DocumentCode :
3287060
Title :
A study of parametric yield estimation by uniform design sampling
Author :
Jing, Minge ; Wang, Jun Ping ; Chen, Ke Dong ; Hao, Yue
Author_Institution :
Res. Inst. of Microelectron., Xidian Univ., Xi´´an, China
Volume :
2
fYear :
2004
fDate :
18-21 Oct. 2004
Firstpage :
1068
Abstract :
A novel yield estimation method is proposed based on uniform design sampling method. Compared with conventional statistical methods based on Monte Carlo sampling method, this method needs only few circuit simulations to have a valuable estimation. Furthermore, the algorithm to generate samples by uniform design method is very simple, which greatly reduces computational complexity. A comparison of uniform design sampling method with some popular Monte Carlo based methods is made in this paper to show the efficiency of the new method. Finally, a numerical and a circuit example are presented to demonstrate the advantages of the proposed method.
Keywords :
Monte Carlo methods; circuit simulation; integrated circuit design; integrated circuit yield; Monte Carlo sampling; acceptability region; circuit simulation; computational complexity; parametric yield estimation; statistical methods; uniform design sampling; Circuit simulation; Density functional theory; Design methodology; Mathematics; Microelectronics; Monte Carlo methods; Sampling methods; Statistical analysis; Very large scale integration; Yield estimation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Solid-State and Integrated Circuits Technology, 2004. Proceedings. 7th International Conference on
Print_ISBN :
0-7803-8511-X
Type :
conf
DOI :
10.1109/ICSICT.2004.1436691
Filename :
1436691
Link To Document :
بازگشت