TinyML for Empowering Low-Power IoT Edge Consumer Devices

IF 4.3 2区计算机科学 Q1 ENGINEERING, ELECTRICAL & ELECTRONIC

IEEE Transactions on Consumer Electronics Pub Date : 2025-01-03 DOI:10.1109/TCE.2024.3482353

Rutvij H. Jhaveri;Hao Ran Chi;Huaming Wu

引用次数: 0

Abstract

Pervasive Artificial Intelligence (AI) has been promoted to be applicable to multiple services and markets, based on the recent surge in AI and machine learning (ML) techniques. Together with the fact that the market size of edge computing has been boosted to 16 billion USD last year (and a forecast to reach more than 200 billion USD by 2030), TinyML will be one of the main forces to embrace the new era of pervasive AI, by embedding the main operations (e.g., training, modeling, and others) in edge computing, relying on its relatively short physical distance to the users/end devices. Therefore, TinyML has promised to support ultra-low latency, enhanced security/privacy, highly demanded scalability, and potentially sustainability by reducing the frequency accessing centralized cloud computing.

查看原文本刊更多论文

TinyML为低功耗物联网边缘消费设备提供支持

基于最近人工智能和机器学习（ML）技术的激增，普及人工智能（AI）被推广为适用于多种服务和市场。再加上去年边缘计算的市场规模已经提升到160亿美元（预计到2030年将超过2000亿美元），TinyML将成为拥抱普及人工智能新时代的主要力量之一，依靠其与用户/终端设备相对较短的物理距离，将主要操作（例如培训，建模等）嵌入边缘计算。因此，TinyML承诺通过减少访问集中式云计算的频率来支持超低延迟、增强的安全性/隐私性、高要求的可扩展性和潜在的可持续性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

IEEE Transactions on Consumer Electronics 工程技术-电信学

CiteScore

7.70

自引率

9.30%

发文量

审稿时长

3.3 months

期刊介绍： The main focus for the IEEE Transactions on Consumer Electronics is the engineering and research aspects of the theory, design, construction, manufacture or end use of mass market electronics, systems, software and services for consumers.