Abstract: In order to solve the problems such as difficult to achieve balance transmission, and the video quality is difficult to meet the needs of user in network video transmission under the conditions of heterogeneous network environment, this paper proposes the Nash equilibrium optimization mechanism of network video data based on feature selection and heterogeneous network bandwidth selection. Firstly, the dynamic evaluation of the network state is conducted to select the appropriate network for transmission by defining the evaluation matrix to calculate the parameters weight vector and selecting the parameters based on the current user video data. Then Nash equilibrium solution is obtained by using the Nash equilibrium optimization mechanism to construct the benefit function and iterative algorithm. The simulation results show that the proposed mechanism can effectively reduce the error level of user??s video experience and improve the user PSNR (peak signal to noise ratio), as well as increasing the efficiency of the network when the user??s video characteristic and network are heterogeneous compared with the BRA-PMEM (bandwidth resource allocation mechanism in pure market economy model).

Key words: heterogeneous network, video characteristics, Nash equilibrium, ruling evaluation matrix, benefit function, optimal allocation bandwidth

摘要： 为解决异构网络环境下网络视频传输中难以实现均衡传输，以及视频质量难以满足用户主观需求等难题，提出了基于特性筛选及异构网络带宽选择的网络视频数据纳什均衡优化机制。首先通过对当前用户视频数据进行基于参数的特性筛选，并通过定义裁决评价矩阵来计算参数权值矢量，从而对网络状况进行动态评估，以便选择适合网络进行传输；然后采用纳什均衡优化机制，构建效益函数并采取迭代算法获取纳什均衡解，从而得到最佳带宽。仿真实验表明，与BRA-PMEM（bandwidth resource allocation mechanism in pure market economy model）机制相比，新机制在用户视频特性及网络处于异构状态时，更能有效降低用户视频体验误差程度，改善用户PSNR（peak signal to noise ratio），提高网络的利用效率。

关键词: 异构网络, 视频特性, 纳什均衡, 裁决评价矩阵, 效益函数, 最佳分配带宽