Ali Hafiz Munsub, Bomgni Alain Bertrand, Bukhari Syed Ahmad Chan, Hameed Tahir, Liu Jun
Information Systems in Computer Science Department, SUNY at Binghamton, Binghamton, NY USA.
Department of Biomedical Engineering, University of South Dakota, Vermillion, SD USA.
Mob Netw Appl. 2023 Mar 3:1-15. doi: 10.1007/s11036-023-02107-9.
Healthcare services become increasingly technology dependent every passing day such as the Internet of Things (IoT), Fog Computing, 5th generation (5G) and beyond communications, etc. They enable the processing and exchange of huge volumes of healthcare data whose integrity and real-time delivery are critical for healthcare services. Optimal power consumption in such essential healthcare infrastructure is critical for the well-being of patients and crucial to reduce the operational cost of healthcare facilities. In this paper, a Fog node has been introduced in an IoT healthcare infrastructure with power consumption as a key deciding factor. This work proposed a mathematical formulation to decide the deployment of two heterogeneous gateways in the healthcare infrastructure. The target of optimization is to minimize transmission power and infrastructure costs. Two swarm intelligence-based algorithms have been used to solve the computationally challenging optimization problem. These evolutionary algorithms are a discrete fireworks algorithm and a discrete artificial bee colony algorithm with an ensemble of local search methods. Their performance is compared against the genetic algorithm. The simulation results demonstrate a saving of up to 33 percent in power consumption in the proposed healthcare infrastructure that can significantly improve healthcare data communications and its operational costs.
医疗保健服务日益依赖技术,诸如物联网(IoT)、雾计算、第五代(5G)及以上通信等。这些技术使得海量医疗数据的处理和交换成为可能,而这些数据的完整性和实时传输对医疗保健服务至关重要。在这种至关重要的医疗基础设施中,优化功耗对于患者的福祉至关重要,同时对于降低医疗机构的运营成本也至关重要。在本文中,在以功耗为关键决定因素的物联网医疗基础设施中引入了一个雾节点。这项工作提出了一种数学公式,用于确定医疗基础设施中两个异构网关的部署。优化的目标是最小化传输功率和基础设施成本。使用了两种基于群体智能的算法来解决计算上具有挑战性的优化问题。这些进化算法是离散烟花算法和带有局部搜索方法集合的离散人工蜂群算法。将它们的性能与遗传算法进行了比较。仿真结果表明,在所提出的医疗基础设施中,功耗可节省高达33%,这可显著改善医疗数据通信及其运营成本。