Adaptive Multi-Layered Elgamal Cryptosystem with Machine Learning-Based Security for Cloud Data Allocation and Access Control

Abstract

The advancements in cloud computing have been happening at a high pace, where there has been an ever-growing need to offer data security. This work proposes an Adaptive Multi-Layered ElGamal Cryptosystem (AMLEC), a new security model incorporating heterogeneous cryptographic methods along with optimal cloud data distribution and access management. AMLEC is stabilized beyond standard ElGamal cryptosystem by supporting different input-output channels of encryption, metadata-oriented binary translation and attribute-based key derivation to protect safe encryption. Adaptive security guard employs machine learning for one-time three-key encryption policy with fuzzy-level security choice selection. To provide secure cloud storage and use, the system uses an Optimized Allocation Strategy (OAS) to dynamically make decisions regarding appropriate storage providers based on security and performance needs. In a multi-layered security framework is suggested with a Cloud Service Provider (CSP) having a sophisticated job scheduler, a Sensitive Document Analyzer for encrypted access control and classification and User/Tenant Tracker for integrity and transparency. The proposed AMLEC framework offers privacy, high-quality cryptography and intelligent resource management and hence forms a robust solution to secure cloud computing environments.