随着无线通信技术的蓬勃发展，多输入多输出(Multiple-Input Multiple-Output， MIMO)技术以其优异的性能受到广泛关注，该技术既可以提供分集增益提高链路可靠性，又可以提供复用增益提高系统传输速率，还可以实现系统各种性能的折中，在扩大网络覆盖范围、节约能耗、降低小区间干扰、提升系统吞吐量等多方面发挥了巨大作用。鉴于MIMO技术的诸多优势，该技术已经被广泛应用于各种无线通信系统当中。然而，MIMO技术需要在终端安装多幅天线，占用额外的空间和功耗，使得该技术在体积和功耗严格受限的小型终端中无法使用，严重限制了这项技术的应用范围。 协作分集技术正是为了解决这个问题而被提出的，该技术可以有效的利用无线信道的广播特性，用户通过彼此共享天线形成一个虚拟的多天线环境，在单天线用户上实现了MIMO技术，从而为MIMO技术的应用提供了全新的思路。从诞生之日起协作分集技术就受到了广泛的研究和关注，其优越的性能已经得到充分的证实，但是，该技术在分布式网络中的推行尚未得到很好的解决。这是因为分布式网络中的用户不受中心节点控制，每个用户都可以自主选择是否要参与协作，如果某个用户选择为其他用户转发信息，那么他需要消耗自己额外的资源(比如能量和带宽等)，而又无法保证会获得其他用户的协作传输。如果用户是理性的或者自私的，他将选择不协作。这个问题的存在严重制约了协作分集技术在分布式网络中的应用。针对上述问题，本论文使用博弈论对网络中用户的行为进行了深入分析，并以此为依据制定行之有效的协作激励机制，保证协作分集技术在分布式网络中得以顺利实施，同时提出有效的资源分配策略，使网络中的资源得到更高效的使用。概括来说，本文的主要工作和贡献如下： (1)针对有多个源节点和多个中继节点的无线分布式网络，研究如何将源节点与中继节点进行两两匹配来实现协作传输。首先引入定价机制来激励协作，当一个源节点与一个中继节点进行匹配时，源节点将获得分集增益，而中继节点获得虚拟货币，中继节点得到的虚拟货币可以在他有信息发送时激励其他节点协作传输；然后在综合考虑源节点和中继节点收益的基础上，提出了用双边一对一匹配博弈(联盟博弈的一个子类)模型对要讨论的问题进行建模。引入著名的延迟接收程序(deferred acceptance procedure)对匹配问题进行了求解，结果证明该匹配策略总是能找到模型的解，并且解在联盟博弈的核心(core)当中，这就意味着用延迟接收程序求得的匹配策略是稳定的，没有节点愿意单方面打破当前匹配状态。仿真结果显示，本文提出的分布式匹配策略具有线性时间复杂度，易于实现，但其性能逼近具有阶乘复杂度的最优匹配策略。 (2)针对分布式无线网络中自私用户不协作的问题，提出了一种基于拍卖的定价机制来激励协作，然后基于拍卖模型设计了一种有效的伙伴选择策略。在模型中，源节点和中继节点分别扮演竞价者和拍卖商的角色。因为源节点之间需要通过公平竞争来确定中继节点的协作资源归属，所以提出的伙伴选择策略是公平的。当一个拍卖过程完成时，赢得拍卖的源节点可以获得分集增益，而中继节点可以获得虚拟货币。两种最常见的拍卖类型，即第二价格拍卖和第一价格拍卖，都被引入了模型当中，并对之进行了深入分析。首先分析了单中继网络，推导了每个源节点的纳什均衡(Nash equilibrium，NE)解和每个中继节点的最佳底价。基于NE，进一步分析了源节点和中继节点的收益情况。然后从整个系统的角度出发，对多中继网络进行了讨论。在多中继网络中，每个源节点都可以通过向多个中继节点喊价的形式获得多中继协作传输。在预算受限的情况下，将源节点应该向哪些中继节点喊价的问题建模为线性整数0-1规划模型。对比了两种拍卖策略的性能，结果证明第一价格拍卖可以更有效的利用系统资源。 (3)在分布式无线网络当中，为了解决下面两个基本问题—什么时候协作和怎么样协作，提出了一种基于重复博弈模型的协作激励机制。在模型当中，考虑了两用户节点向两个不同目的节点传输信息的场景，并且认为系统是对称的，因此每个节点都既是源节点又是潜在的中继节点，并且都有决定是否协作的权利。用重复博弈模型对节点间的协作行为进行了建模，假设网络中的节点都是自私的，即节点以最大化收益为目标，其中收益是一个关于接收信噪比(Signal-to-Noise Ratio，SNR)的单调递增函数，理论分析和仿真结果显示，只要节点足够关心未来的数据传输性能，互相协作将是系统的NE解。 关键词：协作分集，博弈论，伙伴选择，纳什均衡，MIMO 论文类型：应用基础研究类

With the rapid development of wireless communication techniques, the Multiple-Input Multiple-Output (MIMO) technique has attracted much attention for its superior performance. The technique can provide diversity gain to increase the link reliability, and can also provide multiplexing gain to improve the system transmission rate. Furthermore, a trade-off between different functions of a system can also be achieved, and it plays an important role in terms of increasing network coverage, saving energy consumption, reducing interference between different cells, improving system throughput and so on. Given this, MIMO has been widely used in various wireless networks. However, as the technique needs to install multiple antennas on terminals, which means taking up extra space and power of the terminals, and this makes the technique cannot be implemented in the situation where the terminals’ size and power are strictly limited. Thus, the issue restricts the application of this technique. To solve this problem, cooperative diversity has been proposed. The technique can take advantage of the broadcast nature of wireless channels, such that users can share each other’s resources to create a virtual multi-antenna environment and thus realize the technique of MIMO. The technique provides a completely new idea for the applications of MIMO. In recent years, cooperative diversity has been drawing widely attention, and its superior performance has been fully confirmed. However, the technique cannot be properly implemented in distributed wireless networks. This is due to the fact that there is no centralized node in a distributed network, each user can choose whether to participate in cooperation by himself, if a user chooses to forward information for other users, he needs to consume extra resources (such as energy and bandwidth, etc.), and there is no guarantee that he can find a partner who is willing to relay his information. If a user is rational or selfish, he will choose not to cooperate. This problem seriously restricts the application of cooperative diversity in distributed wireless networks. To address the above issues, this paper introduces game theory to analyze the users’ behavior, and then presents effective cooperative incentive mechanisms to make sure the cooperative diversity technique can be achieved in the distributed network. At the same time, efficient resource allocation schemes are proposed, which makes the use of resources more efficiently. The main contribution of this dissertation can be summarized as follows. Consider a distributed wireless network with multiple source and multiple relay nodes, the problem of how to assign source-relay pairs to achieve cooperation is investigated. The pricing-based approach is introduced to stimulate cooperation. And when a source node is matched with a relay node, it will obtain diversity gain and the relay node can get virtual revenue, which can be used to stimulate cooperation when he has data to send. By jointly considering the benefits of the source node and the relay node, we prove that the source-relay assignment problem can be modeled by a two-sided one-to-one matching game, which is a branch of coalitional games. The deferred acceptance procedure is introduced to solve the matching problem, and it turns out that a solution can always be found and is proved in the core of the coalitional game. Consequently, each node is satisfied with its final state and has no incentive to deviate, which leads to a stable matching state. Simulation results demonstrate that the proposed matching scheme has linear time complexity, which means the scheme is easy to implemention. However, it can achieve comparable performance to that employing centralized optimal scheme in terms of total profit of the system. To address the noncooperation problems of selffish nodes in distributed cooperative wireless networks, an auction-based pricing scheme is introduced to stimulate cooperation. And then an efficent partner selection method is designed, in which sources act as bidders and relays act as auctioneers, and the sources compete to obtain the relay’s assistance with monetary incentives. Due to the fact that the sources should compete with each other to determine the recipient of the cooperative resources, the proposed scheme is an example of so-called competitive fairness. When an auction is implemented, the winning source will obtain diversity gains and the relay will get virtual currency. The two most prevalent auction forms, i.e., the second-price auction and first-price auction are both introduced and analyzed in this paper. A single-relay network is considered first, and the Nash equilibrium (NE) for each source and the optimal reserve price for the relay are characterized. Based on the equilibrium, the expected payoff of each source and the expected revenue of the relay are then derived. Furthermore, from the perspective of the whole system, the multiple-relay networks are discussed, in which a source can select multiple partners for cooperative transmission by bidding to several relays. It is proved that the issues that to which relays should each source bid can be modeled as linear 0-1 integer programming problems in both the second- and first-price auctions. We draw the conclusion that a source prefers the first-price auction to the second-price auction when its virtual currency is limited. In distributed wireless networks, to solve the following two basic problems, i.e., when to cooperate and how to cooperate, a cooperation strategy among rational nodes in a wireless cooperative relaying network is proposed. A symmetric system model comprising two users and two destination nodes is presented. In the model, each user plays an equal role and acts as a source as well as a potential relay and has the right to decide whether to cooperate. Cooperative communications is modeled as a repeated game in which the two participating terminals are selfish and seek to maximize their own payoff, a general utility function that monotonically increases with signal-to-noise ratio. Results show if the node care his future payoff , a Nash Equilibrium in which users mutually cooperate can be derived. Keywords: cooperative diversity, game theory, partner selection, Nash equilibrium, MIMO Type of Dissertation: Applied Basic Research