Scholarlink

Alpine livestock monitoring is critical for ecological preservation and agricultural efficiency. However, existing solutions struggle with energy constraints, limited network availability, and intermittent connectivity in remote environments. To address this, we propose an on-device federated learning framework tailored for PV-powered IoT sensors to optimize energy-communication tradeoffs. Our approach introduces staleness-aware aggregation and solar-aware training scheduling to address intermittent connectivity and PV variability in remote alpine environments. Deployed on a real-world testbed with collar sensors, the framework achieves 92% accuracy in time-series location prediction and 89% F1-score in anomaly detection while using 68% less energy than centralized baselines.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

On-Device Federated Learning for Remote Alpine Livestock Monitoring

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Sabtain Ahmad,
  • Thomas Schneidergruber,
  • Ivona Brandic,
  • Johannes Scholz

摘要

Alpine livestock monitoring is critical for ecological preservation and agricultural efficiency. However, existing solutions struggle with energy constraints, limited network availability, and intermittent connectivity in remote environments. To address this, we propose an on-device federated learning framework tailored for PV-powered IoT sensors to optimize energy-communication tradeoffs. Our approach introduces staleness-aware aggregation and solar-aware training scheduling to address intermittent connectivity and PV variability in remote alpine environments. Deployed on a real-world testbed with collar sensors, the framework achieves 92% accuracy in time-series location prediction and 89% F1-score in anomaly detection while using 68% less energy than centralized baselines.