论文部分内容阅读
传统网络终端设备通常以单路径连接网络,因此在路径发生异常时通常严重影响数据传输质量.随着重叠网络和异构多接入网络的发展,多路径传输逐渐成为可能.多路径传输中数据发送端的多路径调度算法是研究多路径传输的核心问题.本文提出了一种基于马尔科夫决策过程的多路径冗余传输调度算法,通过分阶段采用稳态马尔科夫策略并对关键数据进行冗余传输,实现了实时性要求较高的数据传输,提高了数据传输可靠性.仿真实验表明当路径传输质量发生改变时,该调度算法能够正确感知并将数据更多的调度到传输质量较好的路径上,有效降低路径质量变差时对传输的影响.该算法可应用于音、视频等实时业务中,能够有效提高用户使用体验.
Traditional network terminal devices usually connect to the network in a single path, and therefore, the quality of data transmission is often seriously affected in the event of an abnormality.With the development of overlay networks and heterogeneous multi-access networks, multi-path transmission becomes possible gradually. The multi-path scheduling algorithm at sending end is the core issue of multi-path transmission.This paper proposes a multi-path redundancy transmission scheduling algorithm based on Markov decision process, through steady-state Markov strategy and key data Redundant transmission, which realizes the real-time data transmission and improves the data transmission reliability.The simulation results show that when the path transmission quality changes, the scheduling algorithm can correctly perceive and schedule more data to the transmission quality Good path, and effectively reduce the impact on the transmission when the path quality deteriorates.The algorithm can be applied to real-time services such as audio and video, which can effectively improve the user experience.