Improving Digital Forensic Security Using Multi-Key Homomorphic Encryption and DFA-AOKGE

Abstract

The escalating complexity of cyberattacks and the vulnerabilities in current digital forensic frameworks have highlighted the critical need for secure, efficient, and tamper-proof evidence management solutions. This study proposes the Digital Forensic Architecture using Authentication with Optimal Key Generation Encryption (DFA-AOKGE), a novel framework designed to enhance forensic security in cloud and IoT environments. The primary objective is to integrate blockchain-based immutability, dynamic cryptographic key generation, multi-key homomorphic encryption, and secure biometric authentication to overcome the limitations of traditional forensic systems. The proposed methodology utilizes Multi-Key Homomorphic Encryption (MHE) to perform computations directly on encrypted forensic evidence, ensuring end-to-end confidentiality. A lightweight blockchain architecture employing Proof of Authority (PoA) consensus records all evidence operations immutably. Dynamic cryptographic keys are generated using the Enhanced Equilibrium Optimizer (EEO), while user authentication is fortified through a Secure Biometric Verification Mechanism (SBVM). The system was evaluated on a simulated dataset of 5000 forensic records, using an 80:20 train-test split and 5-fold cross-validation. Experimental results demonstrate that DFA-AOKGE achieves a 96.7% authentication accuracy, reduces encryption time to 100 milliseconds, and maintains high performance even under adverse network conditions. Comparative analysis with existing models such as ECC-based Hybrid Encryption and Blockchain Krill Herd Optimization confirms significant improvements in accuracy, efficiency, and scalability. The findings validate DFA-AOKGE’s potential to serve as a resilient and practical digital forensic solution, offering secure, scalable, and real-time capabilities for modern forensic investigations across decentralized infrastructures.