Scholarlink

<p>Cloud computing enables the delivery of computing services and resources over the Internet, with cloud data centers providing the essential infrastructure to support diverse applications and workloads. The Internet of Things (IoT) involves physical devices embedded with sensors that continuously collect and transmit data to cloud data centers for processing and storage. As the number of IoT devices rapidly increases, the massive volume of data generated poses significant challenges for cloud data centers, particularly in terms of network congestion and resource utilization. To address these challenges, a novel Laplacian Kernel Clustering-based Censored Regressive Weighted Queuing (LKC-CRWQ) model is proposed to achieve congestion-aware data distribution in healthcare IoT environments. In this framework, numerous IoT sensors attached to a patient’s body continuously monitor and collect vital health parameters such as temperature, pulse rate, heart rate, respiration rate, and blood pressure. The collected data are transmitted to cloud data centers through the Internet for storage and processing. Within the cloud, a Laplacian Kernelized Expectation Maximization Clustering algorithm is employed to classify incoming data traffic as either normal or abnormal. The data centers store only the normal traffic and discard the abnormal traffic, thereby optimizing storage resources, preventing interference with regular operations, and reducing operational costs. Subsequently, a Censored Regressive Weighted Fair Queuing (CRWQ) strategy is implemented to manage bandwidth effectively across multiple cloud data centers. This congestion-aware mechanism enhances transmission efficiency, minimizes latency, and reduces packet loss. Experimental results demonstrate that the proposed LKC-CRWQ model achieves significant improvements compared to existing methods, including 31% and 6% higher throughput and data delivery rate, respectively. Moreover, 18%, 19%, 27%, and 67% reductions in storage cost, computational cost, latency, and data loss rate, respectively. These results confirm the model’s effectiveness in ensuring efficient, reliable, and congestion-aware data distribution in healthcare IoT cloud environments.</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

Cloud-based congestion-aware data distribution for healthcare IoT using Laplacian kernel clustering and censored weighted queuing

  • 期刊论文

  • 发表日期：

    2026年3月5日

查看全文
  • Ahmed M. Khedr,
  • Sheeja Rani S.

摘要

Cloud computing enables the delivery of computing services and resources over the Internet, with cloud data centers providing the essential infrastructure to support diverse applications and workloads. The Internet of Things (IoT) involves physical devices embedded with sensors that continuously collect and transmit data to cloud data centers for processing and storage. As the number of IoT devices rapidly increases, the massive volume of data generated poses significant challenges for cloud data centers, particularly in terms of network congestion and resource utilization. To address these challenges, a novel Laplacian Kernel Clustering-based Censored Regressive Weighted Queuing (LKC-CRWQ) model is proposed to achieve congestion-aware data distribution in healthcare IoT environments. In this framework, numerous IoT sensors attached to a patient’s body continuously monitor and collect vital health parameters such as temperature, pulse rate, heart rate, respiration rate, and blood pressure. The collected data are transmitted to cloud data centers through the Internet for storage and processing. Within the cloud, a Laplacian Kernelized Expectation Maximization Clustering algorithm is employed to classify incoming data traffic as either normal or abnormal. The data centers store only the normal traffic and discard the abnormal traffic, thereby optimizing storage resources, preventing interference with regular operations, and reducing operational costs. Subsequently, a Censored Regressive Weighted Fair Queuing (CRWQ) strategy is implemented to manage bandwidth effectively across multiple cloud data centers. This congestion-aware mechanism enhances transmission efficiency, minimizes latency, and reduces packet loss. Experimental results demonstrate that the proposed LKC-CRWQ model achieves significant improvements compared to existing methods, including 31% and 6% higher throughput and data delivery rate, respectively. Moreover, 18%, 19%, 27%, and 67% reductions in storage cost, computational cost, latency, and data loss rate, respectively. These results confirm the model’s effectiveness in ensuring efficient, reliable, and congestion-aware data distribution in healthcare IoT cloud environments.