Efficient Task-Offloading in IoT-Fog Based Health Monitoring System

Abstract

Recent advancement applications have more computation-intensive, and data-intensive tasks are delay-sensitive. In IoT-Cloud-based healthcare architecture, data is aggregated using edge devices and sent to the cloud for processing and analysis. Furthermore, we need to transfer the data information out of the network for each event. Hence it is a delay-sensitive process that is not useful for instant processing and is unacceptable for healthcare applications. To overcome this problem, we have focused on a fog layer between smart devices and the cloud layer. Additionally, we use the Bayesian Belief Network's classification technique in the fog layer for task offloading. This paper focuses on reducing the response time using the BBN classifier after task offloading and increasing the system's stability using fog computing. In the simulation result, we compare the cloud-based and fog-based models in which the fog-based model is dominant over the cloud- based. This fog-based approach is based on real-time data processing at the local network. Hence it is practically possible and acceptable to get an instant result.