On-Demand Provisioning of Wearable Sensors Data Processing Services in Edge Computing

The wearable sensor applications and services are highly sensible to computation latency. The infrastructure that processes data closer to its source or end user named as multi-access edge computing (MEC) has emerged to push the data processing services away from the centralized nodes to the network edge, enabling wearable sensor applications with the geolocation awareness. In this paper, we introduce on-demand provisioning of wearable sensors processing services in MEC system, a continuous deployment strategy that offer availability of wearable data processing services packaged as an application ready to be provisioned on-demand by the Docker container virtualization technique at the edge of the network. This allows, thus, fulfilling different requirements, such as low latency and accuracy.