A Survey of Methods and Architectures for Secure Cloud Data Storage and Retrieval Using Giant Trevally Optimizer with Quantum Convolutional Neural Network-Based Encryption Algorithm

Abstract

Cloud computing has become a fundamental infrastructure for modern information systems due to its scalability, flexibility, and cost-efficient data storage capabilities. However, the rapid growth of cloud-based services has also introduced serious security challenges related to data confidentiality, integrity, authentication, and secure data retrieval. Sensitive information stored on remote cloud servers is vulnerable to cyberattacks such as unauthorized access, data breaches, and malicious insider activities. Therefore, designing secure cloud storage and retrieval mechanisms has become a major research priority in recent years. Recent advances in artificial intelligence (AI), metaheuristic optimization algorithms, and quantum computing techniques have opened new opportunities for improving cloud data security frameworks. Optimization algorithms are widely used for solving complex resource management and security optimization problems in distributed computing environments. Among these algorithms, the Giant Trevally Optimizer (GTO) has recently attracted attention as a powerful nature-inspired metaheuristic algorithm that mimics the hunting behaviour of giant trevally fish. The algorithm demonstrates strong exploration and exploitation capabilities for solving global optimization problems and has been successfully applied to complex engineering tasks.  At the same time, deep learning-based encryption models have emerged as promising approaches for protecting cloud data. Convolutional Neural Networks (CNNs) can generate complex transformations and encryption patterns that significantly increase the difficulty of cryptanalysis. Recent studies have proposed neural network-based encryption frameworks capable of protecting cloud-stored data from unauthorized access.