Advanced Industrial SCIence (산업과 과학)

Convergence Society for SMB (중소기업융합학회)
권호 표지
DOI QR코드

DOI QR Code

A Protection Model for Smart Greenhouse Data with Federated Learning and Differential Privacy

연방 학습과 차등 프라이버시를 활용한 스마트 온실 데이터 보호 모델

  • Mang Lu (Dept. of Computer and Media Engineering, Tongmyong University) ;
  • Sung-Hwa Han (Dept. of Information System and Security, Tongmyong University) ;
  • Seung-Soo Shin (Dept. of Computer Engineering, Tongmyong University)
  • 루망 (동명대학교 컴퓨터미디어공학과) ;
  • 한성화 (동명대학교 Dept. of Information System and Security) ;
  • 신승수 (동명대학교 컴퓨터공학과)
  • Received : 2025.12.24
  • Accepted : 2026.03.20
  • Published : 2026.03.30
Download PDF
⟨ Previous Next ⟩

Abstract

Concerns with data privacy in dispersed agricultural IoT applications are raised by the massive amounts of sensitive environmental and crop-related data generated by smart greenhouse systems. In order to improve data protection in greenhouse settings, this study suggests a privacy-enhanced federated learning architecture that incorporates differential privacy. In order to achieve a balanced trade-off between accuracy, privacy budget, and system efficiency, the suggested hierarchical FL+DP strategy significantly stabilizes model updates and lowers communication overhead. The suggested model shows a favorable privacy-accuracy trade-off when compared to ordinary federated learning, achieving up to 88.7% prediction accuracy under moderate noise circumstances (𝜎 = 0.3-0.5) while offering explicit differential privacy guarantees. These findings show that the FLDP framework is a good fit for privacy-sensitive smart greenhouse applications that need high data secrecy and accurate prediction.

분산형 농업 IoT 애플리케이션에서 데이터 프라이버시 문제는 스마트 온실 시스템이 생성하는 방대한 양의 민감한 환경 및 작물 관련 데이터로 인해 제기된다. 본 연구는 온실 환경에서의 데이터 보호를 개선하기 위해 차등 프라이버시를 통합한 프라이버시 강화 FL 아키텍처를 제안한다. 이 접근법은 특히 라즈베리 파이와 같은 제한된 자원을 가진 에지 디바이스를 위해 설계했다. 실제 온실 환경 데이터셋을 대상으로 한 실험에 따르면, 중앙 집중식 훈련은 최대 정확도(92.4%)를 달성하지만 프라이버시 보호 기능이 전혀 제공되지 않는다. 정확도, 프라이버시 예산, 시스템 효율성 간의 균형 잡힌 절충점을 달성하기 위해 제안된 계층적 FL+DP 전략은 모델 업데이트를 크게 안정화하고 통신 오버헤드를 낮춘다. 이러한 결과는 높은 데이터 기밀성과 정확한 예측이 필요한 프라이버시 민감형 스마트 온실 애플리케이션에 FLDP 프레임워크가 적합함을 보여주었다.

Keywords

References

  1. Quy, V. K., Hau,N. V., Anh, D. V., Quy, N. M., Ban, N. T., Lanza, S., & Muzirafuti,. A. (2022). IoT-enabled smart agriculture: architecture, applications, and challenges. Applied Sciences. 12. DOI : 10.3390/app12073396.
  2. Li, L., Fan, Y., Tse, M., &Lin, K. Y. (2020). A review of applications in federated learning. Computers &Industrial Engineering, 149 , 106854. DOI : 10.1016/j.cie.2020.106854.
  3. Yang, M., Guo, T., Zhu, T., Tjuawinata, I., Zhao, J., &Lam, K. Y. (2024). Local differential privacy and its applications: A comprehensive survey. Computer Standards &Interfaces , 89 , 103827. DOI : 10.1016/j.csi.2023.103827.
  4. Ficco, M., Guerriero, A., Milite, E., Palmieri, F., Pietrantuono, R., &Russo, S. (2024). Federated learning for IoT devices: Enhancing TinyML with on-board training. Information Fusion, 104 , 102189. DOI : 10.1016/j.inffus.2023.102189.
  5. Wen, J., Zhang, Z., Lan, Y., Cui, Z., Cai, J., & Zhang, W. (2023). A survey on federated learning: challenges and applications. International journal of machine learning and cybernetics , 14 (2), 513-535. DOI : 10.1007/s13042-022-01647-y.
  6. Li, H., Li, C., Wang, J., Yang, A., Ma, Z., Zhang, Z., &Hua, D. (2023). Review on security of federated learning and its application in healthcare. Future Generation Computer Systems , 144 , 271-290. DOI : 10.1016/j.future.2023.02.021.
  7. Byrd, D., &Polychroniadou, A. (2020, October). Differentially private secure multi-party computation for federated learning in financial applications. In Proceedings of the first ACM international conference on AI in finance (pp. 1-9). DOI : 10.1145/3383455.3422562.
  8. Abadi, M., Chu, A., Goodfellow, I., McMahan, H. B., Mironov, I., Talwar, K., &Zhang, L. (2016, October). Deep learning with differential privacy. In Proceedings of the 2016 ACM SIGSAC conference on computer and communications security (pp. 308-318). DOI : 10.1145/2976749.2978318.
  9. Hassan, M. U., Rehmani, M. H., &Chen, J. (2019). Differential privacy techniques for cyber physical systems: A survey. IEEE Communications Surveys &Tutorials , 22(1), 746-789. DOI : 10.1109/COMST.2019.2944748.
  10. Wu, G., Hu, Q., Liu, X., Qin, H., &Gao, B. (2024). Differential Privacy-Enhanced Blockchain-Based Quality Control Model for Rice. Smart Agriculture , 6(4), 149-159. DOI : 10.12133/j.smartag.SA202311027.
  11. Soheli, S. J., Jahan, N., Hossain, M. B., Adhikary, A., Khan, A. R., &Wahiduzzaman, M. (2022). Smart greenhouse monitoring system using internet of things and artificial intelligence. Wireless Personal Communications, 124(4), 3603-3634. DOI : 10.1007/s11277-022-09528-x.
  12. Patil, A. S., Tama, B. A., Park, Y., & Rhee, K. H. (2018). A Framework for Blockchain Based Secure Smart Green House Farming. Advances in Computer Science and Ubiquitous Computing, 474, 1162-1167. DOI : 10.1007/978-981-10-7605-3_185.
  13. 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. DOI : 10.1109/ICDCECE60827.2024.10548604.
  14. Geyer, R. C., Klein, T., & Nabi, M. (2017). Differentially private federated learning: A client level perspective. arXiv preprint arXiv:1712.07557. DOI : 10.48550/arXiv.1712.07557.
  15. Ponomareva, N., et al. (2023). How to dp-fy ml: A practical guide to machine learning with differential privacy. Journal of Artificial Intelligence Research, 77, 1113-1201. DOI : 10.1613/jair.1.14649.
  16. Adnan, M., Syed, M. H., Anjum, A., &Rehman, S. (2025). A Framework for privacy-preserving in IoV using Federated Learning with Differential Privacy. IEEE Access, 13, 13507-13521. DOI : 10.1109/ACCESS.2025.3526934.
  17. 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. DOI : 10.1016/j.compag.2025.110048.