Scholarlink

<p>Security is becoming more important as the Internet of Things (IoT) continues to grow in popularity. Some of the current efforts are unable to secure data transmitted over wireless channels that are not encrypted. To propose a novel energy-efficient data transfer mechanism for the IoT healthcare system leveraging fog computing, deep learning (DL), and advanced optimization algorithms for solving existing issues. The system pre-processes the data from the Cleveland Heart disease dataset using Data normalization and Missing value imputation, then classifies the health states using Capsule autoencoder (CapsA). Security is ensured through an adaptive proxy homomorphic re-encryption (AProHE) algorithm for enabling privacy preserving computation. The cluster heads are then selected and routed using the Integrated Dwarf mongoose assisted mud ring optimization algorithm (IDM-MAO), which selects the most efficient way to the cloud platform. Experimental results show that the proposed system attains 98.76% accuracy, 97.99% precision, 96.79% F1-score, and 98% energy efficiency, outperforming existing methods. The proposed system offers a scalable, low-latency, and secure solution for real-time IoT healthcare monitoring, addressing existing issues.</p>

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

A novel energy efficient IoT healthcare system using integrated optimization with authentication mechanism

  • 期刊论文

  • 发表日期：

    2025年9月29日

查看全文
  • Arthi Kalidasan,
  • B. Chidhambararajan

摘要

Security is becoming more important as the Internet of Things (IoT) continues to grow in popularity. Some of the current efforts are unable to secure data transmitted over wireless channels that are not encrypted. To propose a novel energy-efficient data transfer mechanism for the IoT healthcare system leveraging fog computing, deep learning (DL), and advanced optimization algorithms for solving existing issues. The system pre-processes the data from the Cleveland Heart disease dataset using Data normalization and Missing value imputation, then classifies the health states using Capsule autoencoder (CapsA). Security is ensured through an adaptive proxy homomorphic re-encryption (AProHE) algorithm for enabling privacy preserving computation. The cluster heads are then selected and routed using the Integrated Dwarf mongoose assisted mud ring optimization algorithm (IDM-MAO), which selects the most efficient way to the cloud platform. Experimental results show that the proposed system attains 98.76% accuracy, 97.99% precision, 96.79% F1-score, and 98% energy efficiency, outperforming existing methods. The proposed system offers a scalable, low-latency, and secure solution for real-time IoT healthcare monitoring, addressing existing issues.