Efficient and Performance Guaranteed Methods for Order Reduction and Analysis of VLSI Interconnect Circuits

用于 VLSI 互连电路阶数降低和分析的高效且性能有保证的方法

• 批准号: 0098275
• 负责人: Dian Zhou
• 金额: $ 39.42万
• 依托单位: University of Texas at Dallas
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2001
• 资助国家: 美国
• 起止时间: 2001-06-15 至 2005-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0098275&HistoricalAwards=false
• 关键词: Efficient; Performance; Guaranteed; Methods; Order

The advance of high-speed deep-submicron VLSI technology requires chipinterconnect and packaging to be modeled by distributed circuits. Such adetailed modeling level eventually results in large scale linear circuits tobe analyzed. Thus, the research objective is to develop new circuit orderreduction approach to evaluate the circuit performance and characteristicsin a reasonable time period and with a guaranteed performance, as requiredby real design practice.This project focuses on the following tasks:a) Develop new order reduction approaches that provide guaranteed performance;b) Derive correspondingly high efficiency numerical methods;c) Investigate the methods of maintaining preferred characteristics in thereduced order system as well as related implementation issues;d) Investigate wavelet method application in model order reduction andanalysis; ande) Contribute to graduate education at UT- Dallas and UNC-Charlotte.The expected research results will provide efficient and performanceguaranteed methods for order reduction and analysis of VLSI interconnectcircuits, and their corresponding algorithms and software for simulationtest. It will also contribute to engineering and science aspects by theexpected research results and software, as well as to graduate education.

高速深亚微米超大规模集成电路（VLSI）技术的发展要求用分布式电路来模拟芯片互连和封装。这种详细的建模水平最终导致大规模的线性电路进行分析。因此，本课题的研究目标是根据真实的设计实践的要求，开发新的电路降阶方法，以在合理的时间段内，在保证性能的前提下，评估电路的性能和特性，本课题的主要任务是：a）开发新的降阶方法，以保证性能;B）推导相应的高效数值方法;（3）研究降阶系统中保持优先特性的方法以及相关的实现问题：（4）研究小波方法在模型降阶与分析中的应用;本文的研究成果将为VLSI互连电路的降阶和分析提供有效的、有性能保证的方法，并提供相应的算法和软件用于仿真测试。 它还将通过预期的研究成果和软件，以及研究生教育，为工程和科学方面做出贡献。