Secure verifiable aggregation for blockchain-based federated averaging
IoT devices’ storage and computation capacities are constantly increasing in recent years, which brings critical challenges in data privacy protection. Federated learning (FL) and blockchain technology are two popular techniques used in IoT data aggregation, where FL enables data training with priva...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-03-01
|
| Series: | High-Confidence Computing |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2667295221000362 |
| _version_ | 1818306333237051392 |
|---|---|
| author | Saide Zhu Ruinian Li Zhipeng Cai Donghyun Kim Daehee Seo Wei Li |
| author_facet | Saide Zhu Ruinian Li Zhipeng Cai Donghyun Kim Daehee Seo Wei Li |
| author_sort | Saide Zhu |
| collection | DOAJ |
| description | IoT devices’ storage and computation capacities are constantly increasing in recent years, which brings critical challenges in data privacy protection. Federated learning (FL) and blockchain technology are two popular techniques used in IoT data aggregation, where FL enables data training with privacy protection, and blockchain provides a decentralized architecture for data storage and mining. However, very few the state-of-the-art works consider the applicability of the combination of FL and blockchain. In this paper, we adopt the federated averaging algorithm to reduce the communication overhead between the blockchain and end users to achieve higher performance. We also apply the double-mask-then-encrypt approach for end users to submit their local updates in order to protect data privacy. Finally, we propose and implement a non-interactive Public Verifiable Secret Sharing (PVSS) algorithm with Distributed Hash Table (DHT) that solves the user-drop-out problem and improves the communication efficiency between blockchain and end-users. At last, we theoretically analyze the security strengths of the proposed solution and conduct experiments to measure the execution time of PVSS on both the server and clients sides. |
| first_indexed | 2024-12-13T06:40:49Z |
| format | Article |
| id | doaj.art-0bd494dd70834f1393449e61669d1e72 |
| institution | Directory Open Access Journal |
| issn | 2667-2952 |
| language | English |
| last_indexed | 2024-12-13T06:40:49Z |
| publishDate | 2022-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | High-Confidence Computing |
| spelling | doaj.art-0bd494dd70834f1393449e61669d1e722022-12-21T23:56:25ZengElsevierHigh-Confidence Computing2667-29522022-03-0121100046Secure verifiable aggregation for blockchain-based federated averagingSaide Zhu0Ruinian Li1Zhipeng Cai2Donghyun Kim3Daehee Seo4Wei Li5Department of Computer Science, Georgia State University, Atlanta, USADepartment of Computer Science, Bowling Green State University, Bowling Green, Ohio, USADepartment of Computer Science, Georgia State University, Atlanta, USADepartment of Computer Science, Georgia State University, Atlanta, USAInformation Security Lab, Sangmyung University, Seoul, KoreaCorresponding author.; Department of Computer Science, Georgia State University, Atlanta, USAIoT devices’ storage and computation capacities are constantly increasing in recent years, which brings critical challenges in data privacy protection. Federated learning (FL) and blockchain technology are two popular techniques used in IoT data aggregation, where FL enables data training with privacy protection, and blockchain provides a decentralized architecture for data storage and mining. However, very few the state-of-the-art works consider the applicability of the combination of FL and blockchain. In this paper, we adopt the federated averaging algorithm to reduce the communication overhead between the blockchain and end users to achieve higher performance. We also apply the double-mask-then-encrypt approach for end users to submit their local updates in order to protect data privacy. Finally, we propose and implement a non-interactive Public Verifiable Secret Sharing (PVSS) algorithm with Distributed Hash Table (DHT) that solves the user-drop-out problem and improves the communication efficiency between blockchain and end-users. At last, we theoretically analyze the security strengths of the proposed solution and conduct experiments to measure the execution time of PVSS on both the server and clients sides.http://www.sciencedirect.com/science/article/pii/S2667295221000362BlockchainFederated learningSecret sharingPrivacy |
| spellingShingle | Saide Zhu Ruinian Li Zhipeng Cai Donghyun Kim Daehee Seo Wei Li Secure verifiable aggregation for blockchain-based federated averaging High-Confidence Computing Blockchain Federated learning Secret sharing Privacy |
| title | Secure verifiable aggregation for blockchain-based federated averaging |
| title_full | Secure verifiable aggregation for blockchain-based federated averaging |
| title_fullStr | Secure verifiable aggregation for blockchain-based federated averaging |
| title_full_unstemmed | Secure verifiable aggregation for blockchain-based federated averaging |
| title_short | Secure verifiable aggregation for blockchain-based federated averaging |
| title_sort | secure verifiable aggregation for blockchain based federated averaging |
| topic | Blockchain Federated learning Secret sharing Privacy |
| url | http://www.sciencedirect.com/science/article/pii/S2667295221000362 |
| work_keys_str_mv | AT saidezhu secureverifiableaggregationforblockchainbasedfederatedaveraging AT ruinianli secureverifiableaggregationforblockchainbasedfederatedaveraging AT zhipengcai secureverifiableaggregationforblockchainbasedfederatedaveraging AT donghyunkim secureverifiableaggregationforblockchainbasedfederatedaveraging AT daeheeseo secureverifiableaggregationforblockchainbasedfederatedaveraging AT weili secureverifiableaggregationforblockchainbasedfederatedaveraging |