A Fog Computing-based mechanism to provide reliability in IoT : A case study in a coffee smart farming environment

Abstract

Reliability is essential in Smart Farming supported by the Internet of Things. A Fog Computing approach is pivotal for Smart Farming since it allows farmers to monitor and improve crop production by getting closer cloud capabilities at the edge of the network. The provisioning of reliability in farms is critical since the failure of a fog node can cause interruptions for farmers’ decision-making services. Smart Farms’ unprotection may cause significant economic losses and low yields of production. Moreover, making decisions based on inaccurate data can diminish the quality of crops and, consequently, lose money. This master dissertation addresses the Fog-based Smart Farms’ unprotection from two approaches: system and data reliability. On the one hand, the dissertation introduces an optimization model for traditional protection schemes 1:1 and 1:N for meeting reliability in Smart Farms to minimize deployment cost to farmers giving heterogeneous fog nodes. On the other hand, we propose an IoT-Fog-Cloud architecture that incorporates a mechanism based on Machine Learning to detect outliers and another based on interpolation for inferring data intended to replace outliers. The proposed approaches were eval-uated by conducting a case study in a network based on the proposed and deployed architecture at a Colombian Coffee Smart Farm. The results show the effectiveness of the proposed approaches regarding protection schemes in FN-based smart farms guaranteeing high reliability to improve the operation of farms; and high Accuracy, Precision, and Recall, as well as low False Alarm Rate and Root, Mean Squared Error when detecting and replacing outliers with inferred data.