Scholarlink

<p>Health-focused Internet of Things (IoT) systems are essential components of smart healthcare infrastructures but face challenges such as limited computational resources, constrained energy supply, rapid growth of medical data, and increased latency under high-traffic conditions. To address these issues, this study proposes A-CAGE (Adaptive Cooperative Game-based Energy Optimization), a novel framework that dynamically manages resources while fostering cooperative interactions among IoT devices. Beyond its cooperative decision-making and optimization modules, A-CAGE uniquely integrates a Worst-Case Energy Drift–based energy shock forecasting module, capable of detecting abrupt and severe energy drops in advance. The core innovation of this module lies in an endogenous, adaptive, and risk-aware safe energy threshold coupled with a continuous energy shock risk indicator, providing a closed-form, computationally efficient O(H) solution that enables proactive decision-making, energy–performance stability, and reduced service interruptions. Simulation results in OMNeT + + demonstrate that A-CAGE achieves 13.42% lower energy consumption, 17.18% reduced latency, and 1.21% higher task success rate compared to state-of-the-art approaches, highlighting its potential for efficient resource management and enhanced service delivery in future healthcare IoT networks.</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-CAGE: a novel no-regret learning and game-theoretic approach for reducing energy consumption in IoHT networks

  • 期刊论文

  • 发表日期：

    2026年6月30日

查看全文
  • Marjan Mahmoudi,
  • Behrang Barekatain,
  • Hamid R. Arabnia

摘要

Health-focused Internet of Things (IoT) systems are essential components of smart healthcare infrastructures but face challenges such as limited computational resources, constrained energy supply, rapid growth of medical data, and increased latency under high-traffic conditions. To address these issues, this study proposes A-CAGE (Adaptive Cooperative Game-based Energy Optimization), a novel framework that dynamically manages resources while fostering cooperative interactions among IoT devices. Beyond its cooperative decision-making and optimization modules, A-CAGE uniquely integrates a Worst-Case Energy Drift–based energy shock forecasting module, capable of detecting abrupt and severe energy drops in advance. The core innovation of this module lies in an endogenous, adaptive, and risk-aware safe energy threshold coupled with a continuous energy shock risk indicator, providing a closed-form, computationally efficient O(H) solution that enables proactive decision-making, energy–performance stability, and reduced service interruptions. Simulation results in OMNeT + + demonstrate that A-CAGE achieves 13.42% lower energy consumption, 17.18% reduced latency, and 1.21% higher task success rate compared to state-of-the-art approaches, highlighting its potential for efficient resource management and enhanced service delivery in future healthcare IoT networks.