CuLOA-based Data Encryption with Tuned Key for Privacy Preservation in the Cloud

Abstract

Listen

Translate article

Preservation of data privacy in cloud computing involves securing sensitive data during analysis and storage. Conventional approaches often use techniques such as encryption and differential privacy, but they can be computationally intensive and may still risk data leakage through indirect inferences. These limitations necessitate advanced methods to balance efficiency and robust privacy. To address this, this study proposes a novel approach using DL-based fine-tuned keys for encrypting the data, aimed at preserving data privacy in the cloud through the Coati Lyrebird Optimization Algorithm (CuLOA) approach. Initially, sensitive data is randomly chosen from the database. Then, the optimal key is derived using the CuLOA approach. This key, along with the sensitive data is input into the SqueezeNet model, which generates a fine-tuned optimal key. Subsequently, the sensitive data are encrypted and stored in cloud storage. Finally, the encrypted data and the optimally tuned key are employed in the data decryption process to recover the original. A comparative experimental analysis showed that the proposed CuLOA approach was better than previous schemes.