A Blockchain-Based Decentralized Architecture for Federated Learning: A Case Study on Consensus Mechanisms

• Francinaldo Barbosa UFPI
• Luis Guilherme Silva UFPI
• Iure Fé UFPI
• Israel Cardoso UFPI
• Alex Vieira UFJF
• Geraldo P. Rocha Filho UESB
• Francisco Airton Silva UFPI

DOI: https://doi.org/10.5753/sbrc.2026.19318

Abstract

Federated Learning has consolidated itself as an alternative to centralized models by mitigating privacy risks, while its integration with Blockchain seeks to eliminate single points of failure and enhance trust in distributed training. However, real-world experiments are costly, and inadequate architectures may introduce inefficiencies and vulnerabilities with significant impacts on performance and security. This work evaluates, through simulations conducted in the FLEX framework, the impact of the Proof-of-Work (PoW) and Proof-of-Federated-Learning (PoFL) consensus mechanisms on Blockchain-based Federated Learning systems. The results indicate that PoFL, driven by model performance, achieves accuracy above 80%, accelerates convergence, and reduces the number of training rounds, at the expense of higher computational cost per round and lower throughput when compared to PoW. In turn, PoW enables faster network validation but results in a slower and less stable learning process. The analysis highlights the trade-offs among model quality, training stability, and operational costs.

References

Beutel, D. J., Topal, T., Mathur, A., Qiu, X., Fernandez-Marques, J., Gao, Y., Sani, L., Li, K. H., Parcollet, T., De Gusmão, P. P. B., et al. (2020). Flower: A friendly federated learning research framework. arXiv preprint arXiv:2007.14390.

Caldas, S., Duddu, S. M. K., Wu, P., Li, T., Konečnỳ, J., McMahan, H. B., Smith, V., and Talwalkar, A. (2018). Leaf: A benchmark for federated settings. arXiv preprint arXiv:1812.01097.

Chen, H., Asif, S. A., Park, J., Shen, C., and Bennis, M. (2021). Robust blockchained federated learning with model validation and proof-of-stake inspired consensus. CoRR, abs/2101.03300.

Herrera, F., Jiménez-López, D., Argente-Garrido, A., Rodríguez-Barroso, N., Zuheros, C., Aguilera-Martos, I., Bello, B., García-Márquez, M., and Luzón, M. V. (2025). Flex: Flexible federated learning framework. Information Fusion, 117:102792.

Issa, W., Moustafa, N., Turnbull, B., Sohrabi, N., and Tari, Z. (2023). Blockchain-based federated learning for securing internet of things: A comprehensive survey. ACM Computing Surveys, 55(9):1–43.

Kairouz, P., McMahan, H. B., Avent, B., Bellet, A., Bennis, M., Bhagoji, A. N., Bonawitz, K., Charles, Z., Cormode, G., Cummings, R., et al. (2021). Advances and open problems in federated learning. Foundations and trends® in machine learning, 14(1–2):1–210.

Li, Q., Wen, Z., Wu, Z., Hu, S., Wang, N., Li, Y., Liu, X., and He, B. (2021). A survey on federated learning systems: Vision, hype and reality for data privacy and protection. IEEE Transactions on Knowledge and Data Engineering, 35(4):3347–3366.

Li, T., Sahu, A. K., Talwalkar, A., and Smith, V. (2020). Federated learning: Challenges, methods, and future directions. IEEE signal processing magazine, 37(3):50–60.

Li, Y., Xia, C., Lin, W., and Wang, T. (2024). Ppbfl: A privacy protected blockchain-based federated learning model. arXiv preprint arXiv:2401.01204.

Liu, H., Zhu, F., and Cheng, L. (2025). Proof-of-data: A consensus protocol for collaborative intelligence. arXiv preprint arXiv:2501.02971.

Liu, L., Hu, Y., Zhao, Y., Zhang, X., Ma, Y., and Chang, G. (2024). A novel federated learning system with privacy protection and blockchain consensus incentive mechanisms in cloud-edge collaboration scenarios. In 2024 IEEE International Conference on Systems, Man, and Cybernetics (SMC), pages 3111–3118.

Qu, X., Wang, S., Hu, Q., and Cheng, X. (2021). Proof of federated learning: A novel energy-recycling consensus algorithm. IEEE Transactions on Parallel and Distributed Systems, 32(8):2074–2085.

Ro, J. H., Suresh, A. T., and Wu, K. (2021). Fedjax: Federated learning simulation with jax. arXiv preprint arXiv:2108.02117.

Rosano, B., Bittencourt, L., Anjos, J., Técnico-IC-PFG, R., and de Graduação, P. F. (2022). Aprendizado federado hierárquico.

Tian, Y., Guo, Z., Zhang, J., and Al-Ars, Z. (2023). Dfl: High-performance blockchain-based federated learning. Distributed Ledger Technologies: Research and Practice, 2(3):1–25.

Wang, Y., Peng, H., Su, Z., Luan, T. H., Benslimane, A., and Wu, Y. (2022a). A platform-free proof of federated learning consensus mechanism for sustainable blockchains. IEEE Journal on Selected Areas in Communications, 40(12):3305–3324.

Wang, Y., Zhou, J., Feng, G., Niu, X., and Qin, S. (2022b). Blockchain assisted federated learning for enabling network edge intelligence. IEEE Network, 37(1):96–102.

Xin, W., Jiaqian, L., Xueshuang, D., Haoji, Z., and Lianshan, S. (2024). A survey of differential privacy techniques for federated learning. IEEE Access.

Yuan, J., Liu, W., Shi, J., and Li, Q. (2025). Approximate homomorphic encryption based privacy-preserving machine learning: a survey. Artificial Intelligence Review, 58(3):82.

Zhao, Y., Qu, Y., Xiang, Y., Chen, F., and Gao, L. (2024a). Context-aware consensus algorithm for blockchain-empowered federated learning. IEEE Transactions on Cloud Computing, 12(2):491–503.

Zhao, Y., Qu, Y., Xiang, Y., Chen, F., and Gao, L. (2024b). Long-term proof-of-contribution: An incentivized consensus algorithm for blockchain-enabled federated learning. IEEE Transactions on Services Computing, 17(5):2558–2570.

Zhu, J., Cao, J., Saxena, D., Jiang, S., and Ferradi, H. (2023). Blockchain-empowered federated learning: Challenges, solutions, and future directions. ACM Computing Surveys, 55(11):1–31.