面向多媒体流的主动分层多播拥塞控制方案

针对现有分层多播拥塞控制方案存在较大的拥塞响应延时,吞吐率抖动剧烈,和不满足TCP友好的问题,给出了一个主动分层多播服务模型,通过引入主动标记分层、可用带宽主动测量以及优先级分层过滤机制,提出了一种新的面向多媒体流的自适应主动分层多播拥塞控制方案AALM。仿真结果表明,该方案能够有效地改进分层多播拥塞控制的性能,具有较快的拥塞响应速度、较好的稳定性和TCP友好特性。     

基于ECN带预测验证的拥塞控制算法

提出了一种新的拥塞控制源算法FAV(Forecast and Verify),这是一种在接收方测量分组到达的延迟来预测网络中是否将出现拥塞的算法,并结合ECN来验证预测的准确性。通过和TCP Reno、TCP Vegas算法进行比较实验,表明FAV算法能够有效地减小丢失率和链路延迟,同时保持较高的链路利用率,从而使网络中的拥塞得到较好控制。     

TCP拥塞控制算法研究

随着互联网规模的增长，互联网上的用户和应用都在快速的增长，拥塞已经成为一个十分重要的问题。近年来，在拥塞控制领域开展了大量的研究工作。拥塞控制算法可以分为两个主要部分：在端系统上使用的源算法和在网络设备上使用的链路算法。在介绍拥塞控制算法的基本概念后，本文总结了在TCP协议方面拥塞控制算法的研究现状，并分析了进一步的研究方向。     

一种能处理无线链路拥塞的TCP/ARQ改进算法

采用链路层自动请求重传（ARQ）方法来对分组差错进行恢复，并对传统的TCP／ARQ算法加以改进，加入了在无线链路拥塞情况下的自适应拥塞避免算法。仿真结果表明，这种能处理无线链路拥塞的TCP／ARQ算法对于提高无线网络的吞吐量非常有效。     

一种改进的RTP自适应算法

提出了一种利用RTP／RTCP协议收集的QoS信息来控制UDP流的变量加增长自适应算法。与常量加增长算法相比，此算法具有更好的平稳性。此外，还对这种变量加增长的算法进行了仿真，并分析了结果。     

基于AT89S52单片机的数控直流电流源设计

科学仪器和电子设备中经常用到稳定性好、精度高、输出可预置的直流电流源。高性能的数控直流电流源不仅能够提高科学仪器的稳定性，还能够缩短产品的研发周期，因此具有很高的经济价值。本电流源能够很好地降低因元器件老化、温漂等原因造成的输出误差，输出电流在20mA～2000mA可调，输出电流可预置、具有“＋”、“-”步进调整、输出电流信号可直接显示等功能。硬件电路以AT89S52单片机为控制核心器，构成闭环控制电路，采用压控电流方式实现稳流输出，步进为1mA，通过DAC控制压控电流步进，ADC采样电压值，软件算出电流的测量值。系统主要由键盘输入、控制处理、压控电流输出、液晶显示等功能模块组成。经实际测试，该系统可靠性好、精度较高、操作简单方便、人机界面友好。     

一种融合MAC层拥塞通告的混合网络TCP协议

在研究无线网络媒体接入控制(MAC)层拥塞测度的基础上,提出了一种跨层的显式拥塞通告(ECN)机制,即:当数据包中记录的请求发送(RTS)次数超过给定阈值时,通过ECN向传输控制协议(TCP)源端发送拥塞通告,从而启动TCP拥塞控制.这种跨层设计是对有线网络中基于主动队列管理(AQM)的拥塞控制的有效补充,由此可以得到一种与已有的协议无缝连接的混合网络TCP模型.通过在网络模拟器NS2中构造多流无线局域网和多跳无线/有线混合网络,对所提出的方法进行了仿真,实验结果说明该方法能够提高混合网络的性能,并且具备良好的扩展性.     

网络拥塞控制算法

随着计算机网络的持续快速发展,各种网络应用需求不断涌现,造成网络数据流量的激增。网络拥塞问题变得越来越严重,网络拥塞控制也一直是网络研究的最关键热点问题之一。在本文中,作者着重阐述了TCP／IP拥塞控制中的典型算法,并指出了这些算法的优缺点。     

Google WebRTC网络拥塞控制策略的研究

伴随着近几年人们对网络视频，语音通讯需求的极大增长，国内网络带宽提速的脚步依然很慢。从而使得网路拥塞的现象更加严重。在本文中，作者阐述了当前TCP拥塞控制的典型算法，着重针对GoogleWebRTC技术中的拥塞解决方案进行分析和研究，达到了理论与实际相互结合的目的。     

基于状态反馈的网络分布式拥塞管理方案

提出了一种新型的面向区分服务网络的分布式拥塞管理方案。其基本思想是利用拥塞状态反馈信息在边缘节点或主机上实施拥塞管理，该方案主要包括三个组成部分；拥塞状态控制分组，早期拥塞检测和流量控制算法，实验结果表明，与标准的区分服务网络相比，该方案能在TCP和UDP聚集之间公平地分配带宽并能显著地降低分组丢失率。     

宽带CDMA无线多媒体接入系统的研究

提出了可变传输速率的宽带CDMA扩频调制方法、非平衡功率控制算法以及自适应接入速率的无线接入方法。基于上述技术,给出了由基站和多媒体终端组成的实验系统参数。该系统实现了舆速率分别为8kbp和144skbps的语音和数据通信。     

An Active Queue Management Algorithm： CRED

By applying the method of average and variance, a new queue management algorithm named the Classified-Random Early Detection （CRED） algorithm is presented which can identify the media streaming, TCP traffic and other UDP traffic at the edge routers. The algorithm discriminates the slow start and the congestion control phase of the TCP traffic and combines the TCP congestion control with the IP congestion control to alleviate the congestion effectively. Simulation shows that CRED can not only make the media streaming obtain the resources needed but also protect the TCP traffic transmitted effectively and reliably.     

TCP Yolanda:无线和有线混合网络中的TCP改进算法

在TCP Westwood基础上提出了一种适合无线有线混合网络的TCP改进算法——TCP Yolanda,该算法以瓶颈链路缓冲区长度测量值为参数,采用分两个子阶段的方式改进拥塞避免阶段以更有效进行传输控制,使传输过程能更长时间保持在稳定的高速率状态,并能在误码率较高的无线网络环境下快速恢复拥塞窗口大小,另外在传输发生错误时能判断丢包原因并进行区别处理。基于ns2的仿真实验表明,改进后的TCP算法在网络传输吞吐率、传输稳定性等方面的性能有较大提升,并且保持较好的连接公平性和友好性。     

TCP-friendly congestion control for streaming real-time applications over wireless networks

Congestion control for streaming real-time applications, which need smoothness of the transmission rate, should be transmission control protocol (TCP)-friendly. Moreover, in wireless networks, TCP-friendly congestion control should be based on differentiation of packet losses due to congestion and wireless link error to improve network utilisation. The authors propose a TCP-friendly congestion control algorithm based on explicit congestion notification over the wireless networks. The simulation results show that the proposed algorithm utilises the link bandwidth efficiently, providing smoothness of the transmission rate.     

一种保证TCP上下行时间公平的新调度算法

针对无线局域网中传输控制协议(TCP)闭环拥塞控制的贪婪性、媒体接入控制(MAC)协议节点接入的公平性和无线信道的异构性导致TCP出现的上下行不公平和时间不公平问题,提出了一种在接入点上实现上下行时间公平的UDTFLAS (Up/Down Time Fair LAS)调度算法.UDTFLAS算法依据信道数据发送速率,给无线信道速率较高的TCP流提供更大的发送概率,以保证上下行各TCP流占用相等的无线信道时间.实验结果表明,UDTFLAS算法可以保护单流的吞吐率,实现上下行流的时间公平,能有效提高网络总吞吐量.     

TCPW BR: A Wireless Congestion Control Scheme Base on RTT

The wireless network is limited by the transmission medium, and the transmission process is subject to large interference and jitter. This jitter can cause sporadic loss and is mistaken for congestion by the congestion control mechanism. The TCP Westwood protocol (referred to as TCPW) is such that it cannot distinguish between congestion loss and wireless jitter loss, which makes the congestion mechanism too sensitive and reduces bandwidth utilization. Based on this, the TCPW protocol is modified based on the estimate of the Round-Trip Time (referred to as RTT) value-called TCPW BR. The algorithm uses the measured smooth RTT value and divides the congestion level according to the weighted average idea to determine congestion loss and wireless jitter loss. The simulation results show that the TCPW BR algorithm enhances the wireless network’s ability to judge congestion and random errors.     

WARD: A deterministic fluid model

Queue management, bandwidth share and congestion control are very important to both the robustness and fairness of the Internet. A new TCP-friendly router-based active queue management scheme, called WARD, approximates the fair queueing policy. WARD is a simple packet dropping algorithm with a random mechanism and discriminates against the flows which submit more packets per second than is allowed by their fair share. By doing this, it not only protects transmission control protocol (TCP) connections from user datagram protocol (UDP) flows, but also solves the problem of competing bandwidth among different TCP versions, such as TCP Vegas and TCP Reno. Furthermore, it is stateless and easy to implement, so WARD controls unresponsive or misbehaving flows with a minimum overhead. In this article, we present a deterministic fluid model of TCP/WARD system, and explain the UDP throughput behaviour with WARD. Also, we prove that, provided the number of TCP flows is large, the UDP bandwidth share peaks at (2e)^-1 = 0.184 when UDP input rate is slightly larger than link capacity and drops to zero as UDP input rate tends to infinity.     

WARD: a transmission control protocol-friendly stateless active queue management scheme

In this article, the problem of providing a fair bandwidth allocation to the flows sharing a congested link in a router is investigated. Queue management, bandwidth share and congestion control are very important to both the robustness and fairness of the Internet. The buffer at the outgoing link is a simple FIFO, shared by packets belonging to the flows. A new transmission control protocol (TCP)-friendly router-based active queue management scheme, termed WARD, is proposed to approximate the fair queueing policy. WARD is a simple packet-dropping algorithm with a random mechanism which discriminates against flows that submit more packets per second than is allowed as their fair share. By doing this, it not only protects TCP connections from user datagram protocol flows, but also solves the problem of competing bandwidth among different TCP versions, such as TCP Vegas and TCP Reno. In addition, WARD works quite well for TCP flow isolation even with different round trip times. In other words, WARD improves the unfair bandwidth allocation properties. Furthermore, as it is stateless and easy to implement, WARD controls unresponsive or misbehaving flows with only a minimum overhead.