Parameter Optimization Of Analog Circuit Implementation Using Neural Network Models
Designing analog integrated circuits is frequently thought of as a time-consuming task due to the fact that transistor and passive component dimensions have a significant impact on analog performance. Several automation techniques have been used over the past ten years to conduct extensive research on the analog circuit front-end design cycle. Machine learning is now a desirable and practical solution for everyone due to significant advancements in high-performance computing technology. The goals of this survey are to provide a comprehensive overview of the most advanced machine learning techniques currently used at the analog circuit scale and analyze how well they perform in terms of achieving the desired goals and identify the most critical areas for future research. Find nodes in a neural network that maximize the objective function and minimize the objective function. To demonstrate the effectiveness of the optimization strategy using outcomes from integrated circuit applications of ANNs, we classify digital data that has been simulated in PSpice.