TCP-ATCA: Improved Transmission Control Algorithm in Satellite Network

An adaptive transmission control algorithm based on TCP （TCP-ATCA） is proposed to reduce the effects of long propagation delay and high link error rate of the satellite network on the performances. The flow control and the error recovery are differentiated by combined dynamic random early detection-explicit congestion notification （DRED-ECN） algorithm, and, moreover, the pertaining congestion control methods are used in TCP-ATCA to improve the throughput. By introducing the entire recovery algorithm, the unnecessary congestion window decrease is reduced, and the throughput and fairness are improved. Simulation results show that, compared with TCP-Reno, TCP-ATCA provides a better throughput performance when the link capacity is higher （ ≥600 packet/s）, and roughly the same when it is lower. At the same time, TCP-ATCA also increases fairness and reduces transmission delay.

针对车地传输TCP协议优化的TTI-RTT拥塞控制算法

为适应铁路5G专用移动通信系统动态易变的网络环境,解决现有算法中存在的吞吐量减少、传输时延增加、检测数据车地传输可靠性难以保证的问题,在分析传输控制协议(TCP)传输拥塞控制过程,以及对比TCP-Reno,TCP-Vegas和TCP-Westwood 3种基础拥塞控制算法存在问题的基础上,提出一种联合传输时间间隔(TTI)和往返时延(RTT)的拥塞控制算法——TTI-RTT算法,即将RTT的成对比较测试统计量和网络当前拥塞因子联合起来作为网络拥塞判断标准,根据反映网络拥塞程度的拥塞因子动态调节加性和乘性因子,并在传输控制协议拥塞控制的不同条件下灵活调整传输时间间隔。采用NS-3仿真软件,构建考虑高速铁路5G-R无线信道误码影响的点对点通信网络拓扑结构,对TTI-RTT拥塞控制算法进行仿真验证,并与3种基础算法进行性能比较。结果表明:不同丢包率下所提算法的平均吞吐量均保持在9 Mb·s^(-1)以上,较基础算法中吞吐量最高且最稳定的TCP-Reno算法增加了约5.9%,且实时吞吐量稳定性具有显著改善。

改进TCP/IP帧结构的多参数信道传输性能评估方法

目前,网络密集化现象加剧了信道传输性能的恶化。而普遍应用的评估方法由于参数考虑有限和适用范围受限,在信道传输性能评估方面面临显著挑战。针对信道传输性能评估困难的问题,提出一种改进传输控制协议/网际协议(TCP/IP)帧结构的多参数信道传输性能评估方法。首先,生成标准化的测试数据,包括伪随机码、基本曲线数据、自定义曲线数据,以确保测试数据遵循统一标准;其次,使用改进的TCP/IP帧结构,将总帧数、帧序号等测试数据信息封装在TCP/IP帧中,并对测试数据进行收发与解析,统计不同帧类型的帧数量、不同帧长度的帧数量、总帧数和有效数据量这4种基础信道传输信息;最后,分析接收数据得到误帧率和误码率这2种高级信道传输信息,完成信道传输性能的整体评估。所设计的方法采用6种不同的参数对信道质量进行评估,方法评估的精度可达0.01%,最小误差为0.01%,且兼容所有使用TCP/IP进行通信的信道。实验结果表明,所提出的信道传输性能评估方法可以完成对6种信道通信信息的统计与分析,准确评估信道传输性能。

TCP Window Based Congestion Control -Slow-Start Approach

Transmission control protocol (TCP) has undergone several transformations. Several proposals have been put forward to change the mechanisms of TCP congestion control to improve its performance. A line of research tends to reduce speed in the face of congestion thereby penalizing itself. In this group are the window based congestion control algorithms that use the size of congestion window to determine transmission speed. The two main algorithm of window based congestion control are the congestion avoidance and the slow start. The aim of this study is to survey the various modifications of window based congestion control. Much work has been done on congestion avoidance hence specific attention is placed on the slow start in order to motivate a new direction of research in network utility maximization. Mathematical modeling of the internet is discussed and proposals to improve TCP startup were reviewed. There are three lines of research on the improvement of slow start. A group uses the estimation of certain parameters to determine initial speed. The second group uses bandwidth estimation while the last group uses explicit request for network assistance to determine initial startup speed. The problems of each proposal are analyzed and a multiple startup for TCP is proposed. Multiple startups for TCP specify that startup speed is selectable from an n-arry set of algorithms. We then introduced the e-speed start which uses the prevailing network condition to determine a suitable starting speed.

AN EFFECTIVE NETWORK CONGESTION CONTROL METHOD FOR MULTILAYER NETWORK

The congestion control problem in a single node network has been solved by the nonlinearfeedback control method,which has been proven to be effective and robust for different router’s queuesize.However,these control models are based on the single layer network architecture,and the sendersand receivers are directly connected by one pair of routers.With the network architecture being moreand more complex,it is a serious problem how to cooperate many routers working in the multilayernetwork simultaneously.In this paper,an effective Active Queue Management(AQM)scheme toguarantee the stability by the nonlinear control of imposing some restrictions on AQM parameter inmultilayer network is proposed.The nonlinear control can rely on some heuristics and network trafficcontrollers that appear to be highly correlated with the multilayer network status.The proposedmethod is based on the improved classical Random Early Detection(RED)differential equation and atheorem for network congestion control.The theorem proposed in the paper proved that the stability ofthe fluid model can effectively ensure the convergence of the average rate to its equilibrium pointthrough many routers in multilayer network.Moreover,when the network capacity is larger,theproposed scheme can still approach to the fullest extensibility of utilization and ensure the stability ofthe fluid model.The paper reveals the reasons of congestion control in multilayer network,provides atheorem for avoiding network congestion,and gives simulations to verify the results.

On Approaches to Congestion Control over Wireless Networks

Congestion control in wireless networks has been extensively investigated over the years and several schemes and techniques have been developed, all with the aim of improving performance in wireless net-work. With the rapid expansion and implementation of wireless technology it is essential that the congestion control problem be solved. This paper presents a survey of five congestion control schemes which are dif-ferent in slow start threshold calculation, bandwidth estimation, and congestion window manipulation. A comprehensive comparison of these approaches is given in relation to assumptions, bandwidth estimation, congestion window size manipulation, performance evaluation, fairness and friendliness and improved throughput.

AN ENHANCEMENT SCHEME OF TCP PROTOCOL IN MOBILE AD HOC NETWORKS:MME-TCP

Transmission Control Protocol (TCP) optimization in Mobile Ad hoc NETworks (MANETs) is a challenging issue because of some unique characteristics of MANETs. In this paper,a new end-to-end mechanism based on multiple metrics measurement is proposed to improve TCP performance in MANETs. Multi-metric Measurement based Enhancement of TCP (MME-TCP) designs the metrics and the identification algorithm according to the characteristics of MANETs and the experiment results. Furthermore,these metrics are measured at the sender node to reduce the overhead of control information over networks. Simulation results show that MME-TCP mechanism achieves a significant performance improvement over standard TCP in MANETs.

TCP/AWM网络系统的事件触发预设性能拥塞控制

研究了具有非响应用户数据报协议流干扰的TCP/AWM网络系统的跟踪控制问题.针对网络拥塞控制系统,给出了一种改进的网络模型.将预设性能控制和事件触发机制相结合,提出了一种新的AWM拥塞控制算法.利用改进的边界性能函数限制TCP/AWM网络系统的队列跟踪误差,提高了闭环系统的暂态和稳态性能,保证了所有闭环信号都是有界的.此外,采用基于一位信号传输的事件触发控制协议,当控制信号更新时,仅通过一个二进制信号(0或1)传输信息.与传统的事件触发控制协议相比,不仅考虑了有限的带宽,而且考虑了网络传输的成本以及安全性问题.最后,通过仿真验证和分析,证明了所提方法的可行性和优越性.

基于TCP/IP的多媒体技术支持系统设计和实现

随着网络技术的发展,网络技术逐渐渗透到各个行业领域,很多领域顺应时代发展进行多样化的变革。其中,实际生产和工作中使用多媒体技术支持系统可以共享网络资源,提高资源的利用率和工作效率。文章介绍了传输控制协议/网际协议(Transmission Control Protocol/Internet Protocol,TCP/IP),并对基于TCP/IP的多媒体技术支持系统进行了设计和实现。

网络通信中嵌入式TCP/IP单片机技术研究

嵌入式传输控制协议/网际协议(Transmission Control Protocol/Internet Protocol,TCP/IP)单片机技术是网络通信中一项重要的技术。主要从嵌入式TCP/IP单片机技术的概述、硬件结构、软件设计进行详细介绍,提出一种嵌入式TCP/IP单片机的软件设计方法,并将其应用于网络通信,以提高网络通信的稳定性和可靠性,并提高数据传输质量。

适用于数据中心低延迟高吞吐量TCP的动态标记配置策略

在数据中心流量传输过程中,传输性能的好坏很大程度上取决于数据中心TCP是否可以满足长短数据流不同的传输性能需求,即长流的高吞吐率和短流的低排队延迟。由于数据中心网络应用主要是在线数据密集型应用(online data intensive application,OLDI)产生大量长流和短流混合的数据流,使长流占用大部分瓶颈链路带宽,造成短流高延迟,影响应用的服务质量。文章提出的动态标记配置(Dynamic-marking Configuration,DMC)策略动态调整瓶颈链路队列长度门限,实现流量的低延迟和高吞吐量传输,并降低数据流错过最后期限的比例。

基于共享存储技术的乱序TCP分段重排电路设计

传输控制协议(Transmission Control Protocol,TCP)处理引擎可以广泛应用于数据中心的各类型服务器,降低处理器负荷。为提高TCP协议处理引擎处理数据的分段效率,设计了一种基于共享存储技术的乱序TCP分段重排电路,其采用静态随机存储器(Static Random Access Memory,SRAM)存储,具有结构简单、存储资源利用率高、低时延的特性,可有效克服内容寻址存储器(Content Addressable Memory,CAM)或三态内容可寻址存储器(Ternary Content Addressable Memory,TCAM)高功耗以及可扩展性差的缺点,同时满足上千条连接的工作需求;分析了电路的具体结构、关键调度算法和工作流程,并进行了仿真。电路基于Xilinx ZYNQ 7000系列的现场可编程门阵列(Field-Programmable Gate Array,FPGA)实现,对主要硬件资源消耗进行了综合分析。

传输控制中的确认机制研究

Internet传输层协议需要依赖于确认(ACK)机制提供的反馈信息,实现拥塞控制和可靠传输等功能.以Internet传输协议演化的历史为线索,回顾传输控制领域中的确认机制,并讨论现有确认机制中需要解决的问题;基于“类型-触发条件-信息”三要素,提出按需确认机制及其设计原则,重点分析确认机制和拥塞控制、丢包恢复等传输协议子模块之间的耦合关系;结合设计原则,详细阐述一种可行的按需确认机制实现——TACK机制,并对相关概念进行系统的、深入的分析和澄清.最后结合按需确认机制面临的挑战,给出几个有意义的研究方向.

MANET中TCP改进研究综述

传统TCP(transmission control protocol)本是为有线网络设计,它假设包丢失全是由网络拥塞引起,这个假设不能适应于MANET (mobile ad hoc network),因为MANET 中除了拥塞丢包以外,还存在由于较高比特误码率、路由故障等因素引起的丢包现象.当出现非拥塞因素丢包时,传统 TCP 将错误地触发拥塞控制,从而引起TCP 性能低下.任何改进机制都可以分为发现问题和解决问题两个阶段.首先概括了 MANET 中影响 TCP 性能的若干问题;然后针对发现问题和解决问题两个阶段,详细地对每一阶段中存在的各种可行方法进行了分类、分析和比较;最后指出了 MANET 中 TCP 性能优化的研究方向.

一种异构网络TCP拥塞控制算法

针对互联网中端对端带宽、时延和丢包率等的差异性日益加剧,导致TCP传输性能严重退化,该文提出一种链路自适应TCP拥塞控制算法(INVS)。INVS在拥塞避免阶段初期采用基于指数函数的凸窗口增长函数,以提高链路利用率;在窗口增长函数中引入了自适应增长因子实现窗口增长速率与链路状态相匹配;采用了自适应队列门限的丢包区分策略以提高无线环境下TCP的性能。性能分析和评估表明,INVS提高了TCP拥塞控制算法的吞吐量、公平性、链路利用率和RTT公平性。

改进的无线网络TCP协议跨层设计方法

针对传统TCP协议的拥塞控制机制在无线网络中不能很好地解决数据丢失的问题,提出一种改进的无线网络TCP协议跨层设计方法——NTCP。NTCP将传统TCP功能分割为传输速率控制和数据完整性控制2个部分,避免使用转发节点信息的方法,最大限度地利用无线资源。该方法与传统TCP-Reno方法相比,能提高无线网络的吞吐量,其稳定性和速率自适应能力也有较好的改善。

Ad Hoc中的TCP改进方案——Adaptive ADTCP

在分析无线自组网特点及其对TCP性能影响的基础上,提出了一种能够自适应无线自组网状态的TCP改进方案(Adaptive ADTCP)。Adaptive ADTCP在明确辨识网络状态的前提下,首先根据前向路径跳数自适应地调整拥塞窗口增长因子,限制源节点TCP拥塞窗口过分增长,避免造成网络拥塞;同时源节点TCP根据当前拥塞窗口自适应改变发送的分组长度,充分利用网络资源。仿真实验表明,在网络重负载和节点高速移动情况下,Adaptive ADTCP对数据传输的吞吐量有较大提升。

无线多跳网络中Semi-TCP协议的实现

为解决传统TCP协议在无线多跳网络中经常误判网络拥塞状况且反应迟钝的问题,提出一种Semi-TCP协议的实现方法。添加虚拟TCP层控制包的过滤和重发,采用逻辑信道传递跨层信息,应用一跳式ACK速率控制策略调节发送速率。仿真实验结果表明,该实现方法能在不改变原TCP协议框架的基础上提高系统性能。

TCP拥塞控制解决方法分析及评价

探讨了网络拥塞出现的原因及其一般解决方法，并主要对TCP的拥塞控制做了较深入的研究，阐述了在TCP网络中处理拥塞的几种算法。通过对原理的分析和性能的比较，指出了在目前网络实现中可以优先采用的拥塞控制算法。

考虑慢启动影响的TCP吞吐量模型

TCP吞吐量模型的研究是网络协议研究的一个重要方面 ,同一些其它模型相比 ,Padhye提出的TCP吞吐量模型比较精确地描述了TCP吞吐量与往返时间、丢包率和超时时限的关系 ,但在丢包率很高的场合 ,Padhye模型误差较大 ,本文分析了高网络负荷下TCP传输的性能 ,提出了一个高网络负荷下TCP吞吐量的改进Padhye模型 ,实验表明 ,该模型在高网络负荷环境下更接近实际情况 .