Estimating the Probability Density Function of Critical Path Delay via Partial Least Squares Dimension Reduction

Yu Ben and Costas Spanos
UC Berkeley


We propose a method based on partial least squares (PLS) regression to estimate the probability density function of the critical path delay. The method works on a reduced problem facilitated by PLS regression and requires only ~10^2 samples to achieve satisfactory accuracy. The method is verified by simulations on ISCAS’85 benchmark circuits.