View in EDS

Secure Aggregation with Computational Scalability Based on Additive Homomorphic Encryption.

Saved in:
Bibliographic Details
Title: Secure Aggregation with Computational Scalability Based on Additive Homomorphic Encryption.
Authors: ZE YANG1,2 gzuyangze@gmail.com, YOULIANG TIAN1,2 yltian@gzu.edu.cn, MENGQIAN LI1,2 mqli88@I63.com, KUN NIU2 kniu@gzu.edu.cn
Source: Journal of Information Science & Engineering. Jul2025, Vol. 41 Issue 4, p987-1007. 21p.
Subject Terms: Distributed computing, Cloud computing, Parallel programming, Computer systems, Big data
Abstract: Cloud-based computing framework resolves the dilemma of users' computational capabilities mismatching the demand for high-quality models. However, the explosive growth of data due to the massive popularity of terminal devices not only imposes higher requirements on the performance of the cloud but also increases the risk of private data leakage in the cloud. In this paper, we focus on cloud-based secure aggregation with private data on cloud devices, where the cloud can only handle fixed bit lengths. First, we propose a truncate-mapping scheme for private data to alleviate the resource limitations of the cloud by encoding big data into different shares for parallel computing. Second, we define the calculations on the encoded data, involving the secure addition and secure comparison, to achieve secure aggregation on private data. It is worth noting that not only the results of the computations on the encoded data are the same as on the raw private data, but also the proposed scheme can make full use of the power of the computing devices. That is the more devices concerning the computation, the more precise results can be obtained instead of steadfastly enhancing the computing power of the cloud. Finally, the theoretical analysis and experiments show that the proposed scheme is secure, effective and suitable for practical applications. [ABSTRACT FROM AUTHOR]
Database: Supplemental Index
Description
Abstract:Cloud-based computing framework resolves the dilemma of users' computational capabilities mismatching the demand for high-quality models. However, the explosive growth of data due to the massive popularity of terminal devices not only imposes higher requirements on the performance of the cloud but also increases the risk of private data leakage in the cloud. In this paper, we focus on cloud-based secure aggregation with private data on cloud devices, where the cloud can only handle fixed bit lengths. First, we propose a truncate-mapping scheme for private data to alleviate the resource limitations of the cloud by encoding big data into different shares for parallel computing. Second, we define the calculations on the encoded data, involving the secure addition and secure comparison, to achieve secure aggregation on private data. It is worth noting that not only the results of the computations on the encoded data are the same as on the raw private data, but also the proposed scheme can make full use of the power of the computing devices. That is the more devices concerning the computation, the more precise results can be obtained instead of steadfastly enhancing the computing power of the cloud. Finally, the theoretical analysis and experiments show that the proposed scheme is secure, effective and suitable for practical applications. [ABSTRACT FROM AUTHOR]
ISSN:10162364
DOI:10.6688/JISE.202507_41(4).0014