MARLTC: A Multi-Agent Reinforcement Learning-Based Interference-Aware Transmission Control for LoRaWAN IoT Devices
IEEE Internet of Things Journal · 2026-08 · 13 (15) · 35121-35138
DOI: 10.1109/jiot.2026.3696353 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Internet of Things Journal, 2026-08. Verfasst von Juan Aznar-Poveda und 8 weiteren Autor:innen. Die Arbeit nennt 64 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Partnership for Research and Innovation in the Mediterranean Area, Agencia Estatal de Investigaci?n.
Abstract (Originalsprache)
Long-range wide-area network (LoRaWAN) has become a foundational technology in the Internet of Things (IoT) landscape due to its long-range (LoRa) communication and energy efficiency (EE). However, its default adaptive data rate (ADR) mechanism struggles to adapt to dynamic environments and dense network deployments, where co-channel interference and the coupling between transmission parameters limit its ability to ensure reliable and energy-efficient communication. To address these challenges, this article proposes MARLTC, an ADR mechanism based on multi-agent reinforcement learning (MARL) that jointly optimizes spreading factor (SF) and transmission power using realistic observable metrics, including recent transmission history, observed signal-to-noise ratios (SNRs), and distribution of SFs in the network. The transmission configuration problem is modeled as a cooperative Markov game and solved using the centralized training with decentralized execution (CTDE) paradigm, where pretrained policies are deployed at end devices to infer suitable transmission parameters with reduced convergence time. The results using a realistic LoRaWAN simulator show that MARLTC achieves up to 67.9% faster convergence, 62.1% higher EE, and a 5.0% better packet delivery ratio (PDR) compared to state-of-the-art approaches, highlighting its scalability and responsiveness in dense deployments. The practical feasibility of MARLTC is further validated using physical LoRaWAN hardware, proving that the resulting policies meet the memory and timing constraints of resource-constrained IoT devices.
Förderung
- Partnership for Research and Innovation in the Mediterranean Area
- Agencia Estatal de Investigaci?n
Zitieren
APA
Aznar-Poveda, J., Acosta-Garcia, L., Margreiter, F., Etheredge, M., Younesi, A., Pedratscher, S., Garcia-Haro, J., Fahringer, T., & Garcia-Sanchez, A. (2026). MARLTC: A Multi-Agent Reinforcement Learning-Based Interference-Aware Transmission Control for LoRaWAN IoT Devices. IEEE Internet of Things Journal, 13(15), 35121-35138. https://doi.org/10.1109/jiot.2026.3696353
Vancouver
Aznar-Poveda J, Acosta-Garcia L, Margreiter F, Etheredge M, Younesi A, Pedratscher S, et al. MARLTC: A Multi-Agent Reinforcement Learning-Based Interference-Aware Transmission Control for LoRaWAN IoT Devices. IEEE Internet of Things Journal. 2026;13(15):35121-35138. doi:10.1109/jiot.2026.3696353
BibTeX
@article{AznarPoveda2026,
author = {Aznar-Poveda, Juan and Acosta-Garcia, Laura and Margreiter, Fabian and Etheredge, Marlon and Younesi, Abolfazl and Pedratscher, Stefan and Garcia-Haro, Joan and Fahringer, Thomas and Garcia-Sanchez, Antonio-Javier},
title = {MARLTC: A Multi-Agent Reinforcement Learning-Based Interference-Aware Transmission Control for LoRaWAN IoT Devices},
journal = {IEEE Internet of Things Journal},
year = {2026},
volume = {13},
number = {15},
pages = {35121--35138},
doi = {10.1109/jiot.2026.3696353}
}
EUVIMED