Background: Federated learning enables multiple participants to train ML models collaboratively without exposing or sharing raw data, making FL attractive approach for privacy-sensible applications. To reduce residual leakage risks in shared model updates, studies have proposed several data-protection mechanisms, including differential privacy, homomorphic encryption, secure multi-party computation, secure aggregation protocols, federated averaging with secure aggregation, model distillation, regularization, and client-side anonymization. While these mechanisms are regularly evaluated for the strength of guarantees they provide, their performance implications: computational overhead, communication cost, model accuracy, and scalability are comparatively under-examined as a harmonized basis for practical selection. Method: The study conducted a systematic comparative analysis of eight data-protection mechanisms in federated learning, evaluating them using performance metrics. Drawing on a structured analysis of the most recent innovations, the study developed a comparison framework, applied it consistently across mechanisms, and presented the resulting trade-offs in tabular and narrative form. Results: The study found that lightweight mechanisms: secure aggregation, model distillation, and regularization techniques offer the most favourable performance profiles for resource-constrained, large-scale deployments, while the cryptographic approach: homomorphic encryption and secure multi-party computation impose substantial computational and communication burdens that limit technique scalability. Conclusion: The study concludes that, with practical guidance for mechanism selection based on deployment context, it outlines directions for future performance-oriented research in federated learning.
References
[1]
Banse, A., Kreischer, J. and Jürgens, X.O.I. (2024) Federated Learning with Differential Privacy. arXiv: 2402.02230. http://arxiv.org/abs/2402.02230
[2]
Fu, J., Hong, Y., Ling, X., Wang, L., Ran, X., Sun, Z., Wang, W.H., Chen, Z. and Cao, Y. (2024) Differentially Private Federated Learning: A Systematic Review. arXiv: 2405.08299.
[3]
Ibrahim Khalaf, O., Ashokkumar, S.R., Algburi, S., Anupallavi, S., Selvaraj, D., Sharif, M.S., et al. (2024) Federated Learning with Hybrid Differential Privacy for Secure and Reliable Cross-IoT Platform Knowledge Sharing. SecurityandPrivacy, 7, e374. https://doi.org/10.1002/spy2.374
[4]
Ren, X., Yang, S., Zhao, C., McCann, J. and Xu, Z. (2024) Belt and Braces: When Federated Learning Meets Differential Privacy. CommunicationsoftheACM, 67, 66-77. https://doi.org/10.1145/3650028
[5]
Saha, S., Hota, A., Chattopadhyay, A.K., Nag, A. and Nandi, S. (2024) A Multifaceted Survey on Privacy Preservation of Federated Learning: Progress, Challenges, and Opportunities. Artificial Intelligence Review, 57, Article No. 184. https://doi.org/10.1007/s10462-024-10766-7
[6]
Wang, B., Li, H., Ren, X. and Guo, Y. (2023) An Efficient Differential Privacy-Based Method for Location Privacy Protection in Location-Based Services. Sensors, 23, Article 5219. https://doi.org/10.3390/s23115219
[7]
Wei, K., Li, J., Ding, M., Ma, C., Yang, H.H., Farhad, F., Jin, S., Quek, T.Q.S. and Poor, H.V. (2019) Federated Learning with Differential Privacy: Algorithms and Performance Analysis. arXiv: 1911.00222. http://arxiv.org/abs/1911.00222
[8]
Gong, Y., Chang, X., Mi？i？, J., Mi？i？, V.B., Wang, J. and Zhu, H. (2024) Practical Solutions in Fully Homomorphic Encryption: A Survey Analyzing Existing Acceleration Methods. Cybersecurity, 7, Article No. 5. https://doi.org/10.1186/s42400-023-00187-4
[9]
Hussien, N., Hussien, N.M., Salman, S.A. and Aljanabi, M. (2023) Secure Federated Learning with a Homomorphic Encryption Model. InternationalJournalPapierAdvanceandScientificReview, 4, 1-7. https://doi.org/10.47667/ijpasr.v4i3.235
[10]
Moshawrab, M., Adda, M., Bouzouane, A., Ibrahim, H. and Raad, A. (2024) Securing Federated Learning: Approaches, Mechanisms and Opportunities. Electronics, 13, Article 3675. https://doi.org/10.3390/electronics13183675
[11]
Munjal, K. and Bhatia, R. (2022) A Systematic Review of Homomorphic Encryption and Its Contributions in Healthcare Industry. Complex&IntelligentSystems, 9, 3759-3786. https://doi.org/10.1007/s40747-022-00756-z
[12]
Nikfam, F., Casaburi, R., Marchisio, A., Martina, M. and Shafique, M. (2023) A Homomorphic Encryption Framework for Privacy-Preserving Spiking Neural Networks. Information, 14, Article 537. https://doi.org/10.3390/info14100537
[13]
Byrd, D. and Polychroniadou, A. (2020) Differentially Private Secure Multi-Party Computation for Federated Learning in Financial Applications. Proceedings of the First ACM International Conference on AI in Finance, New York, 15-16 October 2020, 1-9. https://doi.org/10.1145/3383455.3422562
[14]
Hosseini, S.M., Sikaroudi, M., Babaei, M. and Tizhoosh, H.R. (2022) Cluster Based Secure Multi-Party Computation in Federated Learning for Histopathology Images. In: Albarqouni, S., et al., Eds., Distributed, Collaborative, and Federated Learning, and Affordable AI and Healthcare for Resource Diverse Global Health, Springer, 110-118. https://doi.org/10.1007/978-3-031-18523-6_11
[15]
Liu, F., Zheng, Z., Shi, Y., Tong, Y. and Zhang, Y. (2023) A Survey on Federated Learning: A Perspective from Multi-Party Computation. FrontiersofComputerScience, 18, Article No. 181336. https://doi.org/10.1007/s11704-023-3282-7
[16]
Biswas, S., Kermarrec, A.M., Pires, R., Sharma, R. and Vujasinovic, M. (2024) Secure Aggregation Meets Sparsification in Decentralized Learning. arXiv: 2405.07708.
[17]
Pasquini, D., Francati, D. and Ateniese, G. (2022) Eluding Secure Aggregation in Federated Learning via Model Inconsistency. Proceedings of the 2022 ACM SIGSAC Conference on Computer and Communications Security, Los Angeles, 7-11 November 2022, 2429-2443. https://doi.org/10.1145/3548606.3560557
[18]
Zhang, Y., Behnia, R., Yavuz, A.A., Ebrahimi, R. and Bertino, E. (2024) Uncovering Attacks and Defenses in Secure Aggregation for Federated Deep Learning. 2024 IEEE International Conference on Data Mining Workshops (ICDMW), Abu Dhabi, 9 December 2024, 650-656. https://doi.org/10.1109/icdmw65004.2024.00090
[19]
Casella, B. and Fonio, S. (2023). Architecture-Based FedAvg for Vertical Federated Learning. Proceedings of the IEEE/ACM 16th International Conference on Utility and Cloud Computing, Taormina, 4-7 December 2023, 1-6. https://doi.org/10.1145/3603166.3632559
[20]
Wan, W., Hu, S., Lu, J., Zhang, L.Y., Jin, H. and He, Y. (2022) Shielding Federated Learning: Robust Aggregation with Adaptive Client Selection. Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, Vienna, 23-29 July 2022, 753-760. https://doi.org/10.24963/ijcai.2022/106
[21]
Sun, T., Li, D. and Wang, B. (2023) Decentralized Federated Averaging. IEEETransactionsonPatternAnalysisandMachineIntelligence, 45, 4289-4301. https://doi.org/10.1109/tpami.2022.3196503
[22]
Zhang, H., Wu, T., Cheng, S. and Liu, J. (2024) CC-FedAvg: Computationally Customized Federated Averaging. IEEEInternetofThingsJournal, 11, 4826-4841. https://doi.org/10.1109/jiot.2023.3300080
[23]
Zhou, Y., Qing, Y. and Lv, J. (2021) Communication-Efficient Federated Learning with Compensated Overlap-FedAvg. arXiv: 2012.06706.
[24]
Moshawrab, M., Adda, M., Bouzouane, A., Ibrahim, H. and Raad, A. (2023) Reviewing Federated Learning Aggregation Algorithms; Strategies, Contributions, Limitations and Future Perspectives. Electronics, 12, Article 2287. https://doi.org/10.3390/electronics12102287
[25]
Jiang, D., Shan, C. and Zhang, Z. (2020) Federated Learning Algorithm Based on Knowledge Distillation. 2020 International Conference on Artificial Intelligence and Computer Engineering (ICAICE), Beijing, 23-25 October 2020, 163-167. https://doi.org/10.1109/icaice51518.2020.00038
[26]
Li, D. and Wang, J. (2019) FedMD: Heterogenous Federated Learning via Model Distillation. arXiv: 1910.03581.
[27]
Mora, A., Tenison, I., Bellavista, P. and Rish, I. (2024) Knowledge Distillation in Federated Learning: A Practical Guide. Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence, Jeju, 3-9 August 2024, 8188-8196. https://doi.org/10.24963/ijcai.2024/905
[28]
Song, C., Saxena, D., Cao, J. and Zhao, Y. (2024) FedDistill: Global Model Distillation for Local Model De-Biasing in Non-IID Federated Learning. arXiv: 2404.09210.
[29]
Wu, C., Wu, F., Lyu, L., Huang, Y. and Xie, X. (2022) Communication-Efficient Federated Learning via Knowledge Distillation. NatureCommunications, 13, Article No. 2032. https://doi.org/10.1038/s41467-022-29763-x
[30]
Yang, G. and Tae, H. (2023) Federated Distillation Methodology for Label-Based Group Structures. AppliedSciences, 14, Article 277. https://doi.org/10.3390/app14010277
[31]
Boix-Adsera, E. (2024) Towards a Theory of Model Distillation. arXiv: 2403.09053.
[32]
Kotsilieris, T., Anagnostopoulos, I. and Livieris, I.E. (2022) Special Issue: Regularization Techniques for Machine Learning and Their Applications. Electronics, 11, Article 521. https://doi.org/10.3390/electronics11040521
[33]
Liu, L. and Zhou, D. (2025) Analysis of Regularized Federated Learning. Neurocomputing, 611, Article ID: 128579. https://doi.org/10.1016/j.neucom.2024.128579
[34]
Lv, Y., Ding, H., Wu, H., Zhao, Y. and Zhang, L. (2023) FedRDS: Federated Learning on Non-IID Data via Regularization and Data Sharing. AppliedSciences, 13, Article 12962. https://doi.org/10.3390/app132312962
[35]
Xu, Q., Cohn, T. and Ohrimenko, O. (2023) Fingerprint Attack: Client Deanonymization in Federated Learning. ECAI 2023: 26th European Conference on Artificial Intelligence, Kraków, 30 September-4 October 2023, 2792-2801.
[36]
Ma, C., Li, J., Ding, M., Yang, H.H., Shu, F., Quek, T.Q.S. and Poor, H.V. (2020) On Safeguarding Privacy and Security in the Framework of Federated Learning. arXiv: 1909.06512. http://arxiv.org/abs/1909.06512
