状态无关主动队列管理算法博弈的Nash均衡

状态无关的主动队列管理算法处理分组时不区分分组所在的流的信息,因此在Internet中,它易于设计和部署。文中通过数学分析和仿真方法研究AQM博弈和Nash均衡存在性。假设业务流是Poisson分布的且用户可自由修改发送速率,因而有结论:Drop Tail、RED不能获得Nash均衡,CHOKe可以获得近似Nash均衡。依据判定条件,推导出一种与状态无关且具有效率的Nash均衡AQM算法。     

基于博弈论的P2P系统分析

P2P系统近几年来在网络上有飞速的发展,然而随着用户对网络和P2P系统的了解,使得P2P系统中大量存在欺骗行为,从而导致效率下降。文中以解决P2P系统中的欺骗为目的,通过引入博弈论思想从宏观和微观两方面对P2P系统进行分析:宏观上通过建立数学模型分析P2P系统效率下降甚至崩溃的原因;微观上,通过合理的论证,提出一种以牙还牙机制对P2P系统的欺骗问题进行解决。     

网络路由研究中的博弈论思想

对调整路由过程中存在的一个非合作非零和的动态博弈模型进行了深入的探讨，使之达到一个均衡，使得任何用户不会单方面改变路由策略从而获得更好的性能。分析了两种交互过程以及当前在自私路由中求解均衡的建模方法，对路由研究中的博弈论思想做了进一步的探讨，从而得到求解均衡的一般方法。     

Linear quadratic Nash game-based tracker for multiparameter singularly perturbed sampled-data systems: digital redesign approach

In this paper, a linear quadratic Nash game-based tracker for multiparameter singularly perturbed sample-data systems is developed. A generalized cross-coupled multiparameter algebraic Riccati equation (GCMARE) with two quadratic cost functions is solved by applying the LQR design methodology for the optimal tracker design. Firstly, the asymptotic expansions of the GCMARE are newly established, and the proposed algorithm is able to effectively solve the GCMARE with the quadratic convergence rate. Then, the low-gain digital controller with a high design performance is realized through the prediction-based digital redesign method. Finally, for further improving the tracking performance, the chaos-evolutionary-programming algorithm (CEPA) is utilized to optimally tune the parameters of the tracker. An example is presented to demonstrate the effectiveness of the proposed methodology.     

Equilibrium joining strategies in the single server queues with negative customers

We consider an M/M/1 queue with negative customers. An arriving negative customer will break the server down and the positive customer being served (if any) is forced to leave the system. Once a breakdown occurs, the server is sent immediately for repair while positive customers are not allowed to join the system during the repair process. When the server is available, positive arrivals decide whether to join or balk the system based on a common reward-cost structure. We consider an observable case that the positive arrivals are informed about the number of customers in the system and an unobservable case without any information. The corresponding Nash equilibrium strategies and the socially optimal joining strategies are explored. We get a socially optimal threshold in the observable case and a mixed joining strategy in the unobservable case. The profit maximization issue is studied, and we derive optimal strategies in two information cases. Finally, numerical examples are provided to show the influence of different parameters on the strategies and social benefit.     

机会网络自私节点的Bertrand博弈与激励机制研究

在真实的网络环境中,很多节点可能是自私的,它们不愿意牺牲自己的资源为其他节点转发消息。针对这种情况,提出一种基于博弈论的激励机制,可以激励节点与其他节点相互合作。该机制为二阶段激励,激励节点接收消息以协助其他节点转发,同时激励节点转发更多的消息。把源节点与中继节点之间的竞争与合作模型化为Bertrand（伯特兰德）博弈,定义了源节点和中继节点的效用函数。求解了源节点的最佳定价策略和中继节点最佳的转发计划,验证了源节点与中继节点之间存在唯一的纳什均衡。模拟仿真结果表明提出的激励机制能够鼓励自私节点参与合作,能提高路由算法的传递率,同时降低了消息传递延迟。与基于声誉的激励机制相比,所提激励机制能使消息传递成功率提高31.4%、平均时延降低9.7%。     

An evolutionary bargaining‐based approach for incentivized cooperation in opportunistic networks

Opportunistic networking enables users to communicate in an environment where connectivity is intermittent or unstable. However, such networking scheme assumes that mobile nodes voluntary cooperate, which cannot be guaranteed. Some nodes can simply exhibit selfish behavior and thus diminish the effectiveness of the approach. In this paper, a game scenario is formulated in which the nodes try to convince each other to participate in packets forwarding. Each node is considered as a player in this game. When a node comes in the communication range of another, a bargaining game starts between them as part of the message forwarding process. Both players try to have a mutual agreement on a price for message forwarding. We present a new incentive mechanism called evolutionary bargaining‐based incentive scheme (EBIS) to motivate selfish nodes to cooperate in data forwarding. In EBIS, a node negotiates with other nodes to obtain an agreeable amount of credit for its forwarding service. Nodes apply a sequential bargaining game and then adapt their strategies using an evolutionary model to maximize the probability of reaching an agreement. Unlike classical bargaining games, nodes in our model are boundedly rational. In addition, we use the evolutionary stable strategy (ESS) concept to determine the adaptive strategies for the nodes. The comparison of EBIS with a benchmarked model demonstrates that EBIS performs better in terms of packet delivery ratio and average latency.     

Femtocell网络的鲁棒议价博弈功率控制

针对具有不完美信道状态信息的Femtocell双层网络，为了保证网络效用最大化、最小服务质量要求及Femtocell用户之间的效用公平，提出了基于议价博弈的鲁棒功率控制方案.在保证宏小区用户正常传输情况下，考虑到Femtocell家庭用户的最小服务质量要求，将Femtocell网络的鲁棒功率控制问题转化为议价博弈功率控制问题.为了保证Femtocell家庭用户之间的公平性，给出了Kalai-Smorodinsky（K-S）议价博弈解，证明了议价博弈解的存在性和唯一性，并提出了一种基于二分查找算法的分布式鲁棒功率控制方案.仿真结果表明，所提出的算法不仅具有较好的收敛性能，而且能够较好地保证用户之间的公平性.     

A hierarchical framework of dynamic relay selection for mobile users and profit maximization for service providers in wireless relay networks

Although extensive research has been carried out on the issue of how to optimally select relays in wireless relay networks, relay selection for mobile users is still a challenging problem because of the requirement that the dynamic selection should adapt to user mobility. Moreover, because the selected relays consume their energy on relaying data for the users, it is required that the users have to pay for this relay service. The price of selecting relays will affect the users' decisions. Assuming that different relays can belong to different service providers, we consider the situation that the service providers can strategically set the prices of their relays to maximize their profits. In this paper, we jointly study the dynamic relay selection for mobile users and profit maximization for service providers. Also, we design a Stackelberg‐game hierarchical framework to obtain the solution. At the lower level, we investigate the relay selection problem for the mobile users under given prices of selecting the relays. It is formulated as a Markov decision process problem with the objective to minimize the mobile user's long‐term average cost (which consists of the payment to the relay service and the cost due to packet loss), and solved by applying the linear programming technique. At the upper level, we study the game of setting relay prices for the service providers, with the knowledge that the mobile users will make relay selections based on their given prices. Nash equilibrium is obtained as the solution. Our results can help to provide a guidance for service providers to compete for providing relay services. Copyright © 2012 John Wiley & Sons, Ltd.     

基于不对称Nash协商模型的供应链协作激励机制研究

建立了制造商与零售商组成的两级供应链,研究了供应链分散决策与协调决策下的合作双方收益问题,发现协调决策下整个供应链收益大于分散决策时整个供应链收益,但整个供应链的收益增加并不能保证合作各方的收益均增加,因此,有必要进行利益协调与激励。利用不对称Nash协商模型建立了激励供应商与分销商进行协作的利益协调模型,并通过案例分析验证了这种激励机制的有效性。