Journal of Applied Science and Engineering

Published by Tamkang University Press

ESCI jase impact factor scopus logo open access rate of Scopus journal

Communication-Efficient Improved Federated Learning Framework with Local Privacy Perturbation for Distributed Sensitive Data Privacy Preservation

 2026

 Volume 35

Bi Zhang and Min-Po Jung

Department of Computer and Information Engineering, Graduate School Youngsan University, Busan, 612-022 South Korea

Received: July 03, 2026
Accepted: August 20, 2026
Publication Date: September 25, 2026

上傳圖片

Proposed Fed-LPCC framework

 Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Download Citation: BibTeX | http://dx.doi.org/10.6180/jase.202612_35.056

Download PDF

Federated Learning (FL) emerges as a promising distributed machine learning paradigm, which enables collaborative model training among multiple edge clients without uploading raw sensitive data to the central server. Nevertheless, conventional federated learning systems still suffer from two intractable bottlenecks: severe privacy leakage risks caused by plaintext gradient/model parameter transmission and excessive communication overhead induced by frequent complete parameter updates across heterogeneous edge devices. To tackle the above dual challenges, this paper proposes a communication-efficient improved federated learning framework with adaptive local privacy perturbation, named Fed-LPCC (Federated Learning with Local Privacy Perturbation and Compressed Communication). First, an adaptive mixed Gaussian-Laplace local differential privacy (LDP) perturbation mechanism is designed to inject differentiated noise into local gradient updates, which dynamically adjusts noise amplitude according to gradient sensitivity and residual privacy budget, effectively preventing sensitive data leakage from gradient inversion attacks while reducing unnecessary utility loss. Second, we propose a dual-layer gradient sparsification-quantization compression strategy to eliminate redundant parameters in local updates combined with adaptive client sampling, the framework significantly cuts down uplink and downlink communication traffic during global aggregation. Furthermore, we derive rigorous privacy accounting based on Rényi Differential Privacy (RDP) and complete convergence proof for the proposed Fed-LPCC framework. Extensive comparative experiments are conducted on three mainstream public datasets (MNIST, CIFAR-10, Adult) against state-of-the-art privacy-preserving federated learning methods. The experimental results demonstrate that Fed-LPCC achieves a maximum communication compression ratio of 94.7%, reduces privacy budget consumption by 18.3% compared with traditional fixed-noise LDP-FL methods, and maintains competitive model accuracy with a maximum accuracy drop of only 2.1%. The proposed framework can well balance privacy preservation, communication efficiency and model utility, which is suitable for distributed sensitive data training scenarios in intelligent edge computing, healthcare and financial risk prediction.

Keywords: Federated Learning, Local Differential Privacy, Gradient Compression, Communication Efficiency, Privacy Preservation, Distributed Sensitive Data

  1. [1] M. Ebrahimi, R. Sahay, S. Hosseinalipour, and B. Akram, (2026) “The transition from centralized machine learning to federated learning for mental health in education: A survey of current methods and future directions” IEEE Transactions on Learning Technologies 19: 341-359. DOI: https://doi.org/10.1109/TLT.2026.3685099.
  2. [2] A. Choksi and M. Shah, (2024) “Machine learning based centralized dynamic clustering algorithm for energy efficient routing in vehicular ad hoc networks” Transactions on Emerging Telecommunications Technologies 35(1): e4914. DOI: https://doi.org/10.1002/ett.4914.
  3. [3] S. Yin, H. Li, A. A. Laghari, L. Teng, T. R. Gadekallu, and A. Almadhor, (2024) “FLSN-MVO: Edge computing and privacy protection based on federated learning siamese network with multi-verse optimization algorithm for industry 5.0” IEEE Open Journal of the Communications Society 6: 3443-3458. DOI: https://doi.org/10.1109/OJCOMS.2024.3520562.
  4. [4] M. Gan, L. Li, S. Alam, L. Liu, L. Liu, M. Zhang, H. Zeng, and Z. Cao, (2026) “Geofl: A framework for efficient geo-distributed cross-device federated learning” IEEE Transactions on Networking: DOI: https://doi.org/10.1109/TON.2026.3668183.
  5. [5] M. Tausif, M. W. Iqbal, R. N. Bashir, B. AlGhofaily, A. Elyassih, and A. R. Khan, (2025) “Federated learning based reference evapotranspiration estimation for distributed crop fields” PloS one 20(2): e0314921. DOI: https://doi.org/10.1371/journal.pone.0314921.
  6. [6] S. Li, B. Chen, Y. Li, Z. Wang, Y. Xue, and C. Xu, (2026) “Privacy-Preserving Anomaly Detection in Cloud Services Using Hierarchical Federated Learning with Differential Privacy” In: 2026 9th International Symposium on Big Data and Applied Statistics (ISBDAS). IEEE. 2026, 668-672. DOI: https://doi.org/10.1109/ISBDAS69350.2026.11484374.
  7. [7] D. Domini, G. Aguzzi, N. Farabegoli, M. Viroli, and L. Esterle, (2024) “Proximity-based self-federated learning” In: 2024 IEEE International Conference on Autonomic Computing and Self-Organizing Systems (ACSOS). IEEE. 2024, 139-144. DOI: https://doi.org/10.1109/ACSOS61780.2024.00033.
  8. [8] S. Yin, H. Li, A. A. Laghari, T. R. Gadekallu, G. A. Sampedro, and A. Almadhor, (2024) “An anomaly detection model based on deep auto-encoder and capsule graph convolution via sparrow search algorithm in 6G Internet of Everything” IEEE Internet of Things Journal 11(18): 29402-29411. DOI: https://doi.org/10.1109/JIOT.2024.3353337.
  9. [9] N. Anithadevi, S. Deivarani, and A. Balakrishnan, (2026) “DEF-CRYPT-Q: a quantum-enhanced hybrid encryption framework for privacy-preserving distributed defense communications” Scientific Reports: DOI: https://doi.org/10.1038/s41598-026-53875-9.
  10. [10] M. Rahaman, V. Arya, S. M. Orozco, and P. Pappachan, (2024) “Secure multi-party computation (SMPC) protocols and privacy” In: Innovations in Modern Cryptography. IGI Global Scientific Publishing, 2024, 193-218. DOI: https://doi.org/10.4018/979-8-3693-5330-1.ch008.
  11. [11] S. Yin, H. Li, L. Teng, A. A. Laghari, and V. V. Estrela, (2024) “Attribute-based multiparty searchable encryption model for privacy protection of text data” Multimedia Tools and Applications 83(15): 45881-45902. DOI: https://doi.org/10.1007/s11042-023-16818-4.
  12. [12] L. Shen, Y. Sun, Z. Yu, L. Ding, X. Tian, and D. Tao, (2024) “On efficient training of large-scale deep learning models” ACM Computing Surveys 57(3): 1-36. DOI: https://doi.org/10.1145/3700439.
  13. [13] Y. Zhang, G. Chen, and S. Cui, (2025) “Efficient large-scale scene representation with a hybrid of high-resolution grid and plane features” Pattern Recognition 158: 111001. DOI: https://doi.org/10.1016/j.patcog.2024.111001.
  14. [14] M. V. Devarajan, A. R. G. Yallamelli, R. K. M. Kanta Yalla, V. Mamidala, T. Ganesan, and A. Sambas, (2026) “Attacks classification and data privacy protection in cloud-edge collaborative computing systems” International Journal of Parallel, Emergent and Distributed Systems 41(3): 419-438. DOI: https://doi.org/10.1080/17445760.2024.2417875.
  15. [15] Y. Zhao, J. Zhao, M. Yang, T. Wang, N. Wang, L. Lyu, D. Niyato, and K.-Y. Lam, (2020) “Local differential privacy-based federated learning for internet of things” IEEE Internet of Things Journal 8(11): 8836-8853. DOI: https://doi.org/10.1109/JIOT.2020.3037194.
  16. [16] B. Wang, Y. Chen, H. Jiang, and Z. Zhao, (2023) “Ppefl: Privacy-preserving edge federated learning with local differential privacy” IEEE Internet of Things Journal 10(17): 15488-15500. DOI: https://doi.org/10.1109/JIOT.2023.3264259.
  17. [17] R. Hu, Y. Guo, and Y. Gong, (2023) “Federated learning with sparsified model perturbation: Improving accuracy under client-level differential privacy” IEEE Transactions on Mobile Computing 23(8): 8242-8255. DOI: https://doi.org/10.1109/TMC.2023.3343288.
  18. [18] Z. Zhang and C. Wang, (2022) “MIPD: An adaptive gradient sparsification framework for distributed DNNs training” IEEE Transactions on Parallel and Distributed Systems 33(11): 3053-3066.
  19. [19] P. Mayekar and H. Tyagi, (2020) “Limits on gradient compression for stochastic optimization” In: 2020 IEEE International Symposium on Information Theory (ISIT). IEEE. 2020, 2658-2653. DOI: https://doi.org/10.1109/ISIT44484.2020.9174075.
  20. [20] M. Zhang, Z. Xie, and L. Yin, (2024) “Private and communication-efficient federated learning based on differentially private sketches” arXiv preprint arXiv:2410.05733. DOI: https://doi.org/10.48550/arXiv.2410.05733.
  21. [21] Y. Xia, T. Jahani-Nezhad, and R. Bitar, (2025) “Fed-DPROC: Communication-Efficient Differentially Private and Robust Federated Learning” In: 2025 3rd International Conference on Federated Learning Technologies and Applications (FLTA). IEEE. 2025, 142-149. DOI: https://doi.org/10.1109/FLTA67013.2025.11336422.
  22. [22] U. Islam, H. Ullah, N. Khan, I. Ahmad, and K. Saleem, (2025) “Adaptive federated learning framework for privacy-preserving consumer-centric IoMT: a novel secure data collaboration model” IEEE Transactions on Consumer Electronics. DOI: https://doi.org/10.1109/TCE.2025.3606642.
  23. [23] L. Chen, X. Ding, M. Li, and H. Jin, (2023) “Differentially private federated learning with importance client sampling” IEEE Transactions on Consumer Electronics 70(1): 3635-3649. DOI: https://doi.org/10.1109/TCE.2023.3338464.
  24. [24] C. Ma, J. Li, M. Ding, B. Liu, K. Wei, J. Weng, and H. V. Poor, (2023) “RDP-GAN: A Rényi-differential privacy based generative adversarial network” IEEE Transactions on Dependable and Secure Computing 20(6): 4838-4852. DOI: https://doi.org/10.1109/TDSC.2022.3233580.