A novel approach to efficient yield estimation for microwave integrated circuits

摘要

Latin Hypercube Sampling (LHS) is proposed as an alternative to Primitive Monte Carlo (PMC) for yield analysis. It is shown that the LHS method improves accuracy and confidence in yield estimate and yet requires smaller number of simulations compared to PMC technique. Yield analysis of a 8-12 GHz microwave small signal amplifier demonstrate the viability of the proposed approach.