基于速率自适应的移动网络音视频传输拥塞控制的研究

发布时间：2018-06-07 16:02

本文选题：音视频传输 + 拥塞控制　；参考：《华北电力大学(北京)》2017年硕士论文

【摘要】：随着通信技术及流媒体技术的发展,网络中存在大量的实时音视频内容。网络拥塞会导致网络的传输时延急剧增加,网络性能急速下降。由于实时音视频数据的传输一般采用RTP协议或UDP协议,这两种协议并不提供服务质量保证,也不提供网络的拥塞控制。因此对实时音视频数据的传输提供拥塞控制机制得到了广泛的关注。本文针对音视频应用的特点,重点研究在保障对TCP数据流的友好性的前提下,提高音视频数据传输质量的拥塞控制技术。拥塞控制机制需要考虑以下三点。一是TCP友好性,即从长期来看,实时音视频数据流不过多的抢占TCP流的带宽。二是音视频应用的可用性,即应用拥塞控制机制时,应该避免过度的调整导致的音视频应用不可用。三是对于数据包丢失原因的区分,即在移动网络环境下,误码产生的数据包的丢失不可作为网络拥塞来处理。针对所考虑的问题,本文所做的主要工作如下:1.比较了相关的拥塞控制机制,并具体的分析了TFRC(TCP Friendly Rate Control)的拥塞控制机制。2.在TFRC协议的基础上应用spike算法,使其可以区分数据包丢失原因,避免在移动环境中对发送速率不必要的减少。3.针对音视频应用的服务质量要求,改进TFRC机制,使其在保证音视频应用的可用性的情况下,尽力避免产生严重的网络拥塞。4.对改进后的TFRC机制进行仿真实验,通过实验表明,改进达到了良好的效果。
[Abstract]:With the development of communication technology and streaming media technology, there are a lot of real-time audio and video content in the network. Network congestion will result in a sharp increase in transmission delay and a rapid decline in network performance. Because real-time audio and video data are transmitted by RTP protocol or UDP protocol, these two protocols do not provide quality of service assurance, nor do they provide congestion control in the network. Therefore, congestion control mechanism for real-time audio-video data transmission has received wide attention. According to the characteristics of audio and video applications, this paper focuses on the congestion control technology to improve the quality of audio and video data transmission under the premise of protecting the friendliness of TCP data flow. Congestion control mechanisms need to consider the following three points. One is TCP friendliness, that is, in the long run, real-time audio and video streams do not preempt the bandwidth of TCP streams. Second, the usability of audio and video applications, that is, the application of congestion control mechanism, should avoid excessive adjustment of audio and video applications are not available. The third is to distinguish the reasons of packet loss, that is, in the mobile network environment, the packet loss caused by error can not be treated as network congestion. The main work of this paper is as follows: 1. The relative congestion control mechanism is compared, and the congestion control mechanism of TFRC(TCP Friendly Rate Control) is analyzed in detail. Based on the TFRC protocol, the spike algorithm is applied to distinguish the cause of packet loss and to avoid unnecessary decrease of transmission rate in mobile environment. Aiming at the quality of service (QoS) requirements of audio and video applications, the TFRC mechanism is improved to ensure the availability of audio and video applications and to avoid serious network congestion. Simulation experiments on the improved TFRC mechanism show that the improvement achieves good results.
【学位授予单位】：华北电力大学(北京)
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN919.8

【参考文献】