单源多径路由网络拥塞链路识别

针对多径路由带来的端到端测量路径不确定性以及布尔模型不能很好地解决多拥塞链路的问题,该文在识别端到端测量路径的基础上,提出一种基于扩展状态空间的网络拥塞链路识别算法。首先基于探测流时延相关性进行自适应聚类,进而得到各路径与探测流之间的映射关系。其次采用多门限的方式,将具有不同丢包程度的拥塞路径赋予不同的拥塞状态。最后将拥塞链路识别问题转化为一个约束最优化问题,并提出基于扩展状态空间的拥塞链路识别算法(ESSCLI)算法求解该问题。仿真结果表明,ESSCLI算法能够在多种不同网络场景下取得比当前算法更高的拥塞链路检测率。     

基于局域信息少数者博弈的拥塞控制算法

针对目前已有的拥塞控制算法普遍存在资源利用率低下、RTT不公平和多瓶颈链路不公平等问题,分析了网络拥塞控制中的博弈现象,考虑网络局域信息在拥塞控制过程中的影响,建立了一种基于局域信息少数者博弈的拥塞控制模型,提出并实现了一种基于局域信息少数者博弈的拥塞控制算法。仿真实验表明,该算法在保证瓶颈链路资源利用率的同时,有效地解决了异构RTT流的不公平性问题,实现了多瓶颈链路下的最大最小公平性。     

多个TCP连接的拥塞丢包模型

本文提出了有关TCP连接的拥塞丢包分析模型.网络瓶颈一般承载许多TCP连接,瓶颈处不可避免的拥塞和缓存溢出,是导致网上丢包的主要原因.网络瓶颈处的行为很大程度上左右了网络性能.本文的模型估计了存在大量持续TCP连接时,网络瓶颈的丢包概率和网络传输中断概率,给出了对实际网络的良好近似.这对于研究TCP对网络性能的影响,提出改善网络性能的新算法,以及分析(从长远来看)TCP还应做哪些改进,都是非常有用的.     

无线传感器网络拥塞控制机制研究

本论文在分析无线传感器网络不同于普通有线网络的特征后,提出了一种适用于无线传感器网络的TCP 拥塞控制机制DW-TCP,此机制将TCP拥塞窗口分为拥塞丢包窗口和误码丢包窗口,在链路误码率较高时通过降低发送速率提高数据发送的可靠性,节约无线节点的能源消耗和系统不必要的开销,该机制不但考虑到拥塞丢包对数据发送速率的影响,而且考虑到无线链路中的误码丢包对数据发送速率的影响.     

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。     

适合关键信息可靠传输的节点拥塞避免算法

无线传感器网络节点通信能力有限,有事件发生时,数据产生速率将急剧增大,网络可能会发生拥塞的问题,提出了一种适合关键信息可靠传输的节点拥塞避免算法CAARTKI (Congestion Avoidance Algorithm for Reliable Transmission of Key Information)。算法主要思想是通过引入区分服务,数据包按其重要性不同划分为不同的优先级,高优先级分组优先得到传输。本算法中,节点只有在下一跳节点为其分配了发送窗口才可以发送数据,以避免节点拥塞发生;在链路层考虑拥塞避免的同时,在路由层通过选择可用缓存空间多的邻居节点作为下一跳节点,使关键信息能及时可靠传输到负载较轻的节点,减少由于负载过重时重要信息不能及时传输的可能;在关键信息密集产生时,采用主动丢包策略,丢弃部分低优先级分组,为高优先级分组腾出缓存区间。NS2仿真实验结果表明:CAARTKI可预防拥塞的产生,最高优先级分组的丢包率低,平均网络时延较小,能保证关键信息的及时可靠传输。      

拥塞控制卫星PRMA-AC协议及性能分析

卫星通信中PRMA协议在负荷较重时由于终端竞争加剧会引起信道拥塞,而较长的传播时延更进一步加剧了拥塞引起的丢包。该文提出一种利用话音终端在通话的不同阶段对信道资源的不同需求对终端进行区分的方案PRMA-AC,据此引入一种接入控制机制,以减少信道竞争,提高系统服务质量。文中给出了系统模型,对协议性能进行了理论分析,获得了新协议下的接入阻塞率、丢包率等性能指标,最后通过仿真与几种卫星通信中常用的PRMA协议进行了对比,证明了协议的性能。     

携带虚拟队列长度信息的ABR业务流量控制算法

本文提出了一种新的基于速率检测拥塞的ABR流量控制算法--E-EFCI，它的基本思想是在基于速率检测拥塞的基础上扩展EFCI的功能，使其不仅反映网络阻塞状况，而且还反映网络排队状况。在交换机中通过虚拟队列的长度设置EFCI来传递网络的排队信息。仿真结果表明，本文提出的算法比基本EFCI算法具有更好的控制行为，减小了ACR和排队长度的振荡幅度，提高了链路、缓存利用率，加快拥塞解除，并具有较高的性能价格比。     

基于功率分配的无线传感器网络拥塞控制协议

由于无线传感器网络的资源受限,相比传统网络其传输过程更容易导致丢包和拥塞。为了避免和控制拥塞,本文提出了一种利用功率分配重新选择路径的拥塞控制算法,首先参考路径传输的多种QoS相关参数度量网络的路径代价和节点可用资源的信息,基于测量得到的网络拥塞水平通过传输功率递增和递减选择不同的路径控制网络拥塞。仿真结果表明所提出的协议减少了网络时延,提高了系统吞吐量和网络生命周期。      

无线网络中TCP友好流媒体传输改进机制

为保持无线网络中多媒体业务对TCP的友好性，提出了一种适用于无线网络的动态自适应的流媒体传输速率调节机制。该机制通过在接收端区分网络拥塞丢包和链路错误随机丢包，准确判断网络的拥塞状况结合接收端缓存区占用程度，自适应实施多级速率调节，实现了TCP流友好性和流媒体服务质量（QoS）的折中。由于准确区分出无线链路误码丢包和动态调整流媒体QoS要求，该机制能维持较高的网络利用率。仿真实验结果显示在连接数为2和32，链路误码率从0到0．1变化时TCP,TFRC和吞吐量幅度下降幅度较大，WTFCC幅度下降相对较慢，最大相差达2M；在网络负载重时，尽管链路误码率较低，WTFCC区分链路错误与拥塞丢包，因此，端到端丢包率高于TCP和TFRC，但整体传输吞吐量也高于两者。归一化吞吐量显示WTFCC对TCP流友好。     

A Delay-Aware Congestion Control Protocol for Wireless Sensor Networks

In wireless sensor networks, congestion leads to buffer overflowing, and increases delay. The tradi-tional solutions use rate adjustment to mitigate congestion, thus increasing the delay. A Delay-aware congestion con-trol protocol (DACC) was presented to mitigate congestion and decrease delay. In order to improve the accuracy of the existing congestion detection model which is based on the buffer occupancy of a single node, DACC presents a new model considering both the real-time buffer occupancy and the average transmission time of packets. DACC uses the untapped bits in the IEEE 802.11 Distributed coordination function (DCF) frames header to carry congestion infor-mation. During the congestion alleviation period, DACC presents a channel occupancy mechanism which is based on the real-time buffer occupancy for the purpose of decreas-ing delay and preventing packet loss. Simulation results indicate that in terms of delay, packet delivery ratio, col-lision and buffer load, DACC has comparative advantages than those of 802.11 DCF, Priority-based congestion con-trol protocol (PCCP) and Decoupling congestion control and fairness (DCCF).     

基于无线传感器网络的跨层拥塞控制协议

无线传感器网络(WSN)中由拥塞引起的大量分组重传以及重传多次失败后的分组丢弃会导致较长的时延、较高的分组丢失率和较多的能量消耗.为了准确探测和控制网络拥塞,提出了一种基于跨层设计的拥塞控制协议,即上行拥塞控制(UCC)协议.该协议利用节点在媒质接人控制(MAC)层中未占用的缓冲器区间大小和所预测的通信流量作为该节点的...     

Optimized Packet Scheduling Management: Maximizing Bandwidth Utilization for Next-Generation Mobile Multimedia Communications

Base stations in next-generation broadband mobile networks (NGBMNs) must efficiently schedule different kinds of multimedia packets providing different quality of service (QoS) classes. During the past 10?years, many researchers have experimented with various packet scheduling schemes. In this paper we will propose a batch-arrival queuing model for evaluating NGBMN multimedia packet scheduling systems, and for obtaining three performance measures: packet loss rate (PLR), queuing delay (QD), and bandwidth utilization (BU). The three measures can be used to solve utilization optimization problems with QoS constraints. Specifically, a combination of a traffic statistic plus maximum PLR and QD constraints can be used to maximize BU for a multimedia packet scheduling management architecture. According to results from mathematical tests of the proposed model, it offers an efficient approach to managing scheduling buffers. The model and optimized parameters can be applied to flexible bandwidth deployment and classified buffer size control, thus enhancing profitability.     

Implication of dropping packets from the front of a queue

When congestion occurs in a packet queuing system, packets can be dropped from the rear or the front of the queue. It is demonstrated that the probability of a packet being dropped is the same in systems with rear and front packet dropping. It is shown that the probability of a packet being delayed longer than a given value in a system with front dropping is less than or equal to that in a system with rear dropping. It is further illustrated that front dropping not only improves the delay performance on an internodal link, but also provides the overall loss performance for time constrained traffic such as packet voice     

无线传感器网络中基于拥塞程度分级的速率调节机制

文中提出一种基于拥塞程度分级的速率调节算法.首先,对缓冲区进行多尺度排队分析,计算出缓冲区的溢出概率.其次,根据溢出概率的值,把节点拥塞程度分成三级.最后,针对每一级拥塞采取相应的速率调节方案来缓解拥塞.实验结果表明,该算法可以有效缓解拥塞,提高无线传感器网络的数据包投递率.     

一种基于速率和队列长度的主动队列管理机制

本文研究了拥塞产生的原因及其表现形式,提出了一种基于报文到达速率和队列长度的随机早丢弃算法(RQ).根据拥塞的严重程度和变化趋势将拥塞划分为六个级别,每个级别采用合适的丢弃概率,从而将队列控制在理想工作点附近.利用经典控制理论,分析了系统的稳定性,给出了参数配置的原则.最后,用NS网络仿真器对算法性能进行了验证.     

Queuing Models for Dimensioning Interactive and Streaming Services in High-Speed Downlink Packet Access Networks

We consider modeling the statistical behavior of interactive and streaming traffics in high-speed downlink packet access (HSDPA) networks. Two important applications in these traffic categories are web-browsing (interactive service) and video streaming (streaming service). Web-browsing is characterized by its important sensitivity to delay. Video streaming on the other hand is less sensitive to delay, however, due to its large frame sizes, video traffic is more affected by the packet loss resulting from a limited buffer size at the base station. Taking these characteristics into account, we consider modeling the queuing delay probability density function (PDF) of the Web-browsing traffic, and modeling the queuing buffer size distribution of video streaming traffic. Specifically, we show that the queuing delay of the Web-browsing traffic follows an exponential distribution and that the queuing buffer size of video streaming traffic follows a weighted Weibull distribution. Model fitting based on simulated data is used to provide simple mathematical formulations for the different parameters that characterize the PDFs under consideration. The provided equations could be used, directly, in HSDPA network dimensioning and, as a reference, to satisfy a certain quality of service (QoS).     

Congestion control for multimedia services

The problem of congestion control in high-speed networks for multimedia traffic, such as voice and video, is considered. It is shown that the performance requirements of high-speed networks involve delay, delay-jitter, and packet loss. A framing congestion control strategy based on a packet admission policy at the edges of the network and on a service discipline called stop-and-go queuing at the switching nodes is described. This strategy provides bounded end-to-end delay and a small and controllable delay-jitter. The strategy is applicable to packet switching networks in general, including fixed cell length asynchronous transfer mode (ATM), as well as networks with variable-size packets     

一种基于多径负载均衡的无线互联网TCP拥塞控制策略

针对无线互联网中拥塞控制存在的问题,提出了一种支持多径负载均衡技术的TCP拥塞控制策略。该篆略主要包括拥塞检测和拥塞控制两个部分。其中基于带宽估计的拥塞检测方法,能有效区分差错丢包和拥塞丢包;在检测到拥塞发生后,采用基于多径负载均衡技术的拥塞控制策略,并使用基于时延估计的多径分组分配策略,在缓解拥塞的同时避免了多路通信的包乱序问题,有效的提高吞吐量,减小端到端时延。     

Adaptive Load-Aware Congestion Control Protocol for Wireless Sensor Networks

Congestion control in wireless sensor networks (WSNs) is crucial. In this article, we discuss congestion control and the adaptive load-aware problem for sensor nodes in WSNs. When the traffic load of a specific node exceeds its the available capacity of the node, a congestion problem occurs because of buffer memory overflow. Congestion may cause serious problems such as packet loss, the consumption of power, and low network throughput for sensor nodes. To address these problems, we propose a distributed congestion control protocol called adaptive load-aware congestion control protocol (ALACCP). The protocol can adaptively allocate the appropriate forwarding rate for jammed sensor nodes to mitigate the congestion load. Through the buffer management mechanism, the congestion index of neighboring sensor nodes, and an adjustment of the adaptive forwarding rate, the degree of congestion is alleviated markedly. The performance in allocating the forwarding rate effectively to neighboring sensor nodes also improves. The ALACCP can avoid packet loss because of traffic congestion, reduce the power consumption of nodes, and improve the network throughput. Simulation results revealed that the proposed ALACCP can effectively improve network performance and maintain the fairness of networks.