A Fast and Accurate approach for Full Chip Leakage Analysis of Nano-scale circuits considering Intra-die Correlations

This paper presents an accurate and efficient approach for estimating the full chip leakage in the presence of intra-die variations. We use an accurate model for leakage in which the leakage is exponentially dependent on a quadratic function of the device parameters. The intra-die correlations in the device parameters are accounted by representing the parameters in terms of abstract independent random variables using Karhunen-Loeve expansion. The total circuit leakage is computed using an efficient sum operation. Our results on ISCAS89 benchmark circuits show a speed up of up to 500times compared to Monte Carlo analysis, with average percentage difference in mean and variance being less than 1.5%