Datasets
Standard Dataset
Privacy-preserving approach to edge federated learning based on blockchain and fully homomorphic encryption
- Citation Author(s):
-
Guo BaiqiGuo
Guilin University of Technology - Submitted by:
- Baiqi Guo
- Last updated:
- Wed, 09/04/2024 - 02:42
- DOI:
- 10.21227/5gmq-kd84
- Data Format:
- License:
61 Views- Categories:
- Keywords:
Abstract
The data in this dataset is the experimental data related to the article named Privacy-preserving approach to edge federated learning based on blockchain and fully homomorphic encryption , which contains data such as running time comparison, communication spend comparison, encryption and decryption time comparison, accuracy comparison, etc.
A summary of the article follows:To address the issues concerned with high risk of single-point failure, weak privacy protection and poor resistance to poisoning attacks in edge federated learning, An edge federatied learning privacy protection scheme based on blockchain and full homomorphic encryption is proposed. This approach utilises blockchain to provide edge federatied learning with the characteristics of anti-tampering ,anti-single-point failure and data process transparency combining the CKKS full homomorphic encryption scheme to encrypt relevant computational parameters, thus reduce the risk of privacy leakage. Additionally, an unsupervised model parameter update identification mechanism is designed, using the consistency between historical model updates of edge servers as the identification basis, and enhances the accuracy of aggregation model while identifying malicious edge server updates. The experimental results demonstrate that the proposed method is capable of resisting 70% of poisoning attacks from malicious edge servers while providing privacy protection, encryption, and maintaining the integrity of the blockchain simultaneously. Furthermore, it is able to achieve high model accuracy and meets the stringent requirements for security, accuracy, and traceability commonly associated with edge federated learning scenarios.
The data in this dataset is the experimental data related to the article named Privacy-preserving approach to edge federated learning based on blockchain and fully homomorphic encryption , which contains data such as running time comparison, communication spend comparison, encryption and decryption time comparison, accuracy comparison, etc.
The experimental data contained therein are automatically generated through the experimental code to record and store in the form of EXCEL table, which is convenient for drawing diagrams as well as analyzing the results.
A summary of the article follows:To address the issues concerned with high risk of single-point failure, weak privacy protection and poor resistance to poisoning attacks in edge federated learning, An edge federatied learning privacy protection scheme based on blockchain and full homomorphic encryption is proposed. This approach utilises blockchain to provide edge federatied learning with the characteristics of anti-tampering ,anti-single-point failure and data process transparency combining the CKKS full homomorphic encryption scheme to encrypt relevant computational parameters, thus reduce the risk of privacy leakage. Additionally, an unsupervised model parameter update identification mechanism is designed, using the consistency between historical model updates of edge servers as the identification basis, and enhances the accuracy of aggregation model while identifying malicious edge server updates. The experimental results demonstrate that the proposed method is capable of resisting 70% of poisoning attacks from malicious edge servers while providing privacy protection, encryption, and maintaining the integrity of the blockchain simultaneously. Furthermore, it is able to achieve high model accuracy and meets the stringent requirements for security, accuracy, and traceability commonly associated with edge federated learning scenarios.
Comments
Dear author, when will the dataset files be uploaded? I am eagerly waiting
Hello, my datasets are FASHION-MNIST and MNIST, both of which are classic datasets that you can add directly in pytorch training, the above file only contains my test results.The origin of the test results can be derived in the article.
Thank you