Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

SusFL: Energy-Aware Federated Learning-based Monitoring for Sustainable Smart Farms

Dian Chen, Paul Yang, Ing-Ray Chen, Dong Sam Ha, Jin-Hee Cho

TL;DR

SusFL tackles health monitoring in sustainable smart farms under energy constraints by introducing a hierarchical FL framework with mechanism-design-based client selection. By estimating client utility and data quality, SusFL selects energy-feasible participants and weights updates via data-quality-aware FedAvg, achieving energy savings, improved reliability (MTBF), and robust performance under adversarial attacks. The approach is validated on real mastitis datasets and IoAHT-derived data, showing advantages over multiple FL baselines in accuracy, energy consumption, and social welfare. This work advances practical, secure, and energy-efficient AI-driven farming, enabling scalable deployment in renewable-energy-powered agricultural networks.

Abstract

We propose a novel energy-aware federated learning (FL)-based system, namely SusFL, for sustainable smart farming to address the challenge of inconsistent health monitoring due to fluctuating energy levels of solar sensors. This system equips animals, such as cattle, with solar sensors with computational capabilities, including Raspberry Pis, to train a local deep-learning model on health data. These sensors periodically update Long Range (LoRa) gateways, forming a wireless sensor network (WSN) to detect diseases like mastitis. Our proposed SusFL system incorporates mechanism design, a game theory concept, for intelligent client selection to optimize monitoring quality while minimizing energy use. This strategy ensures the system's sustainability and resilience against adversarial attacks, including data poisoning and privacy threats, that could disrupt FL operations. Through extensive comparative analysis using real-time datasets, we demonstrate that our FL-based monitoring system significantly outperforms existing methods in prediction accuracy, operational efficiency, system reliability (i.e., mean time between failures or MTBF), and social welfare maximization by the mechanism designer. Our findings validate the superiority of our system for effective and sustainable animal health monitoring in smart farms. The experimental results show that SusFL significantly improves system performance, including a $10\%$ reduction in energy consumption, a $15\%$ increase in social welfare, and a $34\%$ rise in Mean Time Between Failures (MTBF), alongside a marginal increase in the global model's prediction accuracy.

SusFL: Energy-Aware Federated Learning-based Monitoring for Sustainable Smart Farms

TL;DR

SusFL tackles health monitoring in sustainable smart farms under energy constraints by introducing a hierarchical FL framework with mechanism-design-based client selection. By estimating client utility and data quality, SusFL selects energy-feasible participants and weights updates via data-quality-aware FedAvg, achieving energy savings, improved reliability (MTBF), and robust performance under adversarial attacks. The approach is validated on real mastitis datasets and IoAHT-derived data, showing advantages over multiple FL baselines in accuracy, energy consumption, and social welfare. This work advances practical, secure, and energy-efficient AI-driven farming, enabling scalable deployment in renewable-energy-powered agricultural networks.

Abstract

We propose a novel energy-aware federated learning (FL)-based system, namely SusFL, for sustainable smart farming to address the challenge of inconsistent health monitoring due to fluctuating energy levels of solar sensors. This system equips animals, such as cattle, with solar sensors with computational capabilities, including Raspberry Pis, to train a local deep-learning model on health data. These sensors periodically update Long Range (LoRa) gateways, forming a wireless sensor network (WSN) to detect diseases like mastitis. Our proposed SusFL system incorporates mechanism design, a game theory concept, for intelligent client selection to optimize monitoring quality while minimizing energy use. This strategy ensures the system's sustainability and resilience against adversarial attacks, including data poisoning and privacy threats, that could disrupt FL operations. Through extensive comparative analysis using real-time datasets, we demonstrate that our FL-based monitoring system significantly outperforms existing methods in prediction accuracy, operational efficiency, system reliability (i.e., mean time between failures or MTBF), and social welfare maximization by the mechanism designer. Our findings validate the superiority of our system for effective and sustainable animal health monitoring in smart farms. The experimental results show that SusFL significantly improves system performance, including a reduction in energy consumption, a increase in social welfare, and a rise in Mean Time Between Failures (MTBF), alongside a marginal increase in the global model's prediction accuracy.
Paper Structure (28 sections, 7 equations, 7 figures, 2 tables)

This paper contains 28 sections, 7 equations, 7 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Motivation & Research Goal
  3. Key Contributions
  4. Background & Related Work
  5. Federated Learning
  6. Energy-Efficient FL for Smart IoT Systems
  7. FL-based Monitoring Systems
  8. FL-based Smart Farms
  9. Problem Statement
  10. System model
  11. Network Model
  12. Node Model
  13. Threat Model
  14. Proposed Approach: SusFL
  15. Clients' Utility Estimation
  16. ...and 13 more sections

Figures (7)

  • Figure 1: Key steps in FL training process
  • Figure 2: Hierarchical FL-based network architecture designed for a wireless solar sensor-based smart farm
  • Figure 3: Key steps for an edge server (i.e., mechanism designer) to aggregate local model updates from selected clients
  • Figure 4: Comparative performance analysis during training time
  • Figure 5: Effect of Varying Attack Severity ($P_A$)
  • ...and 2 more figures