Blockchain Integration with Machine Learning for Securing Fog Computing Vulnerability in Smart City Sustainability
Lukman Adewale Ajao, Simon T. Apeh
Abstract
The advent of a smart city-based industrial Internet of Things (IIoT) is confidently built on the combined protocols of a virtual IPv6 addressing scheme and the fifth generation (5G) mobile network. For better network service and to achieve Quality of Experience (QoE) in the architecture. But this intelligent city architecture is vulnerable to several cyber-attack and malicious actors at the different layers which make it exposed to the same attacks as in the conventional IPv4 wireless sensor networks. However, this work aims to develop a blockchain-based machine learning (BML) security framework that secures the fog computing layer vulnerability in the smart city’s sustainability. The machine learning approach is firstly implemented between the edge layer and fog server nodes of the city architecture for the variants of intrusion detection using different ML algorithms for the attack’s discovery and classification. While the augmented blockchain technology is implemented between the fog layer and cloud computing to enhance the privacy and confidentiality of packet traffic broadcast to the public. The results obtained from ML-IDS show high-performance detection accuracy and low processing time. While the blockchain framework is also evaluated based on the certmcate generation, and retrieval size in bytes and time in milliseconds.