Statistical circuit performance dependency analysis via sparse relevance kernel machine

Design optimization, verification, and failure diagnosis of analog and mixed-signal (AMS) circuits requires accurate models that can reliably capture complex dependencies of circuit performances on essential circuit and process parameters. We present a novel Bayesian learning technique, namely sparse relevance kernel machine (SRKM), for characterizing analog circuits with sparse statistical regression models. SRKM produces more reliable classification models learned from simulation data with a limited number of samples but a large number of parameters, and also computes a probabilistically inferred weighting factor quantifying the criticality of each parameter as part of the overall learning framework. As a result, it offers a powerful tool to enable variability modeling, failure diagnosis, and test development. The effectiveness of SRKM is demonstrated in an example of building a statistical variability model for analyzing the thermal shutdown feature of a data communication AMS system for automotive applications.