Scholarlink

<p>Modern smart healthcare demands computing infrastructures that deliver timely responses while remaining energy and cost-efficient. Cloud-only deployments often fall short for time-critical monitoring and interventions because delays and resource contention inflate latency. This work introduces HEAL-Opt, a multilayer edge, fog and cloud architecture that jointly optimizes latency, energy consumption, operational cost, and service-level agreement compliance in healthcare IoT systems. The proposed architecture is HyPRO, a hybrid predictive reinforcement scheduling mechanism that couples Holt Winters triple exponential smoothing for workload forecasting with an actor–critic policy for adaptive task placement and failure-aware replanning. Through iFogSim evaluations using healthcare workloads, HEAL-Opt reduces critical-task latency by up to <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(29\%\)</EquationSource> </InlineEquation>, lowers energy usage by <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(18\%\)</EquationSource> </InlineEquation>, and decreases SLA violations by <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(42\%\)</EquationSource> </InlineEquation>, outperforming state-of-the-art baselines. The findings prove that the proposed system is a scalable, intelligent foundation for responsive healthcare computing ecosystems.</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

Optimizing smart healthcare systems via integrated edge, fog, and cloud computing

  • 期刊论文

  • 发表日期：

    2026年5月13日

查看全文
  • Navneet Kumar Rajpoot,
  • Prabh Deep Singh,
  • Bhaskar Pant

摘要

Modern smart healthcare demands computing infrastructures that deliver timely responses while remaining energy and cost-efficient. Cloud-only deployments often fall short for time-critical monitoring and interventions because delays and resource contention inflate latency. This work introduces HEAL-Opt, a multilayer edge, fog and cloud architecture that jointly optimizes latency, energy consumption, operational cost, and service-level agreement compliance in healthcare IoT systems. The proposed architecture is HyPRO, a hybrid predictive reinforcement scheduling mechanism that couples Holt Winters triple exponential smoothing for workload forecasting with an actor–critic policy for adaptive task placement and failure-aware replanning. Through iFogSim evaluations using healthcare workloads, HEAL-Opt reduces critical-task latency by up to \(29\%\) , lowers energy usage by \(18\%\) , and decreases SLA violations by \(42\%\) , outperforming state-of-the-art baselines. The findings prove that the proposed system is a scalable, intelligent foundation for responsive healthcare computing ecosystems.