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.     

联合无线网络编码与TCP Vegas的多播协议优化研究

为解决移动自组网中网络编码多播路由协议因业务传输负载增大,而产生的网络拥塞现象,本文提出了一种可靠的基于TCP Vegas窗口拥塞控制的网络编码多播路由协议。该协议的核心思想是发送节点采用发送窗口自调整和反馈消息触发发送窗口调整的机制,综合的调节数据包的发送速率,来改善网络拥塞现象,从而可以降低丢包率。仿真结果表明,当传输负载增大时,基于窗口拥塞控制的网络编码多播路由协议可使得系统的总开销大大降低,分组投递率获得了相对的提升。     

基于带宽测量技术的TCP拥塞窗口在线学习方法

为了改进传输层协议提高吞吐量,本文提出了一种基于ACK时间间隔的在线学习方法（TCP-Learning）,该方法采用可用带宽测量技术,快速学习网络链路中可用剩余带宽并能够迅速调整TCP拥塞窗口。仿真结果表明,在链路处于良好状态下,TCP-Learning吞吐量略优于Reno和Vegas等传统拥塞控制算法,但在链路较差情况下,TCP-Learning吞吐量明显优于Reno和Vegas等传统拥塞控制算法。     

一种基于速率的组播拥塞控制算法及其性能分析

大部分组播拥塞控制机制都是将包丢失作为网络拥塞的信号,存在丢包、响应速度慢以及由此引起的协议间不公平等缺陷.本文提出了一种新的基于速率拥塞控制算法通过对拥塞的早期检测,进行及时反馈,发送端通过调节数据包的发送间隔进行拥塞避免和控制,使网络能够对拥塞做出快速反应,更有效地利用网络资源.实验结果表明,在相同的配置下,采用该拥塞控制算法的网络在吞吐量、灵敏性和公平性等性能上均优于原先的算法.     

即时恢复--一种新的TCP拥塞控制快速恢复算法

针对TCP连接的网络传输中经常会出现同一窗口多个分组丢失的情况,本文提出了一种新的TCP拥塞控制快速恢复算法:即时恢复算法。即时恢复算法能在恢复阶段探测网络的可用带宽,并将之用于窗口拥塞控制。另外,通过动态设定退出恢复阶段的参数域值,除了能恢复首次丢失的多个分组外,该算法还能很好的解决恢复阶段分组进一步丢失的情况。与SACK、FACK TCP等算法需要对TCP协议的发送、接收部分同时修改不同,该算法只需修改协议的发送部分,而接收部分不需作任何改变。在ns仿真环境的仿真结果表明:在同一窗口有多个分组丢失时,即时恢复算法的性能比TCP New-Reno有较大提高,与 SACK TCP性能相当。     

TCP在不可靠网络环境中的性能分析及其服务质量控制

本文对TCP在拥塞丢失和非拥塞丢失同时存在的不可靠网络环境中的性能进行了理论分析,给出了TCP吞吐量,延时以及分组丢失概率的简明表达式.同时针对多个TCP连接共享一条限速链路时服务质量不均衡问题,提出了轮询丢弃策略和惩罚丢弃策略两种服务质量控制策略.仿真结果表明,轮询丢弃策略可以提高限速链路的总吞吐量,而惩罚丢弃策略可以提高TCP连接之间的公平性.     

基于跳到跳信息的卫星网络传输控制协议研究

设计了一种适用于卫星网络的传输控制协议TPSN。该协议使用异步跳到跳确认,快速恢复成段丢失数据,并采用基于检测窗口的端到端选择性否定应答(SNACK)机制,减少协议控制信息,保证数据可靠传输。在跳到跳可靠性保证机制基础上传输网络负载,实现在避免网络拥塞条件下高效利用网络带宽资源,并保证具有不同端到端往返时延的异种数据流之间的公平性。仿真结果表明TPSN能够在长延迟、误码率高、链路频繁切换的卫星网络中保证高效的带宽利用、可靠的数据传输以及各数据流之间的公平。     

基于价格调整和流量工程的软件定义数据中心网络传输优化方法

分析数据中心网络动态价格调整拥塞控制和流量工程传输优化方法的研究现状,提出结合拥塞控制、流量调度和负载均衡的网络传输优化方法,设计一种新型数据中心网络资源动态价格调整策略框架,构建一种两层结构的数据中心网络拥塞控制价格设置算法,平衡和调节各发送端的购买力和链路带宽收益之间的关系,根据各价格指标设计具有QoS路由技术与调度算法实现资源优化利用的流量工程方法,扩展OpenFlow协议算法,整合SDN、FAST协议、ECN三种技术,从运行机制方面找到提高数据中心网络传输性能的方法.     

MP-Start：基于带宽测量的分阶段TCP慢启动机制

针对TCP慢启动机制门限值参数难于设置以及突发流量引起网络抖动的问题，提出了一种基于带宽测量及分阶段实施的TCP慢启动改进算法。该方法利用在线网络带宽测量技术，探测出网络带宽，从而根据网络状态的动态变化实现慢启动门限值的动态更新，同时，采用分阶段方法调节拥塞窗口的增加，在拥塞窗口大于门限值一半时，拥塞窗口增幅为门限值和拥塞窗口差值的一半，逐步迭代逼近门限值，使拥塞窗口在连接启动和过渡到拥塞避免阶段增加幅度比较小，而在中间阶段窗口增加幅度比较大。有效避免了多个分组丢失现象的发生，实现了连接的平滑接入和过渡到拥塞避免阶段；为提高改进算法的性能，设计了一种灵活的参数化的模型和自适应参数设置方案。大量仿真实验结果表明MP-Start能在多种网络环境下有效改善网络性能，如降低分组丢弃数、减少数据传输时间、降低共享链路分组队列长度、保持连接的传输平滑性。     

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。     

Equation-based end-to-end single-rate multicast congestion control

Among the recently proposed single-rate multicast congestion control protocols is transmission control protocol-friendly multicast congestion control (TFMCC; Widmer and Handley 2001; Floyd et al. 2000; Widmer et al. IEEE Netw 15:28–37, 2001), which is an equation-based single-rate protocol that extends the mechanisms of the unicast TCP-friendly rate control (TFRC) protocol into the multicast domain. In TFMCC, each receiver estimates its throughput using an equation that estimates the steady-state throughput of a TCP source. The source then adjusts its sending rate according to the slowest receiver within the session (a.k.a., current-limiting receiver, CLR). TFMCC is a relatively simple, scalable, and TCP-friendly multicast congestion control protocol. However, TFMCC is hindering its throughput performance by adopting an equation derived from the unicast TFRC protocol. Further, TFMCC is slow to react to congestion conditions that usually result in a change of the CLR. This paper is motivated by these two observations and proposes an improved version of TFMCC, which we refer to as hybrid-TFMCC (or H-TFMCC for short). First, each receiver estimates its throughput using an equation that models the steady-state throughput of a multicast source controlled according to the additive increase multiplicative decrease (AIMD) approach. The second modification consists of adopting a hybrid sender/receiver-based rate control strategy, where the sending rate can be adjusted by the source or initiated by the current or a new CLR. The source monitors RTT variations on the CLR path, in order to rapidly adjust the sending rate to network conditions. Simulation results show that these modifications result in remarkable performance improvement with respect to throughput, time to react, and magnitude of oscillations. We also show that H-TFMCC remains TCP-friendly and achieves a higher fairness index than that achieved by TFMCC.     

Research on Throughout with SWIPT in Two-Way Relay System Under Power Splitting Receiver

We consider the Simultaneous Wireless Information and Power Transfer strategy optimization in the two-way transmission relay system under Power Splitting (PS) receiver. In such case, we propose two operational strategies for two-way transmission under PS receiver structure, (1) Delay-Limited-Relay strategy (PS-LDR) and (2) Delay-Tolerated-Relay strategy (PS-TDR). PS-LDR strategy is that both sending node and receiving node need synchronization of information transmission, so communication systems have higher Signal Noise Ratio (SNR) threshold for channels, or communication may break down. In order to explore the performance of PS-LDR strategy, we use outage probability to calculate the analytical expression of the actual maximum throughput (AMT). PS-TDR strategy allows that the information received by the receiving node delays with the information sent by the sending node. Communication system in PS-TDR strategy has lower SNR threshold than that in PS-LDR strategy. The analytical expression of the AMT is calculated by the ergodic capacity in PS-TDR strategy. The influences of system parameters on the optimal throughput in two strategies are discussed.

     

On MAC optimization for large-scale wireless sensor network

The media access control (MAC) performance of a large-scale wireless sensor network (L-WSN) determines the efficiency of the wireless communication channel. A good MAC protocol could reduce network energy consumption and network delay, which are two problems to be solved urgently in L-WSN. In this paper, we proposed a multi-level integrated MAC protocol (MI-MAC) to solve the overall performance optimization problem of L-WSN. Compared with other protocols, MI-MAC has two mainly improved performances: (1) It improved binary exponential backoff algorithm by twice back off strategy; (2) It designed a sending and receiving algorithm based on the threshold value to recognize control frames (small frames), which effectively avoids the collision probability of data frame. The simulation results show that the MI-MAC protocol improves network throughput and delay performance, significantly reduces energy consumption, and obtains overall network optimization.     

On Weldon's ARQ Strategy

In order to maximize the throughput in a new ARQ strategy proposed by Weldon, a number of parameters need to be selected optimally. An efficient method for choosing these parameters is obtained by exploiting the form of a simplified expression for the throughput. It is also shown that for noisy, long delay channels, the throughput of the Weldon scheme can be increased by sending multiple copies of each new data block.     

一种通过节点序寻优进行贝叶斯网络结构学习的算法

针对K2算法过度依赖节点序,遗传算法节点序寻优效率差的问题,该文提出一种直接对节点序进行评分搜索的贝叶斯结构学习算法。该算法以K2算法为基础,首先通过计算支撑树权重矩阵,构建能够定量评价节点序的适应度函数。然后通过提出混合交叉策略和孤立节点处理机制,同时利用动态学习因子和倒置变异策略,提升遗传算法节点序寻优的性能。最后将得到的节点序作为K2算法的先验知识得到最优贝叶斯网络结构。仿真结果表明,该方法解决了K2算法依赖先验知识的问题,相比于其它优化算法,评分值平均增加了13.11%。     

无线电报务训练装置设计

针对以往无线电报务训练装置硬件成本要求高和码字识别率低等问题,设计了一种基于PC的无线电报务训练装置,它具有发报和收报训练功能,对码字识别效率较高。同时,在码字识别算法上,采用了快速傅里叶变换(FFT)算法对装置进行频域识别,有效提高了抗干扰性。分别采用最大类间方差法和机器学习法,有效提高了码字识别的适应能力和准确性。     

Optimal server bandwidth for mobile video on demand

Multimedia services such as video-on-demand or Internet protocol television in mobile environments have established themselves in our daily lives, yet the obtained quality of service still leads to many open issues. One of them consists in minimizing the server bandwidth, and we recently proposed a novel patching scheme for transporting true video-on-demand called Hierarchical Patching, which minimizes the server bandwidth. In this paper, we present a new concept called Low Start, consisting of encoding the first part of a movie with a lower bitrate than the rest. In Hierarchical Patching, video parts at the beginning have a much higher probability to be transmitted than parts at the end. By using Low Start, we show that the overall server bandwidth can be drastically reduced. We furthermore investigate the impact of Low Start on the subjective quality of service as perceived by human observers, and show that, for mobile video, the optimal strategy is to encode a very short start time with a bandwidth as low as possible.     

基于"联姻"策略的并行遗传算法

早熟现象一直是遗传算法(GAs)所面临的主要问题.为了防止单一种群近亲繁殖导致早熟,本文将人类的联姻策略应用于遗传算法中,提出了一种基于"联姻"策略的并行遗传算法. 对比实验显示出该算法极其优异的性能.同时,文中尝试了一种新的遗传算子,为遗传算法的硬件实现奠定了基础.