This paper proposes a new approach to analyze crosstalk of coupled interconnects in the presence of process variations. The suggested method translates correlated process variations into orthogonal random variables by principle component analysis (PCA). combined with polynomial chaos expression (PCE), the technique utilizes Stochastic Collocation Method (SCM) to analyze the system response of coupled interconnects. A finite representation of interconnect crosstalk is obtained by projecting the infinite series representation onto a finite dimensional subspace. Experimental results demonstrate that the approach match well with HSPICE. The differences between the crosstalk obtained from the analytical method and HSPICE is about 2% or less. Furthermore the new approach shows good computation efficiency: much less running time has been observed over Monte Carlo SPICE simulation.