一种新的MPLS流量工程路由算法MSMR

本文提出了一种新的用于多协议标签交换网络流量工程的动态路由算法。先前研究者们提出的MIRA等著名算法试图通过绕开关键链路而最小化不同源-目的节点对之间的干涉。但是,有些情况下,这些算法可能选择过长的路径,或在寻找关键路径时判断不准确。因而本文提出了一种新的最大化其它入出口对之间的最大流之和的启发式算法。模拟结果证明了算法在拒绝率和吞吐量方面达到了更好的性能。     

A System for Weighted Max-Min Congestion Control in MPLS Networks

This paper introduces a framework to support explicit rate congestion control for best-effort service in MPLS networks. We devise the two basic mechanisms required, namely, a signaling component to convey explicit-rate congestion control information and associated algorithm for calculating explicit fair rates. The proposed signaling component uses Resource Reservation Protocol (RSVP) in a novel way to convey explicit rate congestion control information. An algorithm to calculate Weight Proportional Max-Min (WPMM) fair rates appropriate for MPLS networks is introduced. Simulations are presented that show the effectiveness of the RSVP explicit rate signaling method when combined with the proposed max-min fair rate algorithm.     

用"分块矩阵法"快速解算极点配置自适应控制规律

本文提出快速解算极点配置自适应控制规律的"分块矩阵法".计算工作量从O(N3)降至O(N2);其结果亦适用于随机性模型.仿真结果表明,"分块矩阵法"计算量小.对于控制含有未知或时变滞后的过程有很大实用价值.     

A Parallel Algorithm for Traffic Control Problems in Three-Stage Connecting Networks

A parallel heuristic algorithm for traffic control problems in three-stage connecting networks is presented in this paper. A three-stage connecting network consists of an input crossbar switching stage, an intermediate crossbar switching stage, and an output crossbar switching stage. The goal of our algorithm is to quickly and efficiently find a conflict-free switching assignment for communication demands through the network. The algorithm requires n² × m processing elements for the network composed of n input/output switches and m intermediate switches, where it runs not only on a sequential machine, but also on a parallel machine with maximally n² × m processors. The algorithm was verified by 1100 simulation runs with the network size from 10² × 7 to 50² × 27. The simulation results show that the algorithm can find a solution in nearly constant time with n² × m processors.     

ATM网络中一种新的动态优先队列控制方案

ATM网络中支持的突发业务如语音、视频和图像等要求提供不同的服务质量.为了满足这些突发业务的各种服务质量要求,主要以马尔可夫到达过程来模拟ATM网络中的两类具有不同QoS要求的突发业务流,并为各种不同业务提供QoS保证.通过使用嵌入马尔可夫链和灵活的辅助措施,获得用户在离开时刻各类业务的队列长度,并可以推导出任意时刻下的业务队列长度,导出延迟和损耗等性能参数.并对所提出的方案进行了仿真分析,结果表明该方案完全可以满足ATM网络中突发业务的服务质量.     

分层迁移：分层多播拥塞控制中的快速响应方法

1 引言随着多播(multicast)技术的发展以及多播主干(Mbone)的普及,在Internet上出现了许多基于IP多播的音频/视频会话工具和多播流应用,然而这类应用却面临Internet的异构性和缺乏拥塞控制等主要问题。缺乏拥塞控制措施的多播应用无疑会影响行为规范的TCP流,还有导致网络瘫痪的可能。基于收方驱动的分层多播协议(RLM)最先提出分层多     

Distributed Model Predictive Control Method for Optimal Coordination of Signal Splits in Urban Traffic Networks

Coordination and control approaches based on model predictive control (MPC) have been widely investigated for traffic signal control in urban traffic networks. However, due to the complex non‐linear characters of traffic flows and the large scale of traffic networks, a basic challenge faced by these approaches is the high online computational complexity. In this paper, to reduce the computational complexity and improve the applicability of traffic signal control approaches based on MPC in practice, we propose a distributed MPC approach (DCA‐MPC) to coordinate and optimize the signal splits. Instead of describing the dynamics of traffic flow within each link of the traffic network with a simplified linear model, we present an improved nonlinear traffic model. Based on the nonlinear model, an MPC optimization framework for the signal splits control is developed, whereby the interactions between subsystems are accurately modeled by employing two interconnecting constraints. In addition, by designing a novel dual decomposition strategy, a distributed coordination algorithm is proposed. Finally, with a benchmark traffic network, experimental results are given to illustrate the effectiveness of the proposed method.     

An Effective Framework for Fast Expert Mining in Collaboration Networks: A Group-Oriented and Cost-Based Method

The growth of social networks in modern information systems has enabled the collaboration of experts at a scale that was unseen before.Given a task and a graph of experts where each expert possesses some skills,we tend to find an effective team of experts who are able to accomplish the task.This team should consider how team members collaborate in an effective manner to perform the task as well as how efficient the team assignment is,considering each expert has the minimum required level of skill.Here,we generalize the problem in multiple perspectives.First,a method is provided to determine the skill level of each expert based on his/her skill and collaboration among neighbors.Second,the graph is aggregated to the set of skilled expert groups that are strongly correlated based on their skills as well as the best connection among them.By considering the groups,search space is significantly reduced and moreover it causes to prevent from the growth of redundant communication costs and team cardinality while assigning the team members.Third,the existing RarestFirst algorithm is extended to more generalized version,and finally the cost definition is customized to improve the efficiency of selected team.Experiments on DBLP co-authorship graph show that in terms of efficiency and effectiveness,our proposed framework is achieved well in practice.     

WDM全光网络中Multicast的寻径与波长分配算法

在WDM全光网络中实现实时Multicast功能是计算机网络的应用要求,也是现代计算机网络的重要特征。该文提出了一种在WDM全光网络中实现实时Multicast的算法。该算法以辅助的波长图为基础,将寻径与波长分配统一进行,构造满足延迟约束的、具有较低成本的Multicast树,实现Multicast功能。     

WDM网络中实时组播的分布式优化RWA算法

在WDM网络中,由于每条链路上可用波长是动态变化的,在考虑波长转换延迟时间的条件下,实现实时组播连接的路由选择与波长分配是十分困难的。论文提出了一种用于建立实时组播连接的分布式路由选择与波长分配算法。该算法将路由选择与波长分配统一进行,大大减少连接的建立时间。组播路由算法以Prim最小生成树算法为基础,生成一棵满足给定延迟时限的最小成本树。波长分配使用最少波长转换和同一波长最长使用策略。     

基于VDL的附网存储快照及在线备份新技术

针对现有NAS系统在快照及在线备份技术方面存在的问题,该文提出了一种通过虚拟设备层(VDL:VirtualDeviceLayer)实现了NAS的快照和在线备份技术。在VDL中:通过一个快照盘保存在快照和备份期间系统的写操作,实现NAS的快照和在线备份;以设备驱动模块实现,与具体的文件系统无关,使其适用性更为广泛,并让现有的离线备份软件实现在线备份功能;在驱动模块内部对数据的小写进行了优化,较大的提高了办公/工程环境下工作负载的小写性能;备份的初始化时间几乎可以忽略不计,大大提高了存储系统的可用性;实现了存储虚拟化功能,从而将多个存储接点整合成单一的存储池。通过大量的测试表明,VDL对系统整体I/O性能和应用程序影响很小,同时表现出优良的小写性能,达到了预期的效果。     

CDMA网络中综合业务的呼叫接纳控制算法及性能分析

呼叫接纳控制(CAC)是CDMA系统无线资源管理中的重要组成部分,有效的CAC算法能在满足各类业务不同QoS要求的同时,提高资源的利用率。本文对于新呼叫和切换呼叫,提出了一种有效的呼叫接纳控制算法,给予了切换呼叫更高的优先级,并采用Markov链对算法的性能进行了分析。     

FISDR:一种新的故障注入无线传感器网络性能评估系统

在无线传感器网络WSN中,可靠性和容错性是评价WSN稳定性的重要指标。在WSN的实际应用中常会发生很多故障(Fault)和干扰,采用故障注入FI技术可以向WSN人为地注入这些故障和干扰,通过观察注入故障后网络的反应来评价网络的可靠性和容错性,从而对网络机制进行改进来提高网络的可靠性和稳定性。本文提出的FISDR是一种采用故障注入的WSN性能评估系统,基于软件故障注入方法,采用一对一的方式通过特殊接口与WSN节点连接,向WSN节点注入故障命令。该系统一是可以有效地向WSN注入各种实际应用时可能遇到的故障和干扰并观察网络运行的状况;二是可以接收包括WSN节点和其它各种设备通过特殊接口发来的数据,并将其存储;三是配有上位机软件对网络拓扑结构进行监控、对传输成功率进行统计并对存储的大量信息进行分析,从而对WSN网络及其可靠性做出评价。本系统在一栋五层办公楼分别用数十个WSN节点和FISDR节点做实验,实验内容包括使用FISDR向WSN注入大规模的故障并统计网络的反应状况,验证FISDR故障注入的效果,从而对FISDR的性能进行测试和分析。实验结果表明,FISDR可以有效地向WSN注入各种故障以评价其可靠性,在测试WSN及其可靠性评价方面有很高的应用价值。     

WDM全光网络中实时组播的分布式路由与波长分配算法

在WDM网络中，由于每条链路上可用波长是动态变化的，在考虑波长转换延迟的条件下，实现实时组播连接的路由与波长分配是十分困难的．假定WDM网络中每条链路有多根光纤，只有部分结点具有波长转换器且波长转换时间是不可忽略的，据此提出了一种用于建立实时组播连接的分布式路由与波长分配算法．该算法以Prim最小生成树算法为基础，生成一棵满足给定延迟时限的最小成本树．当最小成本树不能包括所有目的结点时，对剩余目的结点生成一棵最短延迟树，然后合并两棵树得到一棵组播树．波长分配使用最少波长转换和负载平衡策略．     

一种新的基于广义细胞自动机的网络快速包交换优化方法

实时优化求解快速包交换问题(FPS)是提高网络性能的重要手段．基于梯度下降法等数学规划方法，不能并行地实时地优化求解FPS问题，而基于Hopfield型神经网络和细胞神经网络的优化方法中，都只有单一粒度的细胞动力学方程和单一粒度细胞之间的相互作用，不仅收敛到平衡点的过程长，而且神经网络参数的选择和修正十分困难．该文提出一种新的具有多粒度宏细胞的广义细胞自动机模型和方法，广义细胞自动机中的小粒度宏细胞聚合成可以独立演化的大粒度宏细胞，通过多粒度群体的不同程度群体智能的相互作用，能够比目前其他方法更快更有效地分布并行地优化求解FPS问题和其它类似的复杂的网络优化问题．     

A New Contention Resolution Procedure for HFC Access Networks and its Performance Evaluation

Bidirectional Cable TV networks using hybrid fiber coaxial (HFC) systems are good examples of broadcast environments where a contention resolution algorithm is needed in order to allocate the multiaccess medium among various customers. The medium access control (MAC) scheme, proposed by DAVIC/DVB, IEEE 802.14 and DOCSIS for the upstream channel of HFC access networks is based on a mixable contention-based/contentionless time slot assignment. Contention-less (CL) slots are assigned by the head-end (HE) to end stations according to a reservation scheme. Contention-based slots (CB) are randomly accessed by active terminals without any prelimanry allocation and so collisions may occur. To resolve contention the contention tree algorithm has been widely accepted by the DVB/DAVIC, IEEE 802.14 and DOCSIS standards for MAC because of higher throughput and lower access delay. In this paper we propose a simple modification to the existing protocol and analyze its performance. We propose to have one slot in the frame exclusively reserved for the new arrivals that wish to access the channel capacity using contention resolution and atleast one more slot reserved for resolving their contention if there was a contention in the arrival slot. This assumption simplifies the protocol to a queuing mechanism and we use the results of the queue to analyze the protocol. The queuing analysis method is used to determine the throughput of the channel and waiting times of the arbitrary customers. Furthermore, we present numerical results and compare that with simulations.     

A New Methodology for Modeling, Analysis, Synthesis, and Simulation of Time-Optimal Train Traffic in Large Networks

From a system-theoretic standpoint, a constrained state-space model for train traffic in a large railway network is developed. The novelty of the work is the transformation or rather reduction of the directed graph of the network to some parallel lists. Mathematization of this sophisticated problem is thus circumvented. All the aspects of a real network (such as that of the German rail) are completely captured by this model. Some degrees of freedom, as well as some robustness can be injected into the operation of the system. The problem of time-optimal train traffic in large networks is then defined and solved using the maximum principle. The solution is obtained by reducing the boundary value problem arising from the time-optimality criterion to an initial value problem for an ordinary differential equation. A taxonomy of all possible switching points of the control actions is presented. The proposed approach is expected to result in faster-than-real-time simulation of time-optimal traffic in large networks and, thus, facilitation of real-time control of the network by dispatchers. This expectation is quantitatively justified by analysis of simulation results of some small parts of the German rail network.     

一种采用拆分组装构造神经网络的新途径

通过比较BP各种模式,提出了BP神经网络的拆分和组装方法,阐述了它的正确性,并采用该方法来构造成一个学习结果达到均方根误差全局最小点的BP神经网络。该方法在有效克服局部极小点这一问题上具有普遍意义。     

一种光传送网分层结构及其业务量疏导优化模型

提出一种包括IP、SDH和WDM三种网络层次的光传送网分层结构及其业务量疏导优化模型。IP和SDH层的逻辑拓扑由优化模型根据这两层的业务请求矩阵生成,IP和SDH层以多种带宽颗粒度串路承栽具有不同请求带宽的业务流,模型以最大化网络流量为目标,定义了三种网络层次上不同颗粒度串路之间和串路与业务流之间的相关性约束。在不同网络资源配置下,利用Lingo优化软件对模型的求解结果验证了模型的有效性。