Scholarlink

This study explores the use of machine learning algorithms to estimate power consumption in CMOS circuits, aiming to offer a quicker and more efficient alternative to traditional simulation-based approaches. The research evaluates several algorithms, including Support Vector Machines (SVM), Linear Regression, and Random Forest, comparing them based on accuracy metrics, Root Mean Square Error (RMSE), and Mean Squared Error (MSE). The analysis also includes residual plots to assess the models’ performance in capturing data patterns. Results indicate that Random Forest outperforms the other algorithms, followed closely by SVM, with Linear Regression delivering acceptable estimates. These findings underscore the promise of machine learning techniques in improving the accuracy and efficiency of power estimation in integrated circuit design.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning for Power Analysis: A New Paradigm in CMOS VLSI Design

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Naiyya Mittal,
  • Srishty Sharma,
  • Tithi Pandey,
  • Shobha Sharma

摘要

This study explores the use of machine learning algorithms to estimate power consumption in CMOS circuits, aiming to offer a quicker and more efficient alternative to traditional simulation-based approaches. The research evaluates several algorithms, including Support Vector Machines (SVM), Linear Regression, and Random Forest, comparing them based on accuracy metrics, Root Mean Square Error (RMSE), and Mean Squared Error (MSE). The analysis also includes residual plots to assess the models’ performance in capturing data patterns. Results indicate that Random Forest outperforms the other algorithms, followed closely by SVM, with Linear Regression delivering acceptable estimates. These findings underscore the promise of machine learning techniques in improving the accuracy and efficiency of power estimation in integrated circuit design.