state consist of idle-busy state and fading level in the next sensing slot, and cognitive users can use the result to re-sort the channel sensing index to achieve the goal of reducing energy consumption.3. Generally, there are multiple cognitive users in a CRN. Hence, how to guarantee the validity and fairness of spectrum resources allocation is the key issue for cognitive radio technology. To prevent the selfish users providing false information to deceive other users to maximize their own profits, a novel two stage spectrum resources allocation strategy based on Vickrey-Clarke-Groves (VCG) mechanism and priority of users in CRN is proposed. This strategy can avoid the unfairness of allocation by encouraging users to provide true information and thereby enhancing the rational spectrum resources allocation and reliability.4. With the reduction in hardware costs, one cognitive user may be equipped with multiple radio transceiver devices to simultaneously sensing and access. Nevertheless, in the face of multiple available channels, the traditional method of channel assignment is clearly not applicable. For this problem, the tool of non-cooperative game theory is used to analyze and prove the necessary condition for Nash equilibrium to multi-radio multi-channel assignment under a single conflict domain and the fairness of the assignment. A novel fairness-based channel allocation algorithm in multi-radio multi-channel environment is designed to ensure the effectiveness and fairness of spectrum resources allocation.