A Hierarchical Reliability Simulation Methodology for AMS Integrated Circuits and Systems

In this paper, we propose a methodology for simultaneously analyzing the ageing effects and process variations. Nominal ageing simulation and statistical methods are applied to reliability simulation of AMS integrated circuits and systems. Response surface modeling (RSM) is used to build direct relationship between process parameters and circuit/system performances. With Varied/fixed RSMs, designers can have reliability information of designed circuit/system. Also, this methodology has been developed with behavioral modeling for reliability consideration of large AMS circuits and systems (e.g., modulator, RF front-end). Conventional Monte-Carlo (MC) method is infeasible in these complex circuits and systems. The methodology is validated with a series of circuits and systems in 65 nm CMOS technology: simple current mirrors, a dynamic comparator and a 2nd order continuous-time analog-to-digital modulator. It is shown that this methodology can provide designers with reliability information graphically with a general perspective. It can achieve better simulation efficiency than traditional Monte-Carlo analysis, while still guaranteeing good simulation accuracy.