To solve the crosstalk noise problem in deep-submicron technologies, a statistical method for analyzing crosstalk noise with reduced distributed RC-π model is proposed in this paper. First, quiet aggressor net and tree branch reduction techniques are introduced into the distributed RC-π model, and a new spatial correlation model for both Gaussian and non-Gaussian process variations among segments is created. Then, principal components analysis (PCA) and independent component analysis (ICA) techniques are applied to reduce correlations of process variations. Finally, the moment matching scheme is used to obtain the probability density function (PDF) of crosstalk noise in victim coupled with multiple aggressors. Experimental results show that our method maintains the efficiency of previous approaches, and significantly improves on their accuracy.
