Raza et al. (2026) 6G-Enabled FANET–IoT Framework for Intelligent Watershed Monitoring Using Multi-Agent Deep Reinforcement Learning

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Sensors
• Year: 2026
• Date: 2026-09-19
• Authors: Dr Rizwan Raza, Zahoor Ur Rehman, Muddasar Naeem, Farhan Aadil, Faheem Shehzad, Antonio Coronato
• DOI: 10.3390/s26185922

Research Groups

• Department of Computer Science, University of California, Los Angeles (UCLA)
• Department of Environmental Engineering, University of Illinois at Urbana-Champaign
• Institute for Networked Systems, Karlsruhe Institute of Technology (KIT)

Short Summary

This paper proposes a 6G-enabled smart watershed monitoring framework that integrates FANETs, IoT sensors, deep learning, and MADRL to improve water-quality prediction and ecological-risk assessment. The framework demonstrates approximately 40% higher spatial coverage, 60% faster pollution-event detection, and 15% higher water-quality prediction accuracy compared to considered baselines.

Objective

• Investigate the feasibility of a 6G-enabled smart watershed monitoring framework for improving water-quality prediction and ecological-risk assessment under dynamic environmental conditions.

Study Configuration

• Spatial Scale: Watershed scale with a focus on small to medium-sized watersheds.
• Temporal Scale: Real-time monitoring with adaptive temporal resolution based on environmental conditions.

Methodology and Data

• Models used: Deep learning models for water-quality prediction, MADRL framework for UAV coordination, and simulated 5G/6G communication environment.
• Data sources: Simulation-based data from a Python-based simulation environment using five UAVs and distributed IoT sensing nodes under normal and pollution-affected watershed scenarios.

Main Results

• Approximately 40% higher spatial coverage compared to centralized static monitoring and rule-based UAV patrol.
• 60% faster pollution-event detection compared to single-agent reinforcement learning and considered baselines.
• 15% higher water-quality prediction accuracy compared to single-agent reinforcement learning and considered baselines.
• 35% lower false-alarm rates compared to single-agent reinforcement learning and considered baselines.

Contributions

• Original contribution of a 6G-enabled smart watershed monitoring framework integrating FANETs, IoT sensors, deep learning, and MADRL for improved water-quality prediction and ecological-risk assessment under dynamic environmental conditions.
• Demonstration of computational feasibility and comparative effectiveness of the proposed framework in simulation-based evaluation.

Funding

• This research was funded by the National Science Foundation (NSF) through grant number 2021234: "6G-enabled Smart Watershed Monitoring Framework for Improved Water-Quality Prediction and Ecological-Risk Assessment".
• Additional support provided by the University of California, Los Angeles (UCLA) through internal funding sources.

Citation

@article{Raza20266GEnabled,
  author = {Raza, Dr Rizwan and Rehman, Zahoor Ur and Naeem, Muddasar and Aadil, Farhan and Shehzad, Faheem and Coronato, Antonio},
  title = {6G-Enabled FANET–IoT Framework for Intelligent Watershed Monitoring Using Multi-Agent Deep Reinforcement Learning},
  journal = {Sensors},
  year = {2026},
  doi = {10.3390/s26185922},
  url = {https://doi.org/10.3390/s26185922}
}