Cognitive Radio Based Key Technology Research In The Dynamic Spectrum Allocation

With the development of wireless communication, the requirement of spectrum is growing dramatically. Because of the utilization inefficiency of static spectrum allocation, both the academic and industrial field began to pay more attention to noval methods for spectrum management. Cognitive Radio, a newly emerging technology, has promoted the development in spectrum management. Moreover, Dynamic Spectrum Allocation (DSA) is considered to be essentially important to manage the hetetogeneous spectrum efficiently and dynamically. As the DSA could break the bottleneck of the limited wireless resource, it is of high theoriotic and practical importance.This thesis adopts mathematical analysis and data simulation to research the noval DSA mechanisms, QoS influence of resource reservation, the coexistence of primary network and the cognitive network and etc. The main innovative achievements of this thesis lie in the following aspects:A. Proposed an Auction based spectrum allocation mechanismThe thesis applies the Second Price Sealed Auction mechanism for spectrum allocation in CR network. Based on this, a VBG spectrum auction method is promoted for higher spectrum utilization and revenue for primary network. In order to obtain the optimal spectrum allocation in auction mechanism, a Phase-Optimal (PO) algorithm is developed, which is the combination of Greedy algorithm and Exhaustive method. PO algorithm could not only achieve a sub-optimal solution, but also equip with low computation complexity.B. Developed a Discrete Stochastic Optimization based DSA mechanismAs in the existing works on spectrum management in CR networks, the channel condition is considered as perfect information or static. In this thesis, I combine the channel estimation in channel allocation. And the traditional optimization problem becomes a Discrete Stochastic Optimization (DSO) problem. Moreover, I develop an Aggressive Discrete Stochastic Approximation (ADSA) method to solve this problem. ADSA could efficiently follow the time-varing environment in spectrum allocation optimization.C. Discussed the performance enhancement mechanism for second userIn traditional CR network, in order to guarantee the performance of primary user (PU), the QoS of second user (SU) is considered poorly. Combined with the newly emergying techniques, such as:spectrum handoff and the cooperation between PU and SU, I develop a channel reservation mechanism, which allows the system to subscribe a paritial of available channels in common spectrum pool for SU's exclusive usage. This mechanism enhances the SU's QoS with only acceptable interference to PU. This part of the thesis is based on the Markov Chain mathematical description. What's more, it gives the optimum number of reserved channels.Important issues of different levels in CR networks'spectrum allocation are discussed in this thesis to improve the spectrum efficiency and regain the ignored QoS for SU. Combining spectrum allocation and other advanced communication technology for further enhancement of spectrum utility is what I will work on in the future.