A Machine Learning Approach to Predict Supply and Temperature Variation Aware RF Integrated Circuit Interference Limitation

Abstract

In this paper we propose an accurate machine learning technique to predict statistical RF integrated interference limitation estimation with supply and temperature variation from artificial neural network and the regression based polynomial regression which exhibits efficient computation and error less than 1% for the modeling RF integrated circuit interference limitation. The accuracy of the proposed technique has been tested over several supply and temperature corners. It provides a bidirectional signoff flow between IC designer and EMC system designer at early design stage and achieving on radio frequency interference system-level success.