最小代价多播生成树的快速算法

本文针对MPH（Minimum Path Cost Heuristic）等多播最小生成树算法存在的问题，通过改进最短路径节点的搜寻过程，以较小的存储空间为代价，获得了计算效率很高的快速最小代价多播生成树算法FMPH（Fast Minimum Path Cost Heuristic），且获得多播生成树与MPH算法完全相同，随机网络模型的仿真结果表明：FMPH算法快速、稳定，是一种值得推广使用的高效算法。     

适于低轨卫星IP网络的核心群合并共享树组播算法

为了解决低轨卫星IP网络中现有典型源组播算法的信道资源浪费问题,该文提出了一种低树代价的组播算法,即核心群合并共享树(CCST)算法,包括动态近似中心(DAC)选核方法和核心群合并组播路径构建方法。DAC方法基于逻辑位置形成的虚拟静态、结构规则的网络拓扑选择核节点。在核心群合并方法中,以核节点作为初始核心群,通过核心群和剩余组成员的最短路径方法逐步扩展直至整棵组播树构建完成,从而使得组播树的树代价最小,大大提高了网络的传输带宽利用率和组播传输效率。最后,与低轨卫星IP网络中的其他几种典型算法进行了性能对比,仿真结果说明,CCST算法的树代价性能比其它算法有较大改善,而端到端传播时延略高。     

可扩展路由器中SPT并行计算的实现

随着互联网的飞速发展,集群结构的下一代核心路由器已经成为研究的重点.在可扩展路由器中(cluster router),并行路由算法是关键问题之一.对于广泛部署的OSPF协议,最短路径树(SPT)的并行计算是其并行化的核心难点.本文提出了一种计算最短路径树的算法－分区Dijkstra算法(D-D),分析了算法性能,并通过模拟实验验证了算法的性能.     

基于时延限制的分布式组播路由选择算法

本文基于最短路径算法提出了主树的概念,并且利用组播树的代价和时延限制的定义给出了一种分布式组播路由选择算法。可以在时延限制的前提下使组播树的代价接近最优。最后,给出了在ＯＰＮＥＴ环境中的仿真结果,并根据结果说明了本文方法的有效性。     

改进的Floyd算法在套牌车辨别中的应用

针对套牌车辨别中用传统Floyd算法计算最短路径距离存在效率低和速度慢等问题,文中提出了一种改进的Floyd算法,减少了矩阵迭代过程中任意两点之间的最短路径计算量,提高了算法效率。在MATLAB中用同一矩阵对改进前后的Floyd算法进行仿真和比较,结果表明,改进后的算法在效率上高于传统算法。运用某市交通数据集辨别套牌车时,改进的Floyd算法效率比传统Floyd算法的效率提升了近30%,说明了改进的Floyd算法的高效性和实用性。     

大规模MIMO下具有反馈的低复杂度检测算法

针对大规模MIMO系统,信号检测算法的复杂度往往呈指数级增长的问题,通过对传统的QRD-M检测算法和树搜索的深度研究,对其进行改进,得出一种具有反馈的低复杂度检测算法,该算法主要运用分支的权重和分支间的反馈对树进行修剪,然后运用树搜索的原理进行搜索输出.仿真结果表明,该改进算法以降低较少系统性能的代价,能够极大地降低计算复杂度,是一种折衷的高效检测算法,完全能运用于大规模MIMO系统中.     

基于最短路径树的节点删除动态路由算法

提出一种基于最短路径树的节点删除动态路由算法。算法建立一个最短路径树更新集合,该集合包括被删除节点的断裂子树所有节点和其它节点连接的边,利用子树的结构信息,对子树节点的直系子孙节点和祖先节点进行更新,采用Dijkstra算法对其它子树节点进行更新。实验结果表明,该算法能有效减少节点更新计算次数。     

一种新的QoS保证的快速组播路由算法

提出了一种新的服务质量保证的组播路由算法。算法使用了一个修改的斯坦利树近似算法来构建时延有界的低代价组播树，再通过最小时延路径与其他尚不在组播树的组内节点连接，并消除环路。理论和仿真表明，新算法与其他一些时延有界组播路由算法一样有较好的性能。并且有较小的复杂度。     

基于层叠模型的动态组播业务疏导算法

流量疏导对于提高波分复用网络的利用率具有重要的意义.文章基于层叠模型的IP over WDM网络结构,研究了动态组播业务疏导问题,给出了两种动态组播业务疏导方法.仿真结果显示,在IP层和光层都计算最短路径来选取最短路径组建组播树的算法能够获得更低的阻塞率.     

基于虚拟现实技术的三维室内设计及人物漫游技术

针对交互型虚拟现实室内场景建设以及人物漫游问题,提出一种基于改进二叉树BSP的室内场景管理方法,并采用遗传算法实现该场景下的人物路径规划。首先,在基于几何建模法的虚拟漫游系统中,利用改进的二叉空间剖分BSP树算法对三维场景进行预处理,从而提高实时图形绘制的效率;然后,对漫游场景进行栅格化处理,并采用遗传算法实现两点间最短漫游路径的计算。室内场景漫游测试结果表明,相比传统的A*算法,提出的BSP-遗传漫游算法在耗时和最短路径长度两方面均表现出一定的优势。     

光组播路由代价与波长使用量的联合优化方法

为解决光组播路由中组播中路由代价和波长资源消耗单一化造成的组播路树路由的代价过高问题,在分光节点约束条件下,提出了光组播路由代价与波长使用量联合优化的长路优先(LPF)方法和短路优先(SPF)方法。算法通过检查最小光组播树是否存在节点分光约束的问题,根据设置的波长使用代价控制因子,使LPF或SPF的路由代价和波长使用量最小。LPF方法首先选择组播树最长路径或新波长通道重路由受分光约束的目的节点,SPF方法先选择组播树中最短路径或新波长通道重路由受分光约束的目的节点,仿真结果表明,本文提出的两种联合优化方法都能实现路由代价较低和波长需求较少的目的。     

Routing to Multiple Destinations in Computer Networks

Algorithms for effectively routing messages from a source to multiple destination nodes in a store-and-forward computer network are studied. The focus is on minimizing the network cost (NC), which is the sum of weights of the links in the routing path. Several heuristic algorithms are studied for finding the NC minimum path (which is an NP-complete problem). Among them are variations of a minimum spanning tree (MST) heuristic and heuristics for the traveling salesman problem, both of which use global network information. Another set of heuristics examined are based on using only the shortest paths to the destinations. While the MST algorithm has the best worst case performance among all algorithms, a detailed, empirical study of the "average" performance of the algorithms on typical, randomly chosen networks reveals that simpler heuristics are almost as effective. The NC cost measure is also compared to the destination cost (DC), which is the sum of weights of the shortest path distances to all destinations. A scheme of algorithms is given which trades off between NC and DC.     

适于低轨卫星IP网络的特定源组播算法

为了解决低轨卫星IP网络中现有特定源组播算法的信道资源浪费问题,本文提出了一套新的特定源组播算法,即基于核心群的特定源组播算法(CSSM)和加权的CSSM算法(w-CSSM).CSSM算法以源节点作为初始核心群,通过核心群和剩余组成员的最短路径方法逐步扩展直至整棵组播树构建完成,所得的树代价最小,从而大大提高了网络的带宽利用率和传输效率.在w-CSSM算法中,加权因子可以自适应调整以适度增大树代价、降低端到端传播时延,以支持某些有严格端到端时延要求的实时组播业务.通过与低轨卫星IP网络中典型特定源组播算法MRA的仿真比较,可以看出CSSM和w-CSSM算法的树代价性能比MRA有较大改善,不过端到端传播时延略高.     

A new delay‐constrained algorithm for multicast routing tree construction

New multimedia applications provide guaranteed end‐to‐end quality of service (QoS) and have stringent constraints on delay, delay‐jitter, bandwidth, cost, etc. The main task of QoS routing is to find a route in the network, with sufficient resources to satisfy the constraints. Most multicast routing algorithms are not fast enough for large‐scale networks and where the source node uses global cost information to construct a multicast tree. We propose a fast and simple heuristic algorithm (EPDT) for delay‐constrained routing problem for multicast tree construction. This algorithm uses a greedy strategy based on shortest‐path and minimal spanning trees. It combines the minimum cost and the minimum radius objectives by combining respectively optimal Prim's and Dijkstra's algorithms. It biases routes through destinations. Besides, it uses cost information only from neighbouring nodes as it proceeds, which makes it more practical, from an implementation point of view. Copyright © 2004 John Wiley & Sons, Ltd.     

Multicast routing, load balancing, and wavelength assignment ontree of rings

There are two steps to establish a multicast connection in WDM networks: routing and wavelength assignment. The shortest path tree (SPT) and minimum spanning tree (MST) are the two widely used multicast routing methods. The SPT method minimizes the delay from the source to every destination along a routing tree, and the MST method is often used to minimize the network cost of the tree. Load balancing is an important objective in multicast routing, which minimizes the maximal link load in the system. The objective of wavelength assignment is to minimize the number of wavelengths used in the system. This paper analyzes the performance of the shortest path tree (SPT) and minimum spanning tree (MST) methods in the tree of ring networks, regarding the performance criteria such as the delay and network cost of the generated routing trees, load balancing, and the number of wavelengths required in the system. We prove that SPT and MST methods can not only produce routing trees with low network costs and short delays, but also have good competitive ratios for the load balancing problem (LBP) and wavelength assignment problem (WAP), respectively     

局部搜索最小路径费用算法

本文在MPH(Minimum Path Cost Heuristic)的基础上,改进了端节点的加入过程,得到了两种改进的MPH算法:局部搜索最小路径费用算法LSMPH(Locally Searching MPH)和简化的LSMPH:最短端节点最小路径费用算法STMPH(Shortest Terminal MPH).在随机网络模型的基础上,我们进一步进行了仿真.仿真结果表明,LSMPH以相对较小的费用增加换取更快的计算速度.如果要求更快的速度,可以采用STMPH.     

Solution to Some Graph Problems in which Edges Have Two Weights

ＳｏｌｕｔｉｏｎｔｏＳｏｍｅＧｒａｐｈＰｒｏｂｌｅｍｓｉｎｗｈｉｃｈＥｄｇｅｓＨａｖｅＴｗｏＷｅｉｇｈｔｓＭａｎｕｓｃｒｉｐｔｒｅｃｅｉｖｅｄＭａｒ．６，１９９７．ＴｈｉｓｗｏｒｋｉｓｓｕｐｐｏｒｔｅｄｂｙＮａｔｉｏｎａｌＳｃｉｅｎｃｅＦｏｕｎｄａｔ...     

无线Mesh网络可靠多播路由

无线Mesh网络多播路由是无线路由必须解决的关键技术。部分研究者对网络资源和服务质量（QoS）进行研究,提出了建立最短路径树、最小开销树、负载感知、信道分配多播等多播算法;有的算法考虑链路可靠性,建立备用路径。将结合网络资源和可靠性对多播路由算法进行研究,提出了建立可靠多播树（RT,Reliable Tree）的多播路由算法：可靠多播树是一个多树结构,由一棵首选多播树和一棵多径树构成,多径树提供可靠多路径,以提高网络吞吐量。     

Distributed formation of core-based forwarding multicast trees in mobile ad hoc networks

A core-based forwarding multicast tree is a shortest path tree rooted at core node that distributes multicast packets to all group members via the tree after the packets are sent to the core. Traditionally, the bandwidth cost consumed by transmitting a packet from the core via the tree is evaluated by the total weights of all the edges. And, the bandwidth cost is minimized by constructing the multicast tree that has minimum total weights of edges to span all group members. However, when the local broadcast operation is used to multicast a packet, we found that the bandwidth cost is supposed to be evaluated by the total weights of all senders that include the core and all non-leaves. Since the multicast tree with the number of nodes greater than or equal to three has minimum bandwidth cost only when the core is not a leaf, it leads us to find the multicast tree with the minimum number of non-leaves when each sender node has a unit weight. However, no polynomial time approximation scheme can be found for the minimum non-leaf multicast tree problem unless P = NP since the problem is not only NP-hard but also MAX-SNP hard. Thus, a heuristic is proposed to dynamically reduce the number of non-leaves in the multicast tree. Experimental results show that the multicast tree after the execution of our method has smaller number of non-leaves than others in the geometrically distributed network model.