FADR-MLP Lite: Compact Neural Adaptive Data Rate for Mobile LoRaWAN

• Fernando J. V. Santos UFPI / IFMA
• Thiago A. R. da Silva UFPI
• Geraldo A. S. Neto UFPI
• Isaac S. S. Ramos UFPI
• Alex F. dos Santos UFRB / UEFS
• Anderson L. Sanches UEFS
• Jose Valdemir dos Reis Jr. UFPI

DOI: https://doi.org/10.5753/sbrc.2026.19367

Resumo

The computational complexity of Adaptive Data Rate (ADR) algorithms in mobile LoRaWAN challenges their execution on resource-constrained devices. This work validates FADR-MLP Lite, a surrogate modeling strategy that compresses a fuzzy controller into a compact neural network. Through cosimulation with a network simulator, technical feasibility was evidenced: a 92% reduction in parameters, a 1.45 KB memory footprint, and only 672 floating-point operations per inference. Under mobility, neural generalization acted as a regularizer, outperforming the expert system that trained the network with a 19% energy gain while sustaining a packet delivery ratio above 60% in high-density scenarios, validating its robustness for low-cost Internet of Things (IoT) devices.

Referências

Ali Lodhi, M., Sun, X., Mahmood, K., Lodhi, A., Park, Y., and Hussain, M. (2025). AIEnhanced Resource Allocation for LPWAN-Based LoRaWAN:A Hybrid TinyML and Deep Learning Approach. IEEE Internet of Things Journal, 12(14):28950–28963.

Bhat, S. A., Huang, N.-F., Hussain, I., and Sajjad, U. (2023). Correlating the Ambient Conditions and Performance Indicators of the LoRaWAN via Surrogate Gaussian Process-Based Bidirectional LSTM Stacked Autoencoder. IEEE Transactions on Network and Service Management, 20(3):3413–3427.

Cheng, Y., Wang, D., Zhou, P., and Zhang, T. (2020). A Survey of Model Compression and Acceleration for Deep Neural Networks. IEEE Signal Processing Magazine, 37(5):160–174.

Farhad, A. and Pyun, J.-Y. (2023). AI-ERA: Artificial Intelligence-Empowered Resource Allocation for LoRa-Enabled IoT Applications. IEEE Transactions on Industrial Informatics, 19(12):11640–11653.

González-Palacio, M., Tobón-Vallejo, D., Sepúlveda-Cano, L. M., Luna-Delrisco, M., Röhrig, C., and Le, L. B. (2024). Machine-Learning-Assisted Transmission Power Control for LoRaWAN Considering Environments With High Signal-to-Noise Variation. IEEE Access, 12:54466–54487.

Jang, J.-S. R. (1993). ANFIS: Adaptive-Network-Based Fuzzy Inference System. IEEE Transactions on Systems, Man, and Cybernetics, 23(3):665–685.

Kufakunesu, R., Hancke, G., and Abu-Mahfouz, A. (2022). A Fuzzy-Logic Based Adaptive Data Rate Scheme for Energy-Efficient LoRaWAN Communication. Journal of Sensor and Actuator Networks, 11:65.

Lodhi, M. A., Wang, L., Farhad, A., Qureshi, K. I., Chen, J., Mahmood, K., and Das, A. K. (2025a). A Contextual Aware Enhanced LoRaWAN Adaptive Data Rate for mobile IoT applications. Computer Communications, 232:108042.

Lodhi, M. A., Wang, L., Mahmood, K., Farhad, A., Chen, J., and Kumari, S. (2025b). Enhanced adaptive data rate strategies for energy-efficient Internet of Things communication in LoRaWAN. International Journal of Communication Systems, 38(2):e5966.

Prakash, A., Choudhury, N., Hazarika, A., and Gorrela, A. (2025). Effective Feature Selection for Predicting Spreading Factor with ML in Large LoRaWAN-Based Mobile IoT Networks. In 2025 National Conference on Communications (NCC), pages 1–6. IEEE.

Santos, F. J. V., da Silva, T. A. R., Sarmento Neto, G. A., Mendes, L. H. O., Abreu, P. F. F., Sanches, A. L., and dos Reis Jr., J. V. (2025). FADR-MLP: A Lightweight Neural Network-Based Surrogate Model for Adaptive Data Rate Control in LoRaWAN. In Extended Proceedings of the XV Symposium on Computing Systems Engineering (SBESC), pages 41–44, Campinas, SP, Brazil. SBC.

Santos, F. J. V., Neto, G. A. S., da Silva, T. A. R., de Abreu, P. F. F., de O. Mendes, L. H., and dos Reis Junior, J. V. (2026). ADR Based on Fuzzy Logic with Dynamic Number of Uplink Transmissions Adjustment for LoRaWAN Networks in Mobility Scenarios. In de Freitas, R. and Furtado, D., editors, Intelligent Systems. BRACIS 2025, pages 74–88, Cham. Springer Nature Switzerland.

Sarmento Neto, G. A., da Silva, T. A. R., Abreu, P. F. F., Veloso, A. F. D. S., Mendes, L. H. D. O., and dos Reis Junior, J. V. (2024). A Fuzzy Logic Approach to Adaptive Data Rate for Mobile LoRaWAN Environments. In 2024 XIV Brazilian Symposium on Computing Systems Engineering (SBESC), pages 1–6.

Wang, H., Pei, P., Pan, R., Wu, K., Zhang, Y., Xiao, J., and Yang, J. (2022). A Collision Reduction Adaptive Data Rate Algorithm based on the FSVM for a Low-Cost LoRa Gateway. Mathematics, 10(21):3920.

Yang, H., Wu, X., Ji, H., Huang, Z., and Fang, J. (2025). A Topology-Aware GNN Learning Approach for Energy Optimization in Multihop LoRa Networks. IEEE Internet of Things Journal, 12(22):46596–46610.