EA
Federated Learning-based Algorithm Design for Privacy Preservation in Cross-domain Data Sharing
TOTAL VIEWS: 238
Abstract
In the context of accelerating digital integration across industries, cross-domain data sharing has emerged as a critical enabler of intelligent decision-making and resource coordination. However, the sensitivity of data and disputes over ownership have made privacy protection the primary obstacle in collaborative environments. Federated learning, as a promising paradigm for distributed collaborative modeling, offers the key advantage of “data usability without visibility,” enabling privacy-aware computation. This study addresses major challenges—including data heterogeneity, regulatory compliance, and security threats—by proposing a privacy-enhanced federated algorithm framework. The design integrates multiple mechanisms such as dynamic aggregation, local perturbation, consistency regularization, and multi-layered security protection. Through this integrated approach, the framework enhances both the efficiency and security of federated models in heterogeneous scenarios, paving the way for a data-sharing ecosystem that safeguards privacy while enabling collaborative intelligence across critical domains.
Keywords
Federated Learning Privacy protection; Cross-domain collaboration Algorithm design Security mechanism
References
[1] Zhu H, Huang L, Xie Z. Privacy attack in federated learning is not easy: an experimental study. Complex Intell Syst. 2025;11(9):391.
[2] Yang L, Tong W, Li Z, et al. PECD-DSIIoT: Privacy-enhanced cross-domain data sharing scheme for IIoT. J Inf Secur Appl. 2025;93:104128.
[3] Guan M, Bao H, Wang J, et al. PEFed: Enhancing privacy and efficiency in federated learning via removable perturbation and decentralized encryption. Inf Fusion. 2025;122:103187.
[4] Jiang J, Pei T, Chen J, et al. CDAS: A Secure Cross-Domain Data Sharing Scheme Based on Blockchain. Information. 2025;16(5):394.
[5] Fan H, Fan X, Wei W, et al. Privacy preserving ultra-short-term prediction in clustered wind farms with encrypted data sharing: A secure multi-party computation approach. Expert Syst Appl. 2025;278:127218.
[6] Yang X, Zhang C. A Location Trajectory Privacy Protection Method Based on Generative Adversarial Network and Attention Mechanism. Comput Mater Contin. 2024;81(3):3781-804.
[7] Linwei F, Liming W, Hongjia L. Iterative and mixed-spaces image gradient inversion attack in federated learning. Cybersecurity. 2024;7(1).
[8] Mansoor A, Hadis K, Muhammad T. Integration of Blockchain and Federated Learning for Internet of Things: Recent Advances and Future Challenges. Comput Secur. 2021;102355.
How to cite this paper
Federated Learning-based Algorithm Design for Privacy Preservation in Cross-domain Data Sharing
How to cite this paper: Yuxin Wu. (2026). Federated Learning-based Algorithm Design for Privacy Preservation in Cross-domain Data Sharing. Engineering Advances, 6(1), 36-40.
DOI: http://dx.doi.org/10.26855/ea.2026.03.008
More Articles
- Exploration of Data Center Cost Optimization Pathways Under Multi-generation CPU and GPU Collaborative Architectures
- Federated Learning-based Algorithm Design for Privacy Preservation in Cross-domain Data Sharing
- System Performance Evaluation Principle of Heterogeneous Optical Devices in Parallel Use
- Design and Implementation of a Pneumatic Compliant Rehabilitation and Piano Instruction Intelligent Glove System
- Research on Warehouse Capacity Optimization Methods Based on Predictive Modeling
- Construction Precision Control and Quality Improvement of Aluminum Formwork Systems in Residential Buildings