Squeeze Belief Network‐Based Key Management and Data Deduplication for Secure Cloud Storage

ABSTRACT Data deduplication is a crucial data compression technique for eliminating duplicate copies of repetitive data that reduces the bandwidth usage and storage space from cloud service providers (CSP). Data deduplication in cloud computing gained vast attention in large‐scale storage systems, where the main issue comes with security concerns. As the confidentiality of sensitive data gets reduced by data deduplication, a Deep Learning (DL) model is used for data security. In this research, Squeeze Fused Belief Network (SFBN)‐DeepkeyGen is proposed for secure data storage in the cloud environment. Initially, the file is uploaded by the user to the cloud server, which is then allowed to check deduplication. Here, the secret key is generated by SFBN‐DeepkeyGen and then the tag is generated. If the tag is not available, then the file is encrypted using the Advanced Encryption Standard (AES) algorithm. If the tag is available, then the Proof of Ownership (PoW) is checked to perform data deduplication. Finally, the experimental results revealed that SFBN‐DeepkeyGen achieved minimal encryption time, decryption time, and maximal throughput of 0.187 s, 0.197 s, and 0.817 Mbps.