Keeping data at the edge of smart irrigation networks: A case study in strawberry greenhouses

This data was imported from Scopus:

Authors: Angelopoulos, C.M., Filios, G., Nikoletseas, S. and Raptis, T.P.

Journal: Computer Networks

Volume: 167

ISSN: 1389-1286

DOI: 10.1016/j.comnet.2019.107039

© 2019 Elsevier B.V. Strawberries are widely appreciated for their characteristic aroma, bright red color, juicy texture, and sweetness. They are, however, among the most sensitive fruits when it comes to the quality of the end product. The recent commercial trends show a rising number of farmers who directly sell their products in the market and are more interested in using smart solutions for a continuous control of the factors that affect the quality of the final product. Cloud-based approaches for smart irrigation have been widely used in the recent years. However, the network traffic, security and regulatory challenges, which come hand in hand with sharing the crop data with third parties outside the edge of the network, lead strawberry farmers and data owners to rely on global clouds and potentially lose control over their data, which are usually transferred to third party data centers. In this paper, we follow a three-step methodological approach in order to design, implement and validate a solution for smart strawberry irrigation in greenhouses, while keeping the corresponding data at the edge of the network: (i) We develop a small-scale smart irrigation prototype solution with off-the-shelf hardware and software equipment, which we test and evaluate on different kinds of plants in order to gain useful insights for larger scale deployments, (ii) we introduce a reference network architecture, specifically targeting smart irrigation and edge data distribution for strawberry greenhouses, and (iii) adopting the proposed reference architecture, we implement a full-scale system in an actual strawberry greenhouse environment in Greece, and we compare its performance against that of conventional strawberries irrigation. We show that our design significantly outperforms the conventional approach, both in terms of soil moisture variation and in terms of water consumption, and conclude by critically appraising the costs and benefits of our approach in the agricultural industry.

The data on this page was last updated at 05:09 on February 24, 2020.