EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Large-Scale IoT Telemetry Framework for Edge-Aware Forecasting and Diagnostic State Tracking in Precision Apiculture

Sebastian Gorecki, Slawomir Hausman, Piotr Korbel, James Brusey

IEEE Access · 2026 · 14 · 138619-138641

DOI: 10.1109/access.2026.3731411 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Sebastian Gorecki und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Institute of Electronics, Lodz University of Technology and Intelligent Hives Sp. z o.o.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Precision apiculture increasingly relies on Internet of Things (IoT) telemetry, yet forecasting, short-horizon state tracking, and embedded deployment are often evaluated separately. This study presents a deployment-oriented BeeHUB framework integrating multimodal sensing, hive-weight forecasting, diagnostic state tracking, and physically validated edge inference. Two non-equivalent analysis cohorts are reported separately: a full-period cohort of 2,531,844 records from 206 device streams and a weather-enriched cohort of 1,645,074 observations from 116 GPS-resolved hives collected between January 2022 and February 2025. In the 72-h full-period experiment, Linear Regression and Ridge Regression achieved a mean absolute error of 2.111 kg and an $R^{2}$ of 0.962, while a persistence reference remained stronger with a mean absolute error of 1.293 kg. The exact frozen linear-model artifacts were converted to allocation-free C implementations and evaluated on an ESP32-D0WD-V3 platform. Each implementation increased compiled Flash usage by 1216 bytes and required no additional static RAM. Median kernel latency was approximately $11.2~\mu $ s, while complete feature processing and inference required approximately $60.1~\mu $ s. Numerical equivalence with the original Python artifacts was confirmed for 1024 test vectors, with maximum absolute differences below $9\times 10^{-14}$ . Direct board-level measurements using a Nordic Power Profiler Kit II yielded approximately $2.46~\mu $ J per kernel inference and $13.79~\mu $ J per complete pipeline. The principal contribution lies in system-level integration, auditable experiment provenance, and physically verified edge deployment rather than in proposing a new forecasting algorithm.

Förderung

  • Institute of Electronics, Lodz University of Technology and Intelligent Hives Sp. z o.o

Zitieren

APA

Gorecki, S., Hausman, S., Korbel, P., & Brusey, J. (2026). A Large-Scale IoT Telemetry Framework for Edge-Aware Forecasting and Diagnostic State Tracking in Precision Apiculture. IEEE Access, 14, 138619-138641. https://doi.org/10.1109/access.2026.3731411

Vancouver

Gorecki S, Hausman S, Korbel P, Brusey J. A Large-Scale IoT Telemetry Framework for Edge-Aware Forecasting and Diagnostic State Tracking in Precision Apiculture. IEEE Access. 2026;14:138619-138641. doi:10.1109/access.2026.3731411

BibTeX
@article{Gorecki2026,
  author = {Gorecki, Sebastian and Hausman, Slawomir and Korbel, Piotr and Brusey, James},
  title = {A Large-Scale IoT Telemetry Framework for Edge-Aware Forecasting and Diagnostic State Tracking in Precision Apiculture},
  journal = {IEEE Access},
  year = {2026},
  volume = {14},
  pages = {138619--138641},
  doi = {10.1109/access.2026.3731411}
}
Kostenlos starten →