无线自组网络中TCP流公平性的分析与改进

研究了TCP(transmission control protocol)流在多跳无线自组网络中的公平性问题,发现IEEE802.11DCF协议在此环境下会导致严重的不公平性,即部分节点垄断了网络带宽而其他节点被饿死.首先,通过仿真分析了产生TCP流不公平性的原因,指出其根源在于MAC(media access and control)协议的不公平性,同时,TCP的超时机制加剧了不公平性的产生;然后,利用概率模型定量分析了TCP不公平性与MAC协议参数之间的关系,发现TCP流的公平性与TCP报文长度直接相关,并且增加MAC协议初始竞争窗口的大小能够有效提高公平性.据此,提出了一种根据TCP报文长度动态调节初始回退窗口大小的自适应回退MAC协议改进算法.理论分析和仿真表明,该算法在很大程度上可以有效缓解不公平性问题的产生,并且不会引起网络吞吐量的严重降低.     

无线自组织网络中TCP稳定性的分析及改进

在无线自组织网络中,基于IEEE 802.11的TCP流存在严重的不稳定性,其原因与MAC协议、路由协议和TCP本身均有一定的关系,但最根本之处在于MAC协议的不公平性以及假的链路失效消息导致了不必要的耗时的路由发现过程.结合IEEE802.11的MAC协议和DSR路由协议,对这些原因进行了深入的理论分析和仿真实验,并提出了针对MAC协议和路由协议的改进算法.仿真结果证明,提出的改进算法不仅能基本上避免TCP流的不稳定性,还能够极大地提高TCP流的平均吞吐量.     

基于TCP拥塞控制的改进Backpressure算法

本文研究了无线网络中在Backpressure算法路由调度下TCP数据流的性能表现,发现该算法无法有效提高网络吞吐量,原因是Backpressure算法基于队列差值的路由调度与TCP的拥塞控制机制之间的不匹配造成的。本文提出了一种适应于TCP数据流调度的改进Backpressure算法(T-BP),并从理论上证明了T-BP算法具有throughput-optimal性能。仿真结果表明,在多种网络拓扑结构下,与传统BP算法相比,TCP数据流在吞吐量和公平性方面的性能得到了有效的提高。     

稳定裕度与网络特征参数无关的AQM算法

针对现有基于控制论的主动队列管理(AQM)算法的不足,设计一种稳定裕度与网络特征参数(往返时延、通过瓶颈链路的TCP连接数、瓶颈链路的容量)无关的基于PI控制器的AQM算法GPM-PI。该算法响应速度快、计算开销小,能用于大时滞网络环境,抗干扰能力强(对TCP短流和UDP流有良好的控制能力),鲁棒性好,能更好地适应Internet环境下的主动队列管理。     

具有最小时间加权平方误差积分的主动队列管理算法ISTE-PI

针对现有基于控制论的主动队列管理(AQM)算法的不足,在不忽略TCP/AQM传递函数模型中时滞环节的条件下,将TCP短流和UDP流作为干扰信号引入AQM控制器的设计过程,并借助最小"时间加权平方误差积分(ISTE)"控制方法,设计了一个基于PI控制器的AQM算法ISTE-PI.与现有算法相比,ISTE-PI算法同时具备响应速度快、抗干扰能力强、能用于大时滞网络环境且鲁棒性好等特点.鉴于Internet网络流量由TCP长流、TCP短流和UDP流构成,且是大时滞网络,因此ISTE-PI算法能更好地适应Internet网络环境下的主动队列管理.     

提升无线网络TCP公平性的跨层机制研究

主要研究多跳无线Ad Hoc网络中TCP业务的公平性问题,分析了造成TCP业务不公平的多种原因,并验证MAC层的介质访问延迟能够反映出无线信道的拥塞程度,在此基础上提出基于MAC层传输延迟的邻居区域随机提前检测机制,即Mac-layer Delay and Neighborhood Random Early Detection (MDNRED).该机制主要包括四部分内容:测量MAC延迟;判断拥塞发生位置;发出拥塞通知;节点计算丢包概率.仿真试验证实,在特定场景下MDNRED机制能够在保证一定吞吐率的基础上有效提升各条TCP流的公平性.     

一种采用公平策略的CSFQ算法

研究一种对核心无状态公平队列调度(CSFQ)算法的改进方法.针对CSFQ缓存资源管理的弊病,造成路由器突发性流响应差及对响应流TCP和非响应流UDP不公平的问题,提出了一种公平策略的CSFQ算法(FCSFQ).FCSFQ提出了基于共享存储器模式的动态缓存管理机制,并以此来改善CSFQ突发性响应差问题,同时采用对响应流与非响应流不同的丢包策略,来解决CSFQ对不同数据流不公平问题.NS仿真实验表明,在现实网络环境下,FCSFQ算法对突发性响应及带宽在TCP、UDP数据流之间的公平分配都有明显改善.     

Ad-hoc网络PSD拥塞控制算法

神经元PID算法能较好地控制队列长度,但其神经元增益对被控对象的状态较为敏感,基于试凑和经验的设定往往使控制效果难以保证。基于TCP拥塞窗口加法增大、乘法减小原则和排队机制,推导出拥塞窗口与丢弃概率、队列长度的微分方程,再对方程进行线性化,获得Ad-hoc网络TCP/AQM控制系统模型。基于该模型,将递推计算修正功能引入神经元PID,设计了一种神经元自适应PSD的AQM。该算法可以在线调整神经元增益。NS仿真表明,在无线分组丢失、突发流及链路容量变化的Ad-hoc网络中,PSD队列管理性能优于神经元PID。     

高吞吐量的核心无状态公平队列算法

提出了一种优化的核心无状态公平队列凋度算法(xCSFQ),在CSFQ的基础上,根据缓冲区占用率和数据流到达速率决定丢包概率,缓存管理上采用基于CHOKC原理的机制进行缓存管理,解决了CSFQ链路利用率低的问题,提高了带宽在UDP流和TCP流之间分配的公平性,最后对算法进行了仿真分析。     

基于带宽充分利用的TCP友好拥塞控制机制

当基于TCP的应用和基于UDP的应用共享网络时，由于UDP协议没有拥塞控制机制，为了防止基于UDP的应用占尽所有的带宽，UDP应该具有TCP友好性(TCP_friendly)。提出一种机制以解决该问题。在保证UDP数据流的发送速率不大于TCP数据流的发送速率的基础上，利用剩余可用的带宽发送best_efort数据流，在网关采用改进的支持包标记处理的RED(随机早期检测)队列管理机制对不同流量进行相应的处理。     

TCP performance optimization in multi-cell WLANs

Though significant attention has been given to understanding the performance of a single-cell WLAN, performance evaluation of a group of interfering basic service sets (BSSs) within an extended service set (ESS) is still an open area. In this paper, we first demonstrate that a severe throughput imbalance occurs between downlink TCP flows even in the simplest of multi-cell WLANs via simulation and real world experiments; then, to solve this unfairness problem, we derive an analytical model that describes the interaction between TCP flows at the MAC layer, and formulate a throughput allocation problem as a nonlinear optimization problem subject to certain fairness requirements. Our formulation considers real world complexity such as hidden terminals, packet transmission retry limit, and the unique characteristics of TCP traffic. Solving our optimization problem yields the optimal MAC layer contention window settings that can lead each TCP flow to its target end-to-end throughput without the need for any per-flow queuing nor modification of the TCP sender. Simulation results show that our approach can achieve a fair allocation on the end-to-end throughput and attest to the accuracy of our proposed method.     

An accurate closed-form formula for the throughput of long-lived TCP connections in IEEE 802.11 WLANs

There is a vast literature on the performance modeling of the 802.11 MAC protocol, but the interplay between the TCP dynamics and the self-regulating behavior of the 802.11 CSMA/CA access method has not been sufficiently investigated. In addition, it has been observed that TCP stations in a WLAN are sporadically active due to the TCP flow control mechanisms. This makes traditional saturated models of the network capacity of 802.11 WLANs not very accurate. We presented a rigorous analytical model to calculate the distribution of the number of active TCP stations in a WLAN with multiple long-lived TCP flows [R. Bruno, M. Conti, E. Gregori, Performance modelling and measurements of TCP transfer throughput in 802.11-based WLAN, in: Proceedings of the IEEE MsWiM 2006, Torremolinos, Malaga, Spain, 2006, pp. 4–11, [1]]. Starting from this analysis, in this paper we derive a simple but accurate closed-form expression of the per-connection TCP throughout as a function of the average duration of collisions, the average backoff period and the TCP packet size. We validate this formula through performance tests carried out on a real WLAN. Then, we use our model to mathematically study the optimal minimum contention window size to maximize the TCP throughput. Our analytical results indicate that the initial window size specified in the 802.11b standard is about two times larger than the optimal one. However, we also show that the TCP throughput performance is not very sensitive to changes of the minimum contention window size, and that the TCP flows significantly underutilize the channel bandwidth also in the optimal MAC operating state.     

无线局域网MAC机制对TCP性能的影响分析

通过分析和比较IEEE802.11无线局域网分别采用3种典型MAC机制时的TCP性能,深入研究了MAC机制对无线局域网中TCP性能的影响.仿真结果表明,由MAC机制所决定的信道冲突率、MAC帧丢失率、MAC帧传输时延抖动幅度和频率、网络吞吐量和公平性将直接影响无线局域网中TCP的吞吐量、公平性和稳定性.     

Improving TCP performance over wired–wireless networks

Today there is a growing demand for Internet services over WLAN hotspots. Majority of the Internet services today are based on TCP. However, TCP is not well-suited for heterogeneous networks consisting of wired and wireless networks. Losses in wireless network, which are quite frequent, are often misinterpreted by the TCP sender as loss due to congestion, which leads to poor performance of TCP. Hence, it is important to shield the TCP sender from wireless error. In this paper, we propose an improved method for identifying cases of wireless losses and tackling the loss at the local link level through MAC layer retransmissions. We then evaluate the effect of MAC layer retransmissions on the performance of TCP both analytically and empirically. Our empirical results show that significant improvement in TCP performance is possible through MAC layer retransmission.     

A link layer adaptive pacing scheme for improving throughput of transport protocols in wireless mesh networks

In this paper, we study the performance of a static multihop wireless network, specifically that of the backhaul network of a two-tier Wireless Mesh Network (WMN) operating on IEEE 802.11 Medium Access Control (MAC) protocol. The performance of an IEEE 802.11 based backhaul network is greatly affected by the MAC contention and congestion in the network. If the sources pump data into the network than can be supported, loss rate increases due to MAC contention and congestion in the network. This also leads to the problem of unfairness among flows. In this paper, we propose a Link Layer Adaptive Pacing (LLAP) scheme that adaptively controls the offered load into the network. This improves the performance of higher layer protocols without any modifications to them. Our LLAP scheme estimates the four hop transmission delay in the network path without incurring any additional overhead (Control packets) and accordingly paces the packet transmissions to reduce MAC contentions in the network. We implement the LLAP scheme in ns-2.29 network simulator and extensively study its performance for both User Datagram Protocol (UDP) and Transmission Control Protocol (TCP) traffic in different network scenarios. In all the cases, our scheme shows a significant improvement in the performance of both UDP and TCP traffic.     

无线自组织网络中的IEEE802．11 MAC协议的研究

IEEES02．11 DCF作为无线局域网的MAC层规范被广泛应用于无线自组织网络的仿真和测试中。通过研究发现即使在拓扑不变的条件下，使用表驱动路由协议DSDV，TCP流仍然存在不稳定和不公平现象，本文分析了问题产生的原因，提出了基于网络负载的解决问题方法。     

利用"捎带"技术提高HFC网络TCP性能的方法

作为一种有效的宽带接入方式，HFC的应用越来越广泛，但由于HFC信道是非对称信道，会造成TCP应用性能的恶化，有关文献对HFC上TCP性能进行了研究，给出了在IP层的改进算法，该文在此算法基础上，通过对HFC MAC层协议的分析，提出了利用HFC的AC 协议中的“捎带”技术来减少确认帧在上行信道的传输延时，从而进一步提高HFC网络上的TCP性能。     

一种非TCP友好流惩罚算法

为了改善Internet中各流量带宽分配的不公平性,提出一种适于缓存管理的非TCP友好流的惩罚算法（NFTFP算法）。通过在缓冲区记录每个连接所占用的带宽,对于非TCP友好流量进行识别和惩罚,从而提高不同数据流享用带宽的公平性。仿真实验表明在标准TCP流量和非TCP友好流共存的环境下,NFTFP算法能提高对非TCP友好流的惩罚力度,在保持较高的链路利用率的同时,获得比RED算法更好的公平性。