基于时延传感器网络的预测融合算法研究*

现有的基于网络时延的融合算法大都直接套用传统的同步数据融合方法,因此会产生信息等待、资源闲置以及差的实时性能等问题.针对上述问题,在现有工作基础上,结合预测估计以及顺序加权融合技术,设计出了一种新的能适应网络时延的多传感器预测加权融合算法.该算法不仅能很好地解决现有基于时延的数据融合算法存在的诸多弊端,同时获得了良好的实时预测功能,并给出了新的基于网络时延的预测加权融合算法的推导过程,通过计算机仿真算例和理论分析来显示新算法的实用性和优越性.     

基于分布式控制算法的泵站水闸网络预测的研究

针对目前我国现代泵站水闸综合自动化控制信息系统的发展现状,通过研究分布式系统中的时延和网络阻塞问题,提出了一种基于支持向量回归(ε-SVR)预测的分布式算法模型,该算法能够有效的预测时延变化,预测效果超过同类型的预测方法。实验和应用表明:与同类型预测方法相比,该模型不但能够在预测效果上达到现有预测水平,而且能大大提高预测速度,适应实时预测的需要。     

基于支持向量回归（ε-SVR）预测的分布式机器视觉算法的研究

针对目前我国现代分布式机器视觉系统的发展现状,通过研究分布式视觉系统中的时延和网络阻塞问题,提出了一种基于支持向量回归（ε-SVR）预测的分布式机器视觉算法模型,该算法能够有效的预测时延变化,预测效果超过同类型的预测方法。实验和应用表明：与同类型预测方法相比,该模型不但能够在预测效果上达到现有预测水平,而且能大大提高预测速度,适应实时预测的需要。     

多机器人系统传感器融合技术应用研究

当前应用于机器人的多传感器数据融合研究大都集中在机器人自主导航和定位问题上,很少涉及利用机器人进行目标跟踪的研究。现有的基于网络时延的融合算法大都直接套用传统的同步数据融合方法,因此会产生数据等待、资源浪费以及较差的实时性能等问题。本文提出结合预测估计以及递推加权融合技术,设计出一种新的能适应网络时延的多传感器预测加权融合算法。该算法不仅能很好的解决现有基于时延的数据融合算法存在的诸多弊端,而且拥有良好的实时预测功能。文中给出了新算法的推导过程,并通过计算机仿真算例来显示新算法的实用性和优越性。     

基于随机神经网络的多步网络时延预测模型

网络时延的动态变化反映了网络路径的负载特征,对时延的精确预测是实施网络拥塞控制、路由选择的重要依据,建立了基于随机神经网络的时延预测模型,该模型克服了传统时间序列预测方法受随机干扰因素影响大、模型结构辨识过程繁琐,以及传统神经网络预测方法易于陷入局部极值、偏离全局最优的缺点.仿真实验表明,在提前单步和多步的预测中该模型比AR模型、RBF神经网络预测算法的准确度更高.     

预估控制下的实时网络遥操作移动机器人

构建了能使操作者通过Internet远程实时控制的移动机器人系统．为了补偿网络时延和抵消其对遥操作系统的影响，基于我们以前提出的改进型Smith预估器原理，采用了预估控制策略．为了保证系统稳定性和透明性，基于主从端的传感器信息交换，设计了一个动态模型管理器，其中模型和力反馈误差调节通过模糊控制实现．除了力反馈外，为了增强遥操作的实时性，引入了预估的虚拟显示．为了精确地预测网络时延，提出了一个新颖的时钟同步算法．为了降低时延抖动，结合我们提出的两个算法，实现了数据缓冲策略．最后，通过长距离的网络遥操作实验验证了系统和控制策略的实用性和有效性．     

支持反三角嵌入的松弛度量空间网络坐标算法

网络坐标是根据网络测量时延将节点嵌入虚拟度量空间,根据坐标位置预测节点时延的模型和算法.然而通过网络测量发现互联网中广泛存在非对称和反三角(Triangle Inequality Violation,TIV)时延,度量空间固有的对称性和三角形不等性在本质上无法体现互联网时延的真实特征.特别是对利用反三角属性建立优化中继路由的应用,传统的网络坐标算法无法支持.为此,该文提出一种基于松弛度量空间的网络坐标模型,利用KL散度距离替代欧式距离定义节点时延,并设计了分布式网络坐标算法KL_vivaldi.通过理论分析与仿真表明:算法支持非对称时延和TIV的预测,提高了时延的预测精度,拓展了网络坐标算法的应用领域.     

时延和时延差别受限的最大带宽多播路由分布式算法

本文采用反映网络实时特性的可用带宽代替代价作为第一度量，提出一种基于最大可用带宽路径且满足时延和时延差别约束的QoS实时多播路由分布式启发算法，该算法具有多项式复杂性，并通过分析得到每路径时延和二约束度量之间的关系，有效降低涉及时延和时延差别此类问题的复杂性。仿真实验证明，该算法具有较好的带宽性能。     

基于核方法的移动机器人远程控制时延预测与分析*

构建了基于Internet的移动机器人控制平台,针对该平台中网络时延的非线性特性,提出了利用线性核和RBF核方法对网络时延序列样本进行回归建模和预测。比较了核方法、BP神经网络和RBF神经网络算法预测误差,表明在移动机器人远程控制中,核方法对网络时延序列预测具有更好的函数逼近能力和较高的预测精度,从而可根据预测时延调节移动机器人控制指令和状态信息的网络传输,保证系统的可靠性,提高移动机器人的控制性能。     

面向网络诱导时延和数据包丢失补偿的网络化广义预测控制

针对网络化控制系统中存在的网络诱导时延和数据包丢失, 考虑了网络化广义预测控制问题. 基于多个数据打包传送的通讯方式以及网络诱导时延和数据包丢失预先可知的假设, 提出了一种采用最小预测步长和预测控制增量向量分别补偿网络诱导时延和数据包丢失对系统性能影响的新方法, 给出了相应的网络化模型预测算法和网络化滚动优化算法, 对于被控对象参数未知或缓慢变化的情况, 给出了基于递推最小二乘辨识改进算法的网络化反馈修正算法, 通过仿真验证了所提出网络化算法的有效性.     

Neural network prediction model for a real-time data transmission

Both the real-time transmission and the amount of valid transmitted data are important factors in real-time multimedia transmission through the Internet. They are mainly affected by the channel bandwidth, delay time, and packet loss. In this paper, we propose a predictive rate control system for data transmission, which is designed to improve the number of valid transmitted packets for the transmission of real-time multimedia over the Internet. The one-step-ahead round-trip delay time and packet loss are predicted using a prediction algorithm and then these predicted values are used to determine the transmission rate. A real-time multimedia transmission system was implemented using a TCP-friendly algorithm, in order to obtain the measurement data needed for the proposed system. Neural network modeling was performed using the collected data, which consisted of the round-trip time (RTT) delay and packet loss rate (PLR). In addition, the performance of the neural network prediction model was verified through a validation process. The transmission rate was determined from the values of RTT delay and PLR, and a data transmission test for an actual system was performed using this transmission rate. The experiment results show that the algorithm proposed in this study increases the number of valid packets compared with the TCP-friendly algorithm.     

网络时延预测研究

网络时间延迟是影响网络控制系统性能的一个主要因素,提前知道网络时延对提升网络控制系统的性能有一定的重要作用,网络时延预测准确度的高低直接影响到网络控制系统性能,为了更好的预测数据在网络上传输的时间延迟,满足网络控制系统需要,该文针对互联网中网络时延的预测问题进行了分析,分别用AR模型和Elman神经网络预测网络时延,通过仿真表明,平稳时延的预测AR模型要稍好,但扰动时延的预测Elman神经网络预测准确度及自适应性优于AR模型。     

面向出租车出行规律的预测式可视分析方法

针对传统交通数据可视分析方法缺乏预测分析能力的问题,提出了基于出租车出行数据的预测式可视分析方法,支持用户更有效地探索未来的交通状况.在可视分析模型中,提出了结合天气、星期几等多种非交通因素的预测模型,提高了预测的准确度;提出了基于预测数据和广义地点类型约束的路径规划方法,获得了更优的路径规划结果;以多种可视化手段分析和预测了出租车司机的运营状况,帮助司机进行运营决策.以温州市出租车数据进行的实验结果表明,与传统方法相比,文中方法能更准确地预测交通状况和运营状况,并获得更合理的路径规划结果.     

实时多播流的弹性公平性和基于门限的拥塞控制策略

该文提出了关于实时IP多播流基于速率弹性的公平性定义，并结合有关多播组规模估计的机制和多播拥塞控制算法。实现了一种基于门限的实时多播业务的等级拥塞控制策略，使实时多播流在满足瞬态弹性公平性的同时，也基本满足稳态的比例公平性，并对此进行了仿真验证。     

Oscillation control for n-dimensional congestion control model via time-varying delay

Time delay or round trip time (RTT) is an important parameter in the model of Internet congestion control. On the one hand, the delay may induce oscillation via the Hopf bifurcation. In the present paper, a congestion control model of n dimensions is considered to study the delay-induced oscillation. By linear analysis of the n-dimensional system, the critical delay for the Hopf bifurcation is obtained. To describe the relation between the delay and oscillation analytically, the method of multiple scales (M...     

Score-based Incentive Mechanism (SIM) for live multimedia streaming in peer-to-peer network

A decentralized model is primarily used for communication and file transfer, the peer-to-peer network is based on this model. The field of real-time communication and media streaming has witnessed enormous growth in recent times owing to their use of peer-to-peer network. A significant part of Internet traffic is being created by the peer-to-peer network resulting in an increase in its demand. The rise in prominence of peer-to-peer network can be attributed to its properties like resource utilization and distributed nature. Resource utilization of peer-to-peer network is one foremost motive for selecting this network over the traditional client-server architecture. The availability of selfish peer or free riders affects the total resource utilization and degrades the performance of the network. So, there is a requirement of an incentive-based mechanism to motivate the selfish peer or free riders in the network to improve the performance of the network. In this paper, we will discuss a new score based incentive mechanism to improve the overall resource utilization. This approach is based on a reward-punishment based method, so the peers are encouraged or motivated to share more resources and chastise selfish peers. For calculating the score value of a peer, parameters like upload capacity, video quality, control packets, a time period for which peers stay in the network can be used. Different score values are assigned for different parameters. The simulation results presented in this paper verify the approach and illustrate that the quality of video and performance of network improves using our new score based incentive mechanism. The parameters used for calculating the performances are; end-to-end delay, playback delay, start-up delay, and frame redundancy etc.

     

基于多种服务并存网络的实时队列时延研究

高质量的网络服务是既能保证实时数据的正常传递,又能与Internet连通实现尽力服务,并且还能满足一定的带宽。而实时数据传输的最大时延是考核网络服务质量的重要指标之一。为此,论文给出了引入TCP拥塞控制算法后,同时满足带宽保证、实时与尽力服务三种服务的网络中实时队列最大时延的计算公式。     

Fuzzy logic-based induction motor protection system

The protection is very important to detect abnormal motor running conditions such as over current, over voltage, overload, over temperature, and so on. When a failure is sensed by the protection system, a time delay should be specified to trip the motor. In the classical systems, motors are stopped with the time delay, which is adjusted constantly without considering the fault level. This paper presents a fuzzy logic-based protection system covering six different fault parameters for induction motors. This paper focuses on a new time-delay calculation for stopping induction motor and improves the overall detection performance. The time delay is computed by fuzzy logic method according to various fault parameters when one of the failures occurs on the motor. This system is successfully tested in real-time faults on the motor, and it shows that it provides sensitive protection by fuzzy rules.     

Design of a crossbar VOQ real-time switch with clock-driven scheduling for a guaranteed delay bound

Most commercial network switches are designed to achieve good average throughput and delay needed for Internet traffic, whereas hard real-time applications demand a bounded delay. Our real-time switch combines clearance-time-optimal switching with clock-based scheduling on a crossbar switching fabric. We use real-time virtual machine tasks to serve both periodic and aperiodic traffic, which simplifies analysis and provides isolation from other system operations. We can then show that any feasible traffic will be switched in two clock periods. This delay bound is enabled by introducing one-shot traffic, which can be constructed at the cost of a fixed delay of one clock period. We carry out simulation to compare our switch with the popular iSLIP crossbar switch scheduler. Our switch has a larger schedulability region, a bounded lower end-to-end switching delay, and a shorter clearance time which is the time required to serve every packet in the system.