论文部分内容阅读
文中在用户容量受限的异构无线网络环境下,基于进化博弈理论研究了时延对用户网络选择行为的影响。设计了一种收益函数,以可用带宽和网络负载表征网络性能,并在收益函数中加入了惩罚项,使得理性用户不会去选择不可行的策略;提出了一种集中式的带时延的进化博弈实现算法,分析了迭代步长的选取。数值与仿真结果表明,网络存在时延的条件下,设计的收益函数和算法仍能保证种群演进到均衡状态。
In the context of heterogeneous wireless networks with limited user capacity, the evolutionary game theory is used to study the effect of delay on user network selection behavior. A revenue function is designed to characterize the network performance with available bandwidth and network load. The penalty function is added to the revenue function so that rational users will not choose an impractical strategy. A centralized time-dependent Evolutionary game algorithm, analysis of the selection of iterative steps. The numerical and simulation results show that the designed revenue function and algorithm can still guarantee the population evolution to equilibrium under the condition of network delay.