Research On QoS-based Spectrum Management In Cognitive Radio Ad Hoc Networks

Cognitive radio(CR) is a novel technology that allows the secondary users(SUs)opportunistically use the available spectrum resource while keeping licensed or primary users(PUs). As an important branch of CR, the characterstics of infrastructureless, rapid networking and flexible to carry out of cognitive radio ad hoc networks(CRAHNs)make it adaptable and survivable to special needs of wireless communications. However,it brings new challenges for CRAHNs research.With the exception of infrastructureless and frequent topology changes in the classic Ad Hoc network, CRAHNs has its own characteristic such as changes of spectrum availability and protection of PU activities. The characteristics make the spectrum management one of the most important technical keys in CRAHNs. It means that the multi-hop style should be adopted and each link of the network is unstable. This dissertation will mainly focus on the spectrum management scheme over CRAHNs and carry out from spectrum and power allocation, multi-hop cognitive route design and dynamic control channel allocation to make the Qo S provision scheme be able to react with a dynamic environment so as to improve traffic transmission efficiency and SU’s Qo S.Firstly, this dissertation will analyze and expatiate on the cross-layer scheduling design, dynamic control channel allocation and multi-hop cognitive route in CRAHNs.With an introduction on representative work procedures from different points of view,the achievements and simulation results will go into details. The design of dynamic environment-oriented spectrum management scheme has been considered, and the issues of Qo S provision as well as the relationship between network environment and algorithms devising will be presented, based on which the ideological line of this dissertation will be mapped out.Secondly, the cognitive cross-layer scheduling design for mixed multimedia transmission will be addressed for orthogonal frequency-division multiplexing CRAHNs. To avoid a heavy control information exchange on common control channel(CCC), the full Bayesian model will be devised to update the statistical distribution of subcarrier selection strategy profiles of SU. The target SU will adjust its own subcarrier selection strategy profile according to such a statistical distribution of other SUs’ and it’s also used to formate the delay-based utility function. Finally, by maximizing the delay and throughput-based utility function of each SU, the cognitive cross-layer scheduling design generates the optimal subcarrier selection, power and modulation allocation for each service packet so as to improve traffic transmittingefficiency.Moreover, the dynamic control channel allocation for CRAHNs will be presented.CR nodes find their local stable neighbors according to the spectrum availability during neighbor discovery phase. The problem of dynamically assigning the control channel is formulated as a maximum edge biclique graph problem, for which the algorithm provides different tradeoffs between two con?icting factors: number of commo n channels in a group and the group size. Then, an optimal control channel selection approach based on its utility function, which consists of control information delay and throughput, will be adopted for each neighboring group to choose the best qualit y control channel to exchange control information. The neighboring group-based approach guaranteed to have a desirable number of common channels for control, which facilitates for graceful channel migration when the primary radio(PR) activity is detected, without the need for frequent re-selection.Finally, the issue of cognitive multi-hop route protocol in high-mobility CRAHNs will be studied. Impacts of frequent topology changes caused by SUs’ mobility and spectrum availability influenced by PU activities on the cognitive route will be analyzed in detail. A topology management scheme based on a clustering model will be devised to simplify the topology structure of CRAHNs, and the cluster heads(CH) are selected based on proposed metrics and take the responsibility of routing formation and data communication. Besides, considering the spectrum availability changes according to PU activities, a novel routing metric, named cognitive transmission throughput(CTT), is proposed for route path selection. By maximizing CTT, optimal data channel for routing and relay CH can be jointly deserved. By combination of topology management,optimal spectrum and relay CH selection, the route path will be set and adjusted dynamically according to network environment so as to give a better provision of Qo S.And the route discovery mechanism is also given in this sect ion.