An adaptive strategy based on linear prediction of queue length to minimize congestion in Barabási-Albert scale-free networks

In this paper, we propose an adaptive strategy based on the linear prediction of queue length to minimize congestion in Barabási-Albert (BA) scale-free networks. This strategy uses local knowledge of traffic conditions and allows nodes to be able to self-coordinate their accepting probability to the incoming packets. We show that the strategy can delay remarkably the onset of congestion and systems avoiding the congestion can benefit from hierarchical organization of accepting rates of nodes. Furthermore, with the increase of prediction orders, we achieve larger values for the critical load together with a smooth transition from free-flow to congestion.     

竞争环境下供应链的BOPS合作策略及其对交通拥堵改善的影响

针对由一个拥有线上直销渠道的制造商以及一个线下零售商组成的供应链,考虑消费者对不同渠道的偏好构建了供应链博弈模型,研究了竞争环境下制造商与零售商的最优BOPS合作策略,并分析了BOPS合作在改善交通拥堵方面的作用。研究表明,BOPS合作在缓解供应链双重边际效应的同时会加剧渠道之间的价格竞争,所以并不总是对供应链有利;仅当消费者线上渠道偏好程度比较低且线下麻烦成本比较大时,制造商和零售商才会都从BOPS合作中受益从而自愿达成合作,其他情形下制造商可根据供应链利润变化设计转移支付契约促成与线下零售商的BOPS合作并实现双赢;此外,一定条件下BOPS合作能够通过调整渠道需求结构,在保证供应链成员利润的同时对交通拥堵改善也带来正向作用。     

基于Internet的实时视频流的应用层QoS控制策略

分析了Internet上实时视频传输的特点，提出了基于Internet的实时视频流的应用层QoS控制策略，主要包括拥塞控制策略和差错控制策略以及相应的控制技术。在拥塞控制中，讨论速率控制和速率整形，速率控制主要是根据网络运行状态预测当前可用的带宽，并根据预测值调整视频速率，达到与可用带宽匹配；速率整形则是迫使发送端以码率控制算法规定的码率发送视频流。在差错控制中，则讨论了编码器差错复原、解码器错误隐藏和编码器／解码器交互的差错控制等控制策略。这些控制技术应用于终端系统并不需要路由器和网络的QoS支持，可以最大限度地提高视频质量。     

Experimental Investigation on Transmission Control Protocol Throughput Behavior in Optical Fiber Access Networks

ABSTRACT

This article describes an experimental investigation on the behavior of transmission control protocol in throughput measurements to be used in the verification of the service-level agreement between the Internet service provider and user in terms of line capacity for ultra-broadband access networks typical of fiber-to-the-x architectures. It is experimentally shown different conditions in high bandwidth-delay product links where the estimation of the line capacity based on a single transmission control protocol session results are unreliable. Simple equations reported in this work, and experimentally verified, point out the conditions in terms of packet loss, time delay, and line capacity, that allow consideration of the reliability of the measurement carried out with a single transmission control protocol session test by adopting a suitable measurement time duration.     

一种自适应的混合型无线NoC拓扑结构

针对传统大规模片上网络（Network-on-Chip,NoC）远距离核间多跳通信所带来的高能耗与延时问题,提出了一种基于虚Torus的自适应的混合型无线NoC拓扑结构（VT-AWiNoC）．该结构通过引入链路拥塞测度作为感知参数,基于此采用热点无线链路自动探测与带宽动态分配机制,并设计实现发送器动态分配的控制电路模块,以达到根据不同的通信流量模型,于片内自适应地调整拓扑结构及链路带宽的目的．通过建立混合型无线NoC的延时与功耗评估模型,对该结构的无线NoC进行性能评估．实验结果表明,该自适应拓扑与其它混合型无线NoC相比,在随机流量模型下,网络平均延时降低了16．52～23．27％;在20％的热点流量模型下,包平均能耗节省了39．19％;以真实应用FFT作为基准测试,平均延时降低了17．20％～21．68％,并节省了23．49％的包平均能耗．该结构以较小的面积开销获得了更优的性能．     

一种对用户公平的核心无状态队列管理算法

在对"流"公平的CSFQ(Core-Stateless Fair Queueing)算法基础上进行改进,将算法设计思想由对流速率的公平分配改为对表征用户需求的效用函数的公平分配,从而提出一种对用户公平的核心无状态队列管理算法(Core-Stateless User Fair Queueing,CSUFQ),该算法能近似实现边界及核心路由器上对用户需求的最大最小公平,仿真实验证明了其效果.     

在Ad Hoc网络中TCP-SACK性能研究及改进

在Ad Hoc网中,经常发生链路失效和路由变化。TCP把数据包的丢失归结为链路拥塞,所以TCP在Ad Hoc网中表现很差。在本文中通过应用一种新的方法,基于传输包乱序检测与响应方法,研究TCP在Ad Hoc网中的性能。通过实验结果表明,此算法解决了,提高了网络的吞吐量。     

基于控制论的ATM拥塞控制算法

ATM网络的多业务复用特性及其对QoS的承诺,使得其流量和拥塞控制变得非常复杂。控制论是分析和设计复杂系统的强有力工具,文章运用控制理论闭环反馈理论,设计了PI调节器,实现了基于PI调节器的ATM拥塞控制算法,实验及理论分析结果表明,该算法具有较强的稳定性与鲁棒性。     

Measured average cell rate‐based congestion avoidance scheme

Techniques for congestion control of available bit‐rate (ABR) traffic in ATM (asynchronous transfer mode) networks remain an important issue. Several congestion control schemes have been proposed to adjust the cell rates of sources with a modified or mean allowed cell rate. To make these schemes work effectively in practice, the modified or mean allowed cell rate must converge under all conditions. However, it is not easy to obtain an accurate value, and an inaccurate value could result in network performance degradation such as severe oscillations and considerable unfairness. Therefore, we propose a measured average cell rate‐based congestion avoidance for ABR traffic in ATM networks. The scheme has high throughput and achieves shorter queue lengths without congestion. With measured average cell rate, the scheme provides fast convergence to a start‐up virtual connection (VC) and rate of equalization from different initial conditions of the sources. Thus, this scheme provides better fairness among connections. Copyright © 2001 John Wiley & Sons, Ltd.     

Adaptive ABR congestion control with active buffer control

ATM Forum has defined that the ABR service is designed mainly for data traffic. The design of a simple and efficient congestion control scheme is a problem that network managers have to face. In this paper, we propose a control scheme for ABR service for better scalability and response. The scheme uses both the load level and an increase in queue length to detect the presence of congestion. Continuous control functions are designed for the bandwidth allocation. A better buffer control is achieved by incorporating the percentage of buffer occupancy into the control function. All of the control functions are designed in such a way that they are sensitive to the change of the system states when they are far from the optimal level to get a fast response, but insensitive and change slowly when the states are close to the optimal level to keep the system around the steady state. Our simulation results show that the algorithm is fair to all the connections and it converges fast when the network's condition changes. Copyright © 2001 John Wiley & Sons, Ltd.