AN ADAPTIVE FEDERATED DEEP LEARNING FRAMEWORK FOR SECURE AND PRIVACY PRESERVING PRECISION AGRICULTURE IN DISTRIBUTED IOT NETWORKS

Umair Aslam; Muhammad Aqib; Iqra Arshad; Basit Bashir; Muhammad Daniyal; Muhammad Yousif

doi:10.71146/kjmr964

Authors

Umair Aslam Department of Management Sciences, NUML, Lahore Campus, Pakistan Author
Muhammad Aqib Department of Computer Science, Lahore Garrison University, Lahore, Pakistan Author
Iqra Arshad Department of Computing, NFC Institute of Engineering and Technology, Multan, Pakistan Author
Basit Bashir Department of Computer Science, University of Engineering and Technology, Lahore, Pakistan Author
Muhammad Daniyal Department of Computer Science, Superior College Pak Arab campus-Lahore, Pakistan Author
Muhammad Yousif Department of Computer Science, NUML, Lahore Campus, Pakistan Author

DOI:

https://doi.org/10.71146/kjmr964

Keywords:

Federated Learning, Precision Agriculture, Privacy-Preserving Machine Learning, Distributed IoT Networks, Adaptive Aggregation, Deep Learning, Differential Privacy, Secure Aggregation, Edge Computing, Smart Farming, Non-IID Data, Crop Health Monitoring

Abstract

The Internet of Things (IoT) has revolutionized agriculture data gathering by connecting equipment with increased sophistication and scale, contributing to the smart monitoring and data-driven management in the field. The Internet of Things (IoT) has transformed the way data is collected in agriculture by making the process more intelligent and comprehensive, which is vital for smart monitoring and data-driven management in agriculture. But the spread of IoT applications in farms poses significant challenges in data privacy and security, scalability and efficient collaborative learning. In this paper, an Adaptive Federated Deep Learning (AFDL) framework for secure and privacy-preserving precision agriculture in a distributed IoT network is proposed. The proposed framework utilizes federated deep learning, a technique that allows for distributed model training across various agricultural IoT nodes, thereby preserving the privacy of the sensitive data on the farms. The adaptive aggregation strategy adjusts the contribution of participating nodes dynamically according to data quality, the performance of models, and network conditions, which further improves the efficiency of learning and convergence of the model. In addition, secure communication and privacy preserving schemes are built in to safeguard against unauthorized access to and inference attacks on model updates. The effectiveness of the framework has been assessed with the Smart Farm Sensor Dataset, with deep learning models trained independently at the various nodes of the farming site and then collectively aggregated to form a powerful global model. Experimental results show that the proposed framework can achieve a classification accuracy of 97.4% and enhance the efficiency, scalability, and robustness of communication compared to traditional federated learning methods. Moreover, the adaptive aggregation mechanism decreases communication overhead by 65% and privacy leakage by 83%, which makes it applicable to resource limited IoT environments. In conclusion, the proposed Adaptive Federated Deep Learning framework offers a secure, scalable, and privacy-preserving approach for enabling intelligent farming practices that are decentralized and sustainable in the future of precision agriculture, leveraging collaborative learning within decentralized and distributed IoT networks.

Downloads

Download data is not yet available.

References

[1] P. Shambhavi, K. A. Jyotsna, V. Jyothi, S. Kanakaprabha, M. Anusha, and G. C. Babu, "Privacy-Preserving Edge Intelligence for Agriculture: A 6G-Enabled Federated Learning Architecture," in Proc. ICSCN 2025, pp. 318–324.

[2] M. Kavitha, "Federated Learning Framework for Privacy-Preserving Data Analytics in Smart Agriculture for Rural Environments," National Journal of Smart Agriculture and Rural Innovation, vol. 2, no. 1, pp. 9–16, 2024.

[3] S. Behera, N. Padhy, R. Panigrahi, and S. K. Kuanar, "Crop Disease Prediction Using Deep Learning in a Federated Learning Environment: Ensuring Data Privacy and Agricultural Sustainability," Procedia Computer Science, vol. 254, pp. 137–146, 2025.

[4] P. Kumar, G. P. Gupta, and R. Tripathi, "PEFL: Deep Privacy-Encoding-Based Federated Learning Framework for Smart Agriculture," IEEE Micro, vol. 42, no. 1, pp. 33–40, 2022.

[5] S. M. Shawon, I. A. Adan, M. Moniruzzaman, O. H. Sinan, M. J. Abed, M. Al Qwaid, M. T. Sarker, and H. B. A. Karim, "Federated Learning Approaches in Precision Agriculture: A Systematic Review of Current Trends, Challenges, and Opportunities," IEEE Access, vol. 14, pp. 30513–30538, 2026.

[6] M. Aldossary, J. Almutairi, and I. Alzamil, "Federated LeViT-ResUNet for Scalable and Privacy-Preserving Agricultural Monitoring Using Drone and Internet of Things Data," Agronomy, vol. 15, no. 4, p. 928, 2025.

[7] Janga, R., & Dave, R. (2025). Enhancing smart farming through federated learning: A secure, scalable, and efficient approach for AI-driven agriculture. arXiv preprint arXiv:2509.12363.

[8] Chitradevi, B., Kumar, S. S., Balaji, N. S., Suganya, J., Bharathi, B. M. R., Saleel, C. A., ... & Xu, Y. (2026). Energy-efficient and secure federated learning framework for wireless sensor networks in smart agriculture applications. Scientific Reports.

[9] KAYA, Ş. M. (2026). FEDERATED LEARNING ARCHITECTURES FOR PRIVACY-PRESERVING AI IN IOT-BASED SMART AGRICULTURE. Advanced AI Architectures for Engineering Innovation, 287.

[10] Manoj, T., Makkithaya, K., & Narendra, V. G. (2025). A blockchain-assisted trusted federated learning for smart agriculture. SN Computer Science, 6(3), 221.

[11] Singh, B., Kotiyal, A., Kumar, P., KC, U., Doohan, N. V., & Kumar, R. R. (2026, February). Federated Learning-Driven Crop Disease Diagnosis Framework to Ensure Privacy-Preserving Predictive Modeling Across Distributed Agricultural Zones. In 2026 2nd International Conference on Big Data & Machine Learning (ICBDML) (pp. 1-6). IEEE.

[12] Tripathi, M. K., Reddy, P. K., Nikitha, V., Reddy, N., Gourisetty, A., & Misal, K. (2026). Federated Learning for Decentralized Smart Farm Network Applications: Enhancing Crop Classification Performance. Federated Learning for Smart Agriculture and Food Quality Enhancement, 193-215.

[13] Griffin, S. (2026). Federated Learning Frameworks for Privacy-Preserving Data Sharing in IoT-Enabled Precision Agriculture.

[14] Tripathi, M. K., Reddy, P. K., Nikitha, V., Reddy, N., Gourisetty, A., & Misal, K. (2026). Federated Learning for Decentralized Smart Farm Network Applications: Enhancing Crop Classification Performance. Federated Learning for Smart Agriculture and Food Quality Enhancement, 193-215.

[15] Soni, J. A., Pandya, N. R., Patel, R. B., & Raval, J. S. (2025). Privacy-preserving AI in agriculture: a review of federated learning approaches. Sankalchand Patel University Journal of Science Technology and Management Research, 1.

[16] Dembani, R., Karvelas, I., Akbar, N. A., Rizou, S., Tegolo, D., & Fountas, S. (2025). Agricultural data privacy and federated learning: A review of challenges and opportunities. Computers and Electronics in Agriculture, 232, 110048.

[17] Kondaveeti, H. K., Sai, G. B., Athar, S. A., Vatsavayi, V. K., Mitra, A., & Ananthachari, P. (2024, April). Federated learning for smart agriculture: Challenges and opportunities. In 2024 Third International Conference on Distributed Computing and Electrical Circuits and Electronics (ICDCECE) (pp. 1-7). IEEE.

[18] Behera, S. (2026). Deep Learning based Prediction of Crop Disease in Federated Learning Environment. Journal of Intelligent Decision Making and Information Science, 3(9s), 1354-1376.

[19] Behera, S., Padhy, N., Panigrahi, R., Mahapatro, P. K., & Arangi, D. Secure and Collaborative Crop Disease Prediction with Federated Deep Learning: Safeguarding Data Sovereignty in Agriculture. In Artificial Intelligence in Context of Digital Sovereignty (pp. 111-130). Routledge.

[20] Maruthupermal, S. (2026, March). A Novel Federated Learning using Deformable Convolutional GRU Framework for Smart Irrigation in Next-Gen Agriculture with Encryption. In 2026 8th International Conference on Intelligent Sustainable Systems (ICISS) (pp. 294-301). IEEE.

[21] Rajbongshi, S. K., Ghosh, R., Sarmah, K., & Sarmah, S. (2025). Federated Learning: A Game-Changer in Agricultural Decision-Making and Precision Farming. In Artificial Intelligence and Computer Vision for Ecological Informatics (pp. 231-245). CRC Press.