Title: FFHE-SSC: a robust framework for performing statistical computation on encrypted data

Authors: Abdullah Moonis; Ajeet Singh

Addresses: School of Computing Science and Engineering (SCSE), Vellore Institute of Technology – VIT Bhopal University, Bhopal-Indore Highway, Sehore, Madhya Pradesh, 466 114, India ' School of Computing Science and Engineering (SCSE), Vellore Institute of Technology – VIT Bhopal University, Bhopal-Indore Highway, Sehore, Madhya Pradesh, 466 114, India

Abstract: In the field of data privacy and security, performing computations on encrypted data without compromising confidentiality presents a significant challenge. The fast fully homomorphic encryption for secure statistical computation (FFHE-SSC) framework, introduced in this paper, directly addresses this challenge. Utilising advanced cryptographic techniques, including fully homomorphic encryption (FHE), the framework facilitates robust statistical analysis on encrypted data, thereby ensuring the security of sensitive information throughout the analysis process. Empirical evaluations and analyses have shown that FFHE-SSC not only preserves the integrity and confidentiality of data but also achieves computational performance viable for practical applications. Moreover, the framework's adaptability to various data types and its applicability across diverse sectors which require privacy-preserving data analysis, is also examined.

Keywords: homomorphic encryption; privacy preserving machine learning; private statistical computations; finite field; data confidentiality; cryptographic techniques; integrated AI.

DOI: 10.1504/IJCSM.2024.143208

International Journal of Computing Science and Mathematics, 2024 Vol.20 No.4, pp.308 - 324

Received: 26 Jan 2024
Accepted: 19 Jul 2024
Published online: 09 Dec 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article