共查询到20条相似文献,搜索用时 84 毫秒
1.
传统的性能驱动布线算法受限于树形或固定的布线拓扑结构.本文提出一种回路性能优化布线算法,针对树形线网布线,通过在已存在的布线树上加入回路来减小所选择关键路径的延迟时间或线网的最大延迟.我们将互连线树归结为分布传输线网络并采用Elmore延迟计算方法.本文证明,通过选择适当的RC,在连接节点与关键节点之间加入连线可达到减小所选择线网中关键路径延迟或线网最大延迟的目的.实验结果表明,我们的方法有效且可以集成在现有CAD性能优化布线系统中.本文同时给出了所加入线段长度的计算方法. 相似文献
2.
一种新的基于知识的四边通道布线算法 总被引:1,自引:0,他引:1
本文提出了一种新的基于知识的双层四边通道布线算法,该算法对四边通道的布线是通过以下四步完成的。首先,对四边通道的四个角布线,其次,对关键线网优先布线,接下来,利用线网间相互制约关系进行同步增长布线;最后,对仍然没有完成连接的线网,用李氏算法布线。由于使用了启发式规则,使得该算法具有较高的布通率和布线效率。 相似文献
3.
4.
5.
唐茂林 《微电子学与计算机》1994,(4)
在VLSI布线设计中,线网的分布情况直接影响到VLSI芯片的电学性能、可靠性和制造成本。但是,由于通道区布线问题的计算复杂性很高 ̄[1],布线过程中往往没有考虑到线网的分布问题,因此,布线完成后有必要对布线结果进行一次线网分布优化。本文中提出了两种工艺的双层通道区布线线网分布优化问题,并给出了相应的算法。 相似文献
6.
H—V布线模式下产生通孔的根本原因是由于不同线网间存在几何重叠或交叉;标准单元内布线主要应用多晶硅层与单金属层的布线层资源,考虑到不同层间相异的导电特性,文章提出了一种算法,基于网段拓扑交叉分析实现最大化金属层及最小化多晶层分配,优化线网通道分配与线长,同时满足通孔最小化。 相似文献
7.
8.
唐茂林 《微电子学与计算机》1994,11(4):5-8
在VLSI布线设计中,线网的分布情况直接影响到VLSI芯片的电学性能、可靠性和制造成本。但是,由于通道区布线问题的计算复杂性很高^[1],布线过程中往往没有考虑到线网的分布问题,因此,布线完成后有必要对布线结果进行一次线网分布优化。本文中提出了两种工艺的双层通道区布线线网分布优化问题,并给出了相应的算法。 相似文献
9.
《电子技术与软件工程》2015,(1)
随着集成电路的发展,芯片的特征尺寸变得越来越小,模拟电路中的电流也变得越来越大,因大电流使互连线出现短路或开路的现象称为电迁移现象。针对电迁移现象,本文提出了一种单主干权重排序布线算法,它根据各端点电流值的大小将端点的布线顺序进行排序,使电流较大的线网先完成布线,以达到缩小布线面积的目标。这种算法满足了布线设计当中的基尔霍夫电流定律,最小设计规则以及避开障碍物等约束条件,最终实现多源多漏的线网拓扑结构。 相似文献
10.
11.
《Journal of Location Based Services》2013,7(4):273-295
AbstractWLANs are prone to flooding denial-of-service (DoS) attacks. Present research work proposes a novel method to enhance WLAN security by restricting flooding DoS effect on the access point (AP). Each node in the proposed method listens transmission of other neighbouring nodes and sends at regular interval a linked list to AP, containing the identifiers (IDs) of all nodes from which it hears. AP stores the linked list information and runs location-based coordinate calculation (LBCC) algorithm. LBCC finds the expected coordinates of each node and probable location area with the expected coordinates as centre. Flooding DoS attack by the attacker is restricted at the AP using node’s probable location area. The method is validated by simulation experimentations with varying topologies. Further, geometrical analysis of the method is also provided. 相似文献
12.
针对传统的线段检测方法使线段宽度信息丢失且造成部分图形轮廓失真的问题,提出了一种基于行程矢量快速连接原理的线段检测方法。该方法采用四方向行程表示图像边缘点,依据同类行程的连通性建立前趋行程搜索模型,形成容忍断裂范围内的轮廓行程链;采用链表结构存储行程链首尾位置并动态清除无效链表结点,节省了内存开销;通过后续检验,合并满足规则的线段,并剔除虚假的候选线段,保证了线段信息的真实性和完整性。该方法通过构造行程链,将一段边缘线是否为直线段的问题简化为简单的共线性判断问题。实验结果表明,该方法实时性强,计算复杂度和空间复杂度小,抗噪声能力强,具有实用意义。 相似文献
13.
14.
In this paper, we propose a procedure that enhances the very low
bit-rate coding performance of Set Partitioning in Hierarchical
Trees (SPIHT)-based image coders. In SPIHT-based algorithms,
all nodes with descendants in the LL band are initially added to
the list of insignificant sets as type-A entries. Our
experiments indicate that type-A nodes are highly likely to be
insignificant during the sorting passes of the high bit planes.
Our new procedure exploits this characteristic so that all these
type-A nodes can be encapsulated into a single virtual node. 相似文献
15.
Y. Zhen M. Q. Wu D. P. Wu B. Sun X. Y. Li C. X. Xu 《International Journal of Communication Systems》2009,22(11):1377-1401
The opportunistic routing mechanism can use several lossy broadcast links to support reliable transmission. In this paper, a simple opportunistic routing mechanism for real‐time multimedia services is proposed. This mechanism is based on the dynamic source routing protocol with some modifications, multiple route request, and route reply messages are used to construct the forwarder list, and the nodes within the forwarder list forward the packets which they overhear. The forwarder list is placed on the packet header in the form of a Bloom filter, which will restrict the size of the forwarder list to a constant value. There are no strict scheduling mechanisms for the forwarding order of the forwarder nodes, thus our opportunistic routing mechanism can be scalable for multiple simultaneous flows. Simulations show that our mechanism can effectively decrease the transmission times and the amount of the control messages for each packet and reduce the end‐to‐end delay for real‐time voice service, the quality of service for these services can be supported well over the unstable wireless channel. Copyright © 2009 John Wiley & Sons, Ltd. 相似文献
16.
在任务调度领域,基于列表的任务调度算法被广泛应用。经典列表调度算法在节点排序阶段会对权值一致的任务节点进行随机排序,但这种节点排序方式过于粗糙,难以取得较好效果。针对这一缺陷,文中提出了一种改进型列表调度算法,通过特殊列表片段将权值一致的任务整合到一起,并对特殊列表片段的调度解空间进行遍历,以迭代产生更优的调度列表获得更好的调度效果。随机DAG图测试结果表明,提出的算法调度效果优异,算法平均增强比最高可达15.3%,不仅适用于多种任务图规模,且在CCR和平均出入度较高的情况下有更好的调度性能。 相似文献
17.
All finite-element mesh-generation packages output a so-called universal file which, as a minimum, contains a list of the node coordinates, a table of the nodes which form each element, and a list of the materials filling the element. Algorithms that make use of this information to generate additional information or tables, which are typically required in any electromagnetic analysis code, are given. The given code fragments are in C to avoid explicit dimensioning of arrays and facilitate the execution of recursive routines 相似文献
18.
19.
在大型数据集群网络中,业务逻辑节点和数据库节点分布在不同的地理位置,导致在该网络中创建或检索用户数据将经历较大的网络延迟.如何快速找到用户数据的地理位置节点(服务器识别号)将是减少网络延迟的关键.介绍一种动态索引算法,基于简单正则表达,建立用户数据和服务器组之间的映射关系,并引入动态多叉树,实现动态更改映射关系.引入一元组数据节点和二元组数据节点的概念,应用于多叉树,通过分析一元组多叉树和二元组多叉树的时间效率和空间效率,证明二元组多叉树随着树深的增长,检索时间复杂度保持更好的线性特性.通过一些性能测试的实验数据的比较,得出二元组方案的综合性能更优的结论.最后,简要地介绍该算法的应用领域. 相似文献
20.
This paper presents a distributed storage system which is built by lots of low-cost nodes. Low-cost node is divided into two parts: metadata nodes and storage nodes. The storage system will use a group of low-cost metadata nodes instead of the Google File System's (GFS's) single master server machine. These metadata nodes are linked by high-speed and high-density local peer-to-peer (P2P) network. We also designed the method organized metadata in the metadata cluster. The storage nodes are also linked by high-speed local P2P network. Metadata cluster and storage cluster are linked through high-speed reliable network. User files are organized into a tree structure by using a virtual storage component called “unit”. Storage system is built on the top of every node's operating system. This paper also gives the specific way of file addressing. 相似文献