AUTHOR=Cappiello Cinzia , Meroni Giovanni , Pernici Barbara , Plebani Pierluigi , Salnitri Mattia , Vitali Monica , Trojaniello Diana , Catallo Ilio , Sanna Alberto 

TITLE=Improving Health Monitoring With Adaptive Data Movement in Fog Computing

JOURNAL=Frontiers in Robotics and AI

VOLUME=Volume 7 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/robotics-and-ai/articles/10.3389/frobt.2020.00096

DOI=10.3389/frobt.2020.00096

ISSN=2296-9144

ABSTRACT=Pervasing sensing is increasing the possibility to monitor the status of patients, not only when they are hospitalized but also during home recovery. As a consequence, lots of data are collected, thus they are available for multiple purposes. If, on the one hand, operations can take advantage of timely and fine-grained data, on the other hand the huge amount of collected data can be useful also for analytics. Nevertheless, these data could become useless for two reasons: data quality and performance issues. Firstly, if the quality of the gathered values is low, the processing tasks might yield not significant results. Secondly, if the system does not guarantee appropriate performance, the results might not be delivered at the right time.

The goal of this paper is to propose a data utility model that both considers the impact of the  quality of the data sources (e.g., sensed data, biographical data, clinical history) on the expected results and allows improving performance through a utility-based data management in a Fog environment.
As regards data quality, our approach aims to consider that it is a context-dependent issue: a given dataset can be considered useful for an application and inadequate for another application. For this reason, we suggest a context-dependent quality assessment considering dimensions such as accuracy, completeness, consistency, and timeliness and claiming that different applications have different quality requirements to consider. Managing data in Fog computing also requires a particular attention to quality of service requirements. For this reason, we include QoS aspects in the data utility model, like availability, response time, and latency. Based on the proposed data utility model, we propose an approach based on a goal model able to identify, when one or more quality of service or data quality dimensions are violated, which is the best action to be enacted for addressing such violation.

The proposed approach is evaluated with a real dataset, properly anonymized, obtained within the experimental procedure of a research project in which a device with a set of sensors is used to collect motion and environmental data associated with the daily physical activities of healthy young volunteers.