An Efficient Data Encryption Framework for Secure Cloud Storage Systems

Abstract

Cloud storage reduces local infrastructure cost and improves data availability, but outsourcing sensitive information to infrastructure outside the data owner’s direct control creates persistent confidentiality, integrity, and access-control risks. This paper presents a methodology for an efficient hybrid data-encryption framework designed to secure files before they are transferred to cloud storage. The framework separates bulk-data encryption from key protection: file contents are protected with AES-256 in Galois/Counter Mode (AES-GCM), while the per-file symmetric key is wrapped with RSA-2048 using OAEP. Authenticated encryption detects unauthorized ciphertext modification, and an independent policy/key service prevents the cloud-storage provider from receiving plaintext encryption keys. Fine-grained authorization concepts are incorporated at the key-release layer so that storage access and decryption permission are treated as separate decisions. The evaluation methodology combines security-property analysis with a reproducible prototype benchmark across 1, 5, 10, and 25 MB files. Results show nearly linear processing time with file size and negligible storage expansion because public-key cryptography is applied only to the small data-encryption key. The framework therefore provides a practical balance between confidentiality, integrity, controlled key distribution, and computational efficiency for cloud storage applications.