A Dynamic Cost-Efficient Task Offloading Framework for Resource-Constrained Edge-Based Smart Healthcare Systems

The advancements in Internet of Healthcare Things ( $IoHT$ ) has significantly transformed the healthcare industry. To meet the computing needs of the healthcare industry, the Multi-access Edge Computing (MEC) serves a favorable solution towards processing of offloaded computational workloads from resource constrained IoHT devices. As the offloaded data to MEC servers increases in volume, it becomes difficult for the resource constrained servers to process these tasks for advanced health analytics. In this view, the present work proposes a dynamic cost-aware solution for handling the computationally intensive healthcare data packets offloaded by resource constrained IoHT applications. Here, an adaptive offloading technique is put forth that makes use of the $M$ / $M$ / c queuing model and models the system as a non-birth death stochastic process. The effectiveness of the suggested CARE framework is investigated in conjunction with benchmark computing platforms like MEC and the conventional remote Cloud computing platform under various sensitivity criteria, like the ideal workload, the response time, and the system processing cost. Our simulation results showed that the suggested framework outperformed the other two approaches, making it appropriate for time-sensitive IoHT applications.