首页> 外文学位 >Global compact passive macromodeling algorithms for high-speed circuits.
【24h】

Global compact passive macromodeling algorithms for high-speed circuits.

机译:用于高速电路的全局紧凑型无源宏建模算法。

获取原文
获取原文并翻译 | 示例

摘要

In the recent years, signal integrity has been at the forefront of new developments in CAD algorithms focussed on high-speed electronic circuits. With the continuous rise in operating frequencies, the interconnect effects such as delay, distortion, crosstalk and ringing become the dominant factors limiting the performance of VLSI systems. At relatively higher frequencies and increased level of integration, it is not always possible to find analytical or simulatable models for high-speed modules such as packages, vias, connectors, nonuniform transmission lines etc. In such a scenario, these devices are characterized by tabulated multiport parameters (S, Y, Z or T), obtained either through measurements or from the physical layout using rigorous full-wave electromagnetic simulations. However, transient simulation of such frequency-dependent tabulated data in the presence of nonlinear devices to obtain a global electrical assessment is a CPU expensive process due to the mixed frequency/time problem. Recently, several macromodeling algorithms have been proposed for this purpose. However, a major difficulty associated with these algorithms is that, they are either not able to guarantee the passivity of the resulting macromodel or when guaranteeing the passivity, they become prohibitively CPU expensive. In addition, further challenges are posed by modern designs which tend to contain multiple diverse high-speed modules. Such an emerging scenario typically leads to large sized circuit matrices and warrants unified passive reduction algorithms to facilitate fast simulations. However, the currently available algorithms cannot perform global passive reduction of such an environment with multiple diverse high-speed modules.;In this thesis, novel algorithms are presented to address the above issues. A systematic methodology is presented for the passive macromodeling of tabulated data networks. For this purpose, fast passivity verification and enforcement algorithms are presented. These algorithms are suitable for small as well as large sized macromodels of tabulated networks. In addition, an algorithm with a guaranteed search direction is presented, to reduce the number of iterations during the passivity enforcement process. Also, a new algorithm is presented for a global reduction of circuits containing diverse high-speed modules. This algorithm permits a novel formulation to guarantee the passivity of the resulting reduced-order model. Also, an efficient algorithm is presented for the second-level reduction, to obtain further compact macromodels. The developed algorithms are robust and practical, and suitable for implementation in state-of-the-art design automation tools.
机译:近年来,信号完整性一直是针对高速电子电路的CAD算法新发展的最前沿。随着工作频率的不断提高,诸如延迟,失真,串扰和振铃之类的互连效应成为限制VLSI系统性能的主要因素。在相对较高的频率和更高的集成度下,并不总是能够找到用于高速模块(例如封装,通孔,连接器,不均匀的传输线等)的分析或仿真模型。在这种情况下,这些设备的特征在于列表多端口参数(S,Y,Z或T),可以通过测量或使用严格的全波电磁仿真从物理布局获得。但是,由于存在频率/时间混合问题,在存在非线性设备的情况下对此类与频率相关的列表数据进行瞬态仿真以获得全局电气评估是一项CPU昂贵的过程。最近,已为此目的提出了几种宏建模算法。但是,与这些算法相关的主要困难是,它们要么不能保证所得宏模型的无源性,要么在保证无源性时变得非常昂贵。另外,现代设计提出了进一步的挑战,这些现代设计倾向于包含多个不同的高速模块。这种新兴情况通常会导致大型电路矩阵,并需要采用统一的无源归约算法来促进快速仿真。然而,目前可用的算法无法在具有多个不同的高速模块的环境中执行全局被动还原。本文针对上述问题提出了新颖的算法。提出了一种用于列表数据网络的被动宏建模的系统方法。为此,提出了快速被动验证和实施算法。这些算法适用于列表网络的小型和大型宏模型。另外,提出了一种具有保证搜索方向的算法,以减少被动实施过程中的迭代次数。此外,提出了一种新算法,用于全局减少包含各种高速模块的电路。该算法允许一种新颖的公式来保证所得降阶模型的无源性。此外,提出了一种用于第二级归约的有效算法,以获得进一步的紧凑型宏模型。所开发的算法健壮且实用,并且适合在最新的设计自动化工具中实施。

著录项

  • 作者

    Saraswat, Dharmendra.;

  • 作者单位

    Carleton University (Canada).;

  • 授予单位 Carleton University (Canada).;
  • 学科 Engineering Electronics and Electrical.
  • 学位 Ph.D.
  • 年度 2006
  • 页码 241 p.
  • 总页数 241
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 无线电电子学、电信技术;
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号