Title :
Monte Carlo simulation in VDDmin modeling across Fab process
Author :
Guo, Hugo ; Yang, Yang ; Bowen, Jiang
Author_Institution :
Shanghai Product Eng. Team, LSI Corp., Shanghai, China
Abstract :
VDDmin is defined as the lowest supply voltage at which a device can produce correct logic states. It is typically obtained by performing a binary or linear supply voltage search. VDDmin or very-low voltage (VLV) testing is a non-IDDQ defect based test method that has been shown to be an effective method to screen manufacturing defects [1, 2, 3, 4]. It is taken as an indicator of device performance and reliability. Monte Carlo simulation is a random sampling based computational methodology, used extensively for engineering analysis [5, 6, 7]. This paper discuss a VDDmin determination methodology base on-Monte Carlo simulation, case study is presented to illustrate that this approach can precisely predict yield impact due to process shift.
Keywords :
Monte Carlo methods; low-power electronics; random processes; sampling methods; semiconductor device reliability; semiconductor device testing; Fab process; Monte Carlo simulation; VDDmin determination methodology; VDDmin modeling; VLV testing; binary suppy voltage; linear supply voltage; logic state; nonIDDQ defect; random sampling; very-low voltage testing; Computational modeling; Correlation; Foundries; Monte Carlo methods; Performance evaluation; Testing; Isat; Monte Carlo; VDDmin; Yield;
Conference_Titel :
Quality Electronic Design (ASQED), 2012 4th Asia Symposium on
Conference_Location :
Penang
Print_ISBN :
978-1-4673-2687-2
DOI :
10.1109/ACQED.2012.6320499