Homomorphic Encryption-Based Secure Data Retrieval in Cloud Storage

Abstract

The rapid adoption of cloud storage services has transformed data management by offering scalable, flexible, and cost-effective solutions. However, outsourcing sensitive data to third-party cloud providers raises significant privacy and security concerns, particularly regarding unauthorized access and data breaches. Secure data retrieval from encrypted cloud storage is therefore a critical challenge. Homomorphic encryption (HE), a cryptographic technique enabling computations directly on encrypted data without requiring decryption, provides a promising approach to address this issue. This paper explores the application of homomorphic encryption for secure data retrieval in cloud storage environments, focusing on preserving data confidentiality while supporting efficient query execution. We analyze different homomorphic encryption schemes—partially, somewhat, and fully homomorphic encryption—and evaluate their feasibility for enabling secure keyword search and range queries over encrypted datasets. A novel retrieval framework is proposed that integrates homomorphic encryption with optimized indexing techniques to reduce computational overhead and improve response times. Experimental evaluations on benchmark datasets demonstrate that the proposed framework effectively balances security, privacy, and efficiency. The results show that homomorphic encryption-based retrieval methods can prevent data leakage during search operations, even against semi-honest cloud providers. Furthermore, the study discusses practical considerations, including key management, scalability, and resistance to various attack models. This research contributes to advancing privacy-preserving cloud storage solutions, empowering users to securely store and retrieve sensitive information without compromising confidentiality or performance.