|
|||||
|
|
| 嵌套式稀疏网格随机配置法及其在随机门延时建模中的应用 | |
| 引用本文： | 罗旭,杨帆,朱恒亮,陶俊,蔡伟,周电,曾璇.嵌套式稀疏网格随机配置法及其在随机门延时建模中的应用[J].计算机辅助设计与图形学学报,2010,22(1). |
| 作者姓名： | 罗旭 杨帆 朱恒亮 陶俊 蔡伟 周电 曾璇 |
| 作者单位： | 1. 复旦大学专用集成电路与系统国家重点实验室,物质计算科学教育部重点实验室,上海,201203 2. Department of Mathematics, University of North Carolina at Charlotte, NC 28223 USA 3. Department of Electrical Engineering, University of Texas at Dallas, TX 75083 USA |
| 基金项目： | 国家自然科学基金(60976034,60676018,60806013,60673029);;国家十一五重大科技专项项目(2008ZX01035-001-06,2009ZX02023-4-3);;国家重点基础研究发展计划项目(2005CB321701);;教育部高等学校博士学科点专项科研基金(200802460068);;上海市领军人才项目上海市科学技术委员会国际科技合作基金项目(08510700100) |
| 摘 要： | 为了提高随机工艺偏差下门延时建模的计算精度和效率,提出一种基于扩展Gauss积分理论及嵌套式稀疏网格技术的随机配置门延时建模方法.首先采用参数空间中具有指数收敛特性的随机正交多项式对随机门延时进行逼近;然后针对现有的基于传统Gauss积分理论的稀疏网格随机配置法所用的配置点不具有嵌套特性的问题,利用单变量扩展Gauss积分理论及稀疏网格技术构造了一组嵌套式多变量Gauss积分点,将其作为随机门延时建模的配置点.这组配置点既具有Gauss积分点的高精度,又满足嵌套性质,且在低阶积分配置点上已经得到的门延时可以在高阶积分时重复使用.与现有的基于非嵌套式配置点的随机配置法相比,该方法的计算精度和效率可以得到很大的提升,数值实验结果也验证了该方法在计算精度和效率上的优势. |
| 关 键 词： | 门延时 工艺偏差 扩展Gauss积分 嵌套式稀疏网格 随机配置法 |
Nested Sparse-grid Stochastic Collocation Method and Its Application to Gate Delay Modeling under Process Variations |
|
| Luo Xu,Yang Fan,Zhu Hengliang,Tao Jun,Cai Wei,Zhou Dian,Zeng Xuan.Nested Sparse-grid Stochastic Collocation Method and Its Application to Gate Delay Modeling under Process Variations[J].Journal of Computer-Aided Design & Computer Graphics,2010,22(1). | |
| Authors: | Luo Xu Yang Fan Zhu Hengliang Tao Jun Cai Wei Zhou Dian Zeng Xuan |
| Affiliation: | State Key Laboratory of ASIC & System/a>;MOE Key Laboratory for Computational Physical Sciences/a>;Fudan University/a>;Shanghai 201203;Department of Mathematics/a>;University of North Carolina at Charlotte/a>;NC 28223 USA;Department of Electrical Engineering/a>;University of Texas at Dallas/a>;TX 75083 USA |
| Abstract: | In this paper,an extended Gaussian quadrature based nested sparse-grid stochastic collocation method(NSSCM) is proposed for further improving the computation accuracy and efficiency of stochastic gate delay modeling considering process variation.Firstly,the orthogonal polynomial bases in the stochastic space of gate parameters are employed in NSSCM to approximate the stochastic gate delay and exponential convergence rate is achieved.Secondly,the proposed NSSCM employs one-dimensional extended Gaussian quadr... |
| Keywords: | gate delay process variations extended Gaussian quadrature nested sparse grid stochastic collocation method |
| 本文献已被 CNKI 万方数据 等数据库收录！ | |