abstract

• Agriculture has seen several technological transformations in recent years, allowing accelerated growth in production to meet the greater consumer demand for food. In particular, the adoption of the internet of things (IoT) is rapidly transforming the future of agriculture. In this paper, we present a data-driven climate prediction model for greenhouses using an effective and low-cost IoT-based spatial monitoring system. The IoT infrastructure comprises four stages: data gathering, transmission, analysis and processing of information, and visualization. This system was deployed and tested in a real greenhouse for three months, monitoring the temperature, relative humidity, and CO2 levels. The behavior of these environmental variables were predicted 24h in advance, obtaining an advantage in prediction accuracy. Additionally, we developed a spatial monitoring strategy based on packing density theory as a solution to the climate variability within greenhouses, offering a compromise between effectiveness and cost.

authors

• Delgado Sánchez Pablo
• Guadiana Alvarado Zoé Arturo
• Luna Rivera José Martín
• Villarreal-Guerrero F.

publication date

• 2023-01-01

published in

• IEEE Latin America Transactions  Journal

keywords

• Climate prediction, Internet of things, LPWAN, Machine learning, Precision agriculture, Smart greenhouse

Digital Object Identifier (DOI)

• 10.1109/TLA.2023.10128933

PubMed ID

start page

• 602

end page

• 611

volume

• 21

issue

• 4