Research On Cooperative Resource Allocation For Distributed Wireless Networks

The next generation wireless communication network (NGWCN) is evolving towards a wideband, heterogeneous, flat and universal full-IP direction. Distributed wireless network will become an important part in the future mixed networks because of its special characteristics such as cheap cost, rapid deployment, strong resistibility and so on. In order to resolve the contradiction between scarce wireless resources including spectral and energy and fast increasing requirements of high-speed multimedia, data and saving energy, radio resource management techniques as efficient as possible are taken as one of the key research topics. Furthermore, cooperative communication can convert wireless multipath effects into favored factors for communication, and exploit spatial diversity gains by using relays, which makes higher traffic transmission efficiency and higher reliability possible. Actually, it draws widely attention nowadays.Holding such background, it is necessary to obtain the tremendous gains of resource optimization from cooperative communication technique in distributed wireless networks. This thesis will focus on this topic with the mathematic help of academic modeling, queue theory and so on. From performance analysis of distributed cooperative transmission system, cooperative medium access and resource allocation based on NUM, a series of rational resource allocation schemes are proposed jointly with error control, power control and relay selection scheduling. The major contributions of this thesis are as the following:(1) Considering resource utilizing efficiency, performance parameters for opportunistic cooperative transmission system are well analyzed over independent but not identically distributed Nakagami fading channels. Exact closed-form performance expressions are derived and compared for the end-to-end average symbol error probability (ASEP) of both partial and full CSI relaying schemes. The reason of performance degration caused by distributed condition is obtained, and exact closed-form expression in terms of outage probability of selection cooperation system with partial CSI is obtained too. Opportunistic cooperative transmission system is improved by error control in data link layer. Based on this, a Markov performance analytical model is established for this cross-layer cooperative framework. The obtained model is further applyed into multi-relay system, resulting in a optimal power allocation algorithm proposed. The correctness and validity are verified by Monte Carlo simulations, which has an important theoretical guidance significance to the following research of cooperative resource allocation algorithms in distributed wireless networks. The research productions are published in IEEE Transaction on Vehicular Technology, etc.(2) According to medium access problem, a series of MAC algorithms are proposed with different aims. An improved backoff algorithm based on the tradeoff of fairness and efficiency is proposed for IEEE802.11. Holding such idea, a distributed MAC scheme supporting real-time traffic with power control is proposed by integrating saving energy, increasing efficiency of channel utilisation and fairness, resolving hidden-terminal and exposed-terminal problems, improving network throughput and so on. An energy consumption model is proposed for distributed wireless network. Besides, the key influence factors of energy consumption are fully analyzed and possible improved schemes are discussed based on classical MAC protocols. Holding obtained conclusions, an energy-aware distributed MAC algorithm is presented. Finally, a cooperative MAC algorithm with multiuser diversity is proposed through network throughput maximization framework based on random access probabilistic model instead of traditional determinate interference avoidance model. To increase efficiency of spectral utilisation, a virtual link scheduling policy is presented for the primal links and multi-relay virtual cooperative links. Theoretical analysis and experiment results show that a series of proposed algorithms are valid, and the research productions are published in IET Communications, etc.(3) Finally, according to the whole system, a series of cooperative resource allocation schemes are proposed based on NUM. A power control game theoretic approach is presented with QoS constraints. The influences of MAI with variations of channel condition on throughput and delay are fully analyzed. Power resources are allocated efficiently under QoS requirements with the obtained probability of successfully packet transmitting and delay. Extending this idea into energy-constrained cooperative networks, based on double Oligarch Game, an adaptive power pricing scheme is proposed by fair power marketing. The aims of improving fairness, reliability and aggregate throughput are achieved by novel opportunistic relay selection and power allocation method. Besides, a utility-optimal cross-layer cooperative resource allocation algorithm is proposed. Via dual function, relaxing, and subgradient method, system problem is solved by decomposing functional modules into differrnt layers of protocol stack and using their interaction. Using a concept of pseudochannel gain, resource allocation algorithm of primal logical links is obtained by employing reverse-engineering. The research productions are published in Multimedia Tools and Applications, etc.In summary, this thesis conducts a research deeply into cooperative resource optimization in distributed wireless networks. The results of this research work will provide partial theoretical evidences and technical bases for the development of future wireless communication networks.