Worst case analysis of random circuits using Taylor models and Bernstein polynomials techniques

Yuchao Guo, Jingyi Cao, Xiaoxu Wang, Tongyu Ding^*, Liang Zhang

 

College of Information Engineering, Jimei University

401 Keli Building, 183 Yinjiang Road

Xiamen, China

tyding@jmu.edu.cn

 

Abstract: 

In this paper, a novel paradigm to optimize the Taylor models based worst-case analysis results of random circuits responses in time domain has been proposed. The methodology leverages the so-called Bernstein polynomials to yield a conservative, yet tight, prediction of the worst case bounds. This proposed framework has been used and verified in the time-domain analysis of an exemplary linear circuit, which demonstrated its feasibility and strength.

 

Keywords: 

worst case analysis, Taylor models, Bernstein polynomials, reliability, uncertainty

pages: 

71-74

DOI: 

10.35745/icice2018v2.018

Year: 

2018

Published in: 

2nd International Conference on Information, Communication and Engineering (ICICE 2018)

Full paper download: 

71-74.pdf