基于MPUDP的多链路信道分配协议设计与实现  被引量:2

Design and implementation of multi-link channel allocation protocol based on MPUDP

在线阅读下载全文

作  者:王卉 占敖 吴呈瑜 夏雨峰 WANG Hui;ZHAN Ao;WU Chengyu;XIA Yufeng(School of Information and Technology,Zhejiang Sci-Tech University,Hangzhou 310018,China)

机构地区:[1]浙江理工大学信息学院,杭州310018

出  处:《智能计算机与应用》2022年第7期172-177,共6页Intelligent Computer and Applications

基  金:国家自然科学基金(61801430);省自然科学基金(LY18F010022)。

摘  要:在如今网络直播爆发的时代背景下,传统的常规用户数据报协议(User Datagram Protocol, UDP)存在链路性能不稳定、可靠性低、高延时等问题,而多链路并行传输是克服该问题的重要技术途径之一。本文基于UDP传输协议设计了一种多路径协作传输协议,即MPUDP(Multi-path User Datagram Protocol)。协议通过数据训练对每条信道的丢包率、RTT和乱序情况进行实时监测,对每条路径是否参与传输进行管理,并且对传输路径进行速率分配,有效改善系统的传输带宽。仿真结果表明:本文提出的多路径协作传输协议相比于UDP传统协议,在相同的误码率情况下,带宽近似扩大链路数目的倍数,与MPTCP协议相比,在损失丢包率0.1%的情况下,带宽提升了链路数量倍的吞吐量。In the current era of webcasting, the traditional user datagram protocol(User Datagram Protocol, UDP) has lots of problems such as unstable link performance, low reliability, and high latency. Multi-link parallel transmission is one of the important technical approaches which can overcome these problems. Based on the UDP transmission protocol, a multi-path cooperative transmission protocol called MPUDP(Multi-path User Datagram Protocol) is designed. The protocol uses data training to monitor the packet loss rate, RTT and disorder of each channel in real time. The participation of path in the transmission management and the transmission path rate are allocated, which effectively improves the transmission bandwidth of the system. The simulation results show that compared with the traditional UDP protocol, the bandwidth is expanded about a multiple of the number of links when MPUDP protocol has the same bit error rate. Compared with the MPTCP protocol, when the loss of packet loss rate is 0.1%, the bandwidth is increased by nearly multiple links times.

关 键 词:UDP 速率分配 多路径 协作传输 

分 类 号:TP393[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象