Piecewise-Linear Modeling of Analog Circuits using Trained Feed-Forward Neural Networks and Adaptive Clustering of Hidden Neurons

This paper presents a new technique for automatically creating analog circuit models. The method extracts piecewise linear models from trained neural networks. A model is a set of linear dependencies between circuit performances and design parameters. The paper illustrates the technique for an OTA circuit an amplifier circuit widely used in filters and A/D converters for which models for gain and bandwidth were automatically generated. As experiments show, the obtained models have a simple form that accurately fits the sampled points and the behavior of the trained neural networks. These models are useful for fast simulation of systems with non-linear behavior and performances.