Automated Design of Analog Circuits using Machine Learning Techniques
R. Sindhiya Devi, Gourav Tilwankar, Rajesh Zele
Abstract
This work presents methodology for an automated design of analog circuits using global Artificial Neural Network (ANN) for an optimised dataset. The optimised dataset is generated using simulation based g <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m</inf> /I <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">d</inf> technique, which reduces the dataset size and also the time required for data collection and analysis. Automated analog circuit design is implemented using ANN based supervised learning technique for a common source amplifier and a two stage single-ended opamp. The results obtained are compared with unsupervised (Reinforcement Learning algorithm) and supervised learning technique (Genetic Algorithm based local ANN). The comparison results shows that the proposed g <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m</inf> /I <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">d</inf> technique based ANN model gives a better accuracy in terms of score and mean square error (MSE).