Performance robustness analysis of VLSI circuits with process variations based on Kharitonov's theorem

In today's VLSI technology, the process variations are unavoidable. This paper proposes an efficient analysis approach for exploring the worst case performance for VLSI circuits with severe parameter value variations due to nano-scale process. Inspired by Kharitonov's theorem, the described method dramatically reduces the computational burden to only evaluate several critical Kharitonov-type interval transfer functions. The computational efficiency of the method is demonstrated by two practical VLSI circuits.