The Application Of Importance Sampling In Rare Events For SRAM

The estimation of rare event probabilities is probably one of the most populartopics recently. Many events in our daily life are rare events such as the happening ofseismic sea wave, the bankrupt of a insurance company or a air crash… all these eventscan cause catastrophic losses, so it is very important for us to do research on theprobabilities of rare events.In this paper, we mainly study the problem of SRAM design. We first introducewhat is rare events and the importance of studying rare events, the Standard MonteCaro and the efficiency of estimations. Then we introduce the Importance Samplingand the Mixture Importance Sampling. The key point in this paper is that we do somechanges for the design of Mixture Importance Sampling, and then set up two examplesto compare the efficiency of these different methods. In the last section, we get theconclusion of this paper and some progresses remain after this paper.