Novel Homomorphic Encryption for Mitigating Impersonation Attack in Fog Computing

Abstract

Fog computing is a rapidly growing technology that aids in pipelining the possibility of mitigating breaches between the cloud and edge servers. It facilitates the benefits of the network edge with the maximized probability of offering interaction with the cloud. However, the fog computing characteristics are susceptible to counteract the challenges of security. The issues present with the Physical Layer Security (PLS) aspect in fog computing which included authentication, integrity, and confidentiality has been considered as a reason for the potential issues leading to the security breaches. In this work, the Octonion Algebra-inspired Non- Commutative Ring-based Fully Homomorphic Encryption Scheme (NCR-FHE) was proposed as a secrecy improvement technique to overcome the impersonation attack in cloud computing. The proposed approach was derived through the benefits of Octonion algebra to facilitate the maximum security for big data-based applications. The major issues in the physical layer security which may potentially lead to the possible security issues were identified. The potential issues causing the impersonation attack in the Fog computing environment were identified. The proposed approach was compared with the existing encryption approaches and claimed as a robust approach to identify the impersonation attack for the fog and edge network. The computation cost of the proposed NCR-FHE is identified to be significantly reduced by 7.18%, 8.64%, 9.42%, and 10.36% in terms of communication overhead for varying packet sizes, when compared to the benchmarked ECDH-DH, LHPPS, BF-PHE and SHE-PABF schemes.