基于MPTCP的非对称多链路传输调度方法述评

随着通信及网络技术的发展,移动终端设备往往配备多个网络接口,具有多链路的网络接入能力。为实现多链路传输,互联网工程任务组提出了新型的多路径传输控制协议（Multipath Transmission Control Protocol,MPTCP）。如何合理进行MPTCP数据包调度为用户提供更大带宽、提高数据传输可靠性,并最大化网络资源利用率,成为网络通信领域的重要课题。首先,介绍了MPTCP的基本功能和其在数据调度中面临的主要挑战;然后,针对非对称多链路下的MPTCP传输调度算法,从减少数据包乱序、降低传输延迟、提高链路利用率,以及结合强化学习四个角度分析和比较了近年来的多路径传输相关研究;最后,结合当前研究热点展望了MPTCP的未来发展趋势。     

一种智慧协同网络多参数的多路径路由算法

多路径路由技术采用多条路径同时传输,作为优化资源配置和负载均衡的重要技术,在路由可靠性、QoS路由、传输效率等多方面比单路径传输具有优势.现有互联网网络资源配置和路由机制相对静态和僵化,导致多路径技术的发展存在发展的局限性.现有多路径技术考虑在路由层面不利于多路径路由选择和计算,难以保证传输性能,降低网络传输效率.智慧协同网络能够动态感知网络需求,灵活适配网络资源,更好的支持路由可扩展性.本文在智慧协同网络架构下提出了一种智慧协同网络多参数的多路径路由算法.该算法制定了智慧协同网络多参数的多路径路由协议,对网络性能参数CPU占用率、往返时延(RTT)、带宽进行加权计算得到路径权重值,根据权重值进行流量分配.采用图论理论对网络流量分配及模型进行了分析.并在Mini-Net平台上进行了开发和实验,结果表明,该算法能够优化网络配置,减小往返时延,提高网络吞吐量,从而提高网络性能,实现负载均衡.     

多路径传输控制协议技术综述

随着互联网的应用发展,用户对带宽的需求日益增大。同时,伴随着宽带接入技术的发展,终端可以同时具有多条网络链接,然而传统传输控制协议(TCP)采取单路通信,因而造成资源浪费。为此,IETF专门提出了多路径TCP(MPTCP)来实现TCP的多路传输,从而提高链路利用率和协议鲁棒性。本文对国内外MPTCP的最新研究成果进行了总结,包括MPTCP的体系结构、路由和拥塞控制等内容,可为国内研究者进一步深入研究提供参考。     

移动流媒体并行多路传输机制研究

随着我国移动终端的多样化以及人们对网络需求的不断增长,我国移动网络数据的传输数量在近5年增长迅速。然而人们利用移动终端进行网络传输时,通常会受到网络带宽、网络拥塞、网络时延等因素的影响。传统的传输控制协议(Transmission Control Protocol, TCP)与用户数据报协议(User Datagram Protocol, UDP)使用单条路径进行网络传输,已不能很好地满足人们对高速率、实时播放、大吞吐量等网络传输的需求。为了解决这一问题,文章提出一种移动流媒体并行多路传输机制,即多路径传输协议(Multipath Transmission Control Protocol, MPTCP)。该协议可充分发挥多路径传输特性,一方面优化了网络传输的性能,另一方面提高了用户的体验感。     

基于网络效用最大化的一体化网络服务层映射模型

端到端多连接多路径技术能够极大地提高传输的吞吐量,增强数据传输的安全性和可靠性,因此得到了很多学者的关注. 一体化网络中提出了从服务到连接和从连接到路径的映射机制,实现了并行多连接多路径的数据传输. 本文研究了一体化网络服务层的映射机制,基于网络效用最大化思想,给出了从服务到连接和从连接到路径的多对多映射的数学模型. 经连接由服务到路径的映射过程就是如何合理有效地将路径带宽分配给各个服务,从而使得所有服务的效用之和达到最优. 当网络中同时存在弹性服务与非弹性服务时,给出了映射模型的一般形式,确保了非弹性服务能够获得一定的服务质量.     

一种基于端到端多路径的最优化流量控制算法

 随着网络技术的发展,端到端多路径传输正在成为一个热点研究问题.本文研究了基于端到端多路径的最优化流量控制,提出了一种进行最优网络资源分配的分布式算法.该算法充分利用端到端多路径的路径多样性,达到有效的带宽利用和带宽聚合.通过证明,该算法在无时延条件下是全局渐进稳定的.仿真结果证明了该算法的快速收敛性和实用性.     

面向下一代网络的端到端多路径传输层架构

为了解决传统网络无法有效同时使用多家乡终端的多个接口传输数据的问题,提出了一种面向下一代网络的端到端多路径传输层架构-E2EMP.E2EMP通过自适应的根据路径特性分发数据,采用双层序列空间,实施灵活的端到端路径管理,提高了多家乡终端的传输性能.实验仿真表明,E2EMP能够有效地聚合终端多家乡的出口带宽,同时提高了数据传输的安全性和可靠性.     

多路径环境下电信流量模型研究

在下一代通信网络中,一个终端可同时拥有多个网络接口,在一对源、目的终端间存在多条可用路径。这为网络通信提供了一种全新的思路,即利用多条路径并发进行数据传输。针对多路径传输出现的问题,文章为多路径传输建立了流量模型,分析了多路径传输流媒体的优势,提出了＂流体带宽＂和＂流体调度＂的概念,并引入网络流理论计算了多路径环境下的目标流量。     

一种基于链路传输延迟的MPTCP拥塞控制算法*

由于传统TCP拥塞控制算法直接应用到MPTCP(Multipath Transport Control Protocol)中存在公平性问题,以及不能有效地发挥多路径传输的优势,因而从公平性方面对MPTCP现有拥塞控制算法进行研究.研究发现,现有的MPTCP拥塞控制算法均受到相同的回路时间限制.提出一种基于链路延迟的RTT补偿算法(Compensating for RTT mismatch,C-RTT ).该算法通过设置网络带宽占用比参数以及对MPTCP连接的子流设置侵略因子,从而保证瓶颈链路处MPTCP数据流和TCP数据流公平地共享可用带宽.最后通过NS3仿真实验证明,该算法能够有效地补偿链路中因RTT不等引起的公平性问题,并避免链路之间数据的非周期抖动,且保证了多路径传输的优越性.     

基于跳到跳信息的卫星网络传输控制协议研究

设计了一种适用于卫星网络的传输控制协议TPSN。该协议使用异步跳到跳确认,快速恢复成段丢失数据,并采用基于检测窗口的端到端选择性否定应答(SNACK)机制,减少协议控制信息,保证数据可靠传输。在跳到跳可靠性保证机制基础上传输网络负载,实现在避免网络拥塞条件下高效利用网络带宽资源,并保证具有不同端到端往返时延的异种数据流之间的公平性。仿真结果表明TPSN能够在长延迟、误码率高、链路频繁切换的卫星网络中保证高效的带宽利用、可靠的数据传输以及各数据流之间的公平。     

数据中心网络中的MPTCP Incast（英文）

In recent years,dual-homed topologies have appeared in data centers in order to offer higher aggregate bandwidth by using multiple paths simultaneously.Multipath TCP(MPTCP) has been proposed as a replacement for TCP in those topologies as it can efficiently offer improved throughput and better fairness.However,we have found that MPTCP has a problem in terms of incast collapse where the receiver suffers a drastic goodput drop when it simultaneously requests data over multiple servers.In this paper,we investigate why the goodput collapses even if MPTCP is able to actively relieve hot spots.In order to address the problem,we propose an equally-weighted congestion control algorithm for MPTCP,namely EW-MPTCP,without need for centralized control,additional infrastructure and a hardware upgrade.In our scheme,in addition to the coupled congestion control performed on each subflow of an MPTCP connection,we allow each subflow to perform an additional congestion control operation by weighting the congestion window in reverse proportion to the number of servers.The goal is to mitigate incast collapse by allowing multiple MPTCP subflows to compete fairly with a single-TCP flow at the shared bottleneck.The simulation results show that our solution mitigates the incast problem and noticeably improves goodput in data centers.     

URLM: Utilising reinforcement learning to schedule subflows in MPTCP

Parallel transmission in multiple access networks using MultiPath TCP (MPTCP) greatly enhances the throughput. However, critical packet disorder is commonly observed due to traffic fluctuation and path diversity. Although several predictive scheduling algorithms have been proposed to solve this problem, they cannot accommodate prediction accuracy and real-time adaptation simultaneously in a dynamic network environment. The time overhead in modifying scheduling parameters to adapt to network changes leads to performance degradation in throughput and packet disorder. In this study, we propose a scheduling algorithm called U tilising R einforcement L earning to Schedule Subflows in M PTCP (URLM). We apply reinforcement learning to select an optimal scheduling parameter in real time, which brings significant time benefits for modifying the parameters. The simulation comparison experiments show that URLM reduces the average number of out-of-order packets and the time overhead in adapting to network changes while improving global throughput.     

面向车联网按需驱动的多宿主多链路TCP拥塞控制算法

针对车联网终端设备中网络应用QoS的多样化需求,保证与安全应用相关的高优先级数据报文发送实时性,提出了一种新的基于数据优先级与吞吐量评估的按需驱动的MPTCP拥塞控制算法PTLIA。首先,算法采用数据报文优先级以及吞吐量占比因子,表征各个数据报文的发送权重;其次,利用分批估计理论模型对MPTCP中路径状态进行实时评估;最后,依据算法模型设计,实现对网络拥塞时间窗的按需动态调整。实验与仿真验证了PTLIA算法在满足大部分网络应用需求的前提下,提高了高优先级应用的网络传输实时性。     

Highly available network design and resource management of SINET4

The Japanese academic backbone network has been providing a variety of multilayer network services to support a wide range of research and education activities for more than 700 universities and research institutions. The new version, called SINET4, was launched in 2011 in order to enhance the service availability and the network bandwidth as well as to expand the service menu. Its enhanced service availability was unexpectedly verified by the disastrous March 11 Great East Japan Earthquake, when the network managed not to stop service operation even after the earthquake. This paper describes the design and implementation of SINET4 in terms of multiple service provision, network resource control and management, and high reliability from physical level to network management level. The impacts of the huge earthquake are also reported.     

可信度动态感知的间断连接无线网络数据转发机制

提出了一种节点可信度动态感知的间断连接无线网络数据转发机制,节点根据运动过程中所获知的局部网络状态信息以分布式的方式估计其他节点恶意度及协作度,并根据历史信息预测节点的连通状态,进而获知各个节点的信任程度,以合理地选择中继节点,实现高效的数据转发。仿真结果表明,所提出的机制能够有效提高数据传输的可靠性,并大幅降低网络负载率,提高网络资源利用率。     

基于边缘云计算的光纤无线网络优化设计

为了提升网络边缘数据处理能力,满足终端大带宽和低时延的要求,构建了基于边缘基础设施的云计算平台,设计了具有动态带宽调整的光纤网络模型。提出了一种基于边缘云计算的时序优化算法,并将其应用于光纤无线网络。通过OPNET软件仿真分析了时序优化算法的传输时延均值,结果显示,优化后最大时延为43.1 ms,仅为传统方法的34.2%。实验对局域网内多个终端之间的数据通信进行分析,讨论了三种算法的传输能效、光纤信道利用率及传输能耗。实验结果显示,采用时序优化算法的测试结果具有明显改善,其传输能效提升了近1倍,边缘云数据传输时延均值信道利用率提升了约6.2%,网络传输能耗均值最优。该光纤无线网络模型及其优化算法在传输时延、信道利用率以及网络能耗方面具有明显提升。其在提升光纤通信链路选择及边缘端数据交互中具有一定的优势。     

Increasing lifetime of wireless sensor networks using controllable mobile cluster heads

In a static wireless sensor network (WSN), sensors close to the base station (BS) run out of energy at a much faster rate than sensors in other parts of the network. This is because the sensor close to the BS always relays the data for other sensors, resulting in an unequal distribution of network residual energy. In this paper, we propose a scheme for enhancing the network lifetime using multiple mobile cluster heads (CHs) that can move in the WSN in a controllable manner. The CH controllably moves toward the energy‐rich sensors or the event area, offering the benefits of maintaining the remaining energy more evenly, or eliminating multihop transmission. Therefore, the proposed scheme increases the network lifetime. We theoretically analyze the energy consumption in our scheme and propose three heuristical mobility strategies. We further study the collaboration among CHs in order to maintain their connectivity to the BS to ensure the delay requirement for real‐time applications. Simulation shows that network lifetime is increased by upto 75% over existing approach by making CHs always move toward a stable equilibrium point. Our connectivity algorithm provides a best case improvement of 40% in transmission delays over existing schemes. Copyright © 2009 John Wiley & Sons, Ltd.     

Topics in Optical Communications - Service Differentiation and Traffic Engineering in IP over WDM Networks

The emergence of WDM-based optical transport has brought about unprecedented potential for ultra-high-speed IP core networks. It also holds great challenges, not least of which is to determine the optimal way of merging the optical domain with the IP domain. Beyond this issue, still a subject of research, the path to an efficient and resilient optical transmission IP core network is beset by numerous difficulties caused by shortcomings inherent to IP. Simply merging WDM and IP will not automatically produce a core network capable of serving the needs and adapting to the nature of diverse IP traffic. This article demonstrates the complexities of and proposes a possible solution for optimizing the next-generation core network, evolving the management plane IP/WDM integration model proposed in the 1ST WINMAN project. We concentrate on making the core network efficient for transporting differentiated service traffic, adaptive to changes in traffic patterns and resilient against possible failures, which would disrupt its operation. To these ends we introduce MPLS TE and DiffServ QoS functionality in our network model, thus ensuring maximum utilization of resources and automated guarantees of optimal quality for different types of transported traffic.