A Fog Cluster-based Interpretable AI Framework for Human Activity Recognition and Personalized Healthcare Monitoring

The innovations made possible by the Internet of Things have been of enormous use to the healthcare sector. These developments have made it feasible for a multitude of applications to collect huge amounts of information from users for evaluation in the cloud. On the other hand, immediate medical care need very low latency as well as which is something that traditional cloud computing is not able to supply. Fog computing is a solution that has emerged as a result of this issue. It allows for the processing of data that is nearer to its origin, which in turn reduces latency and the amount of bandwidth that is used. In the paper, a novel system is proposed that takes use of fog clusters. Fog clusters are groupings of fog nodes that collaborate with one another to accelerate and distribute the processing of tasks. According to the findings, fog clusters significantly cut down on latency and energy usage, which demonstrates the usefulness of fog technology in Internet of Things (IoT) systems that have stringent latency requirements. This study was carried out with the assistance of a simulation of an individual with a COPD monitoring system.