网络速率控制的博弈模型

Network Flowrate Control Model Based on Game Theory

传统的网络速率控制方案需要端系统用户合作以达到最优的网络性能.但是,当存在不合作端系统用户时,这些方案不可避免地会出现拥塞崩溃.为此,提出了一种非合作博弈网络速率控制框架,该方案基于非合作博弈论的Nash解的思想,博弈的各用户支付网络使用费并选择愿付价格以最大化自己的净收益.文中还设计了一种网络带宽定价机制,驱使自私用户流向社会最优解操作,并证明了该速率控制博弈可达惟一的Nash均衡点且带宽分配是有效与公平的.

Traditional flowrate control schemes require the cooperation of traffic sources to achieve an optimal network performance.However, all of such schemes will lead inevitably to a congestion collapse in the presence of noncooperative end-users.In order to solve this problem,a flowrate control framework is proposed based on the noncooperative game theory.In this scheme,the idea of the Nash solution from noncooperative game theory is taken into account,and each user has to pay for the network and has to choose a willingness-to-pay to maximize his own net profit.Moreover,a pricing mechanism for network bandwidth is designed to drive the user flow to the social optimal solution.The results prove that the proposed flowrate control game admits a unique Nash equilibrium point,at which the bandwidth allocation is efficient and fair.