Ambient Intelligence assisted fog computing for industrial IoT applications

Ambient Intelligence (AmI) is a key concept that uses environmental and contextual information for improving the application experience. The adaptive approach used by AmI offers several benefits for the Industrial Internet of Things (IIoT) such as enhancing machine productivity and improving different processes. IIoT applications involve many tasks that need to be computed in real-time using fog nodes, thus efficient computing techniques are a major challenge in IIoT. In this paper, we propose an ambient intelligence-assisted computing technique for Industrial IoT to maximize the number of served tasks and reduce task outages at fog nodes. We utilize contextual information such as transmission rate and task delay requirements to efficiently offload the tasks from machine-embedded sensors to the fog nodes. We propose an adaptive computing resource unit sizing to serve an individual task at the fog node. Moreover, we propose a many-to-one matching-based algorithm for mapping between tasks and computing resources. We perform extensive simulations to show that the proposed algorithm improves the number of served tasks by 54% and computational resource utilization at the fog nodes by 47%.