中文摘要：

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。

英文摘要：

This paper analyzes the delay＇ s impact on users＇ network selection behavior based on evolution- ary game theory （EGT） in a capacity constrained heterogeneous wireless networks scenario. It describes a utility function to estimate the network performance in terms of bandwidth and network load, then intro- duces a penalty term into the utility function so that the rational users would not select the infeasible strate- gies. A centralized evolutionary game algorithm with delay is proposed in this paper, and the selection of the iteration step size is analyzed. The numerical and simulation results indicate that the proposed utility function and algorithm could still enable the populations evolve to the equilibrium when the networks have delay.