A survey on binary decision diagram approaches to symbolic analysis of analog integrated circuits
A survey on binary decision diagram approaches to symbolic analysis of analog integrated circuits
复制标题
DOI：
10.1007/s10470-011-9773-8
复制
发表时间：
2011-09
影响因子：
1.4
通讯作者：
G. Shi
中科院分区：
文献类型：
--
作者：
G. Shi
将符号技术应用于模拟电路分析是一个传统的研究课题，已经持续了半个多世纪。在过去的十年里，专门为大型模拟集成电路开发的符号技术取得了重大进展。介绍的关键方法是一种称为二叉决策图(BDD)的数据结构，它最初是为逻辑设计和验证而建立的。BDD技术在模拟电路分析中的应用具有以下特点：(1)它是一种紧凑的数据结构，可以消除符号分析中的数据冗余。(2)为隐枚举法提供了一种机制，避免了符号分析中常用的穷举。(3)BDD的数值计算非常有效，使其成为重复分析的极好手段。更高级的功能还有待探索。这篇综述汇集了过去十年发表的重要研究成果，并提供了将BDD应用于模拟电路分析的基本原则的教程概述。讨论了一些对开发未来模拟设计自动化工具有潜在价值的新方向，并给出了一个设计实例来说明符号技术的应用。
Applying symbolic techniques for analog circuit analysis is a traditional research subject, which has lasted for over half a century. The past decade has witnessed a significant advancement of the symbolic techniques developed specifically for large analog integrated circuits. The key methodology introduced is a data structure called binary decision diagram (BDD) which was established originally for logic design and verification. The application of the BDD technique for analog circuit analysis has the following features: (1) It is a compact data structure so that data redundancy in symbolic analysis can be eliminated. (2) It provides a mechanism for implicit enumeration method so that exhaustive enumeration commonly performed in symbolic analysis can be avoided. (3) Numerical evaluation on a BDD can be made extremely efficient, making it an excellent means for repetitive analysis. More advanced features are yet to be explored. This survey brings together the significant research results published in the past decade and provides a tutorial overview on the basic principles of applying BDD to analog circuit analysis. Some new directions that are potentially valuable for developing future analog design automation tools are discussed and a design example is given to illustrate the application of symbolic techniques.