Multi-objectives reinforcement federated learning blockchain enabled Internet of things and Fog-Cloud infrastructure for transport data

Mazin Abed Mohammed; Abdullah Lakhan; Karrar Hameed Abdulkareem; Mohd Khanapi Abd Ghani; Haydar Abdulameer Marhoon; Jan Nedoma; Radek Martinek

doi:10.1016/j.heliyon.2023.e21639

Multi-objectives reinforcement federated learning blockchain enabled Internet of things and Fog-Cloud infrastructure for transport data

Heliyon. 2023 Nov 2;9(11):e21639. doi: 10.1016/j.heliyon.2023.e21639. eCollection 2023 Nov.

Authors

Mazin Abed Mohammed^{1

2

3}, Abdullah Lakhan^{4

2

3}, Karrar Hameed Abdulkareem⁵, Mohd Khanapi Abd Ghani⁶, Haydar Abdulameer Marhoon^{7

8}, Jan Nedoma², Radek Martinek³

Affiliations

¹ Department of Artificial Intelligence, College of Computer Science and Information Technology, University of Anbar, Anbar, 31001, Iraq.
² Department of Telecommunications, VSB-Technical University of Ostrava, 70800 Ostrava, Czech Republic.
³ Department of Cybernetics and Biomedical Engineering, VSB-Technical University of Ostrava, 70800 Ostrava, Czech Republic.
⁴ Department of Cybersecurity and Computer Science, Dawood University of Engineering and Technology, Karachi City 74800, Sindh, Pakistan.
⁵ College of Agriculture, Al-Muthanna University, Samawah 66001, Iraq.
⁶ Department of Software Engineering, Faculty of Information and Communication Technology, Universiti Teknikal Malaysia Melaka (UTeM), Malaysia.
⁷ Information and Communication Technology Research Group, Scientific Research Center, Al-Ayen University, Thi-Qar, Iraq.
⁸ College of Computer Sciences and Information Technology, University of Kerbala, Karbala, Iraq.

Abstract

For the past decade, there has been a significant increase in customer usage of public transport applications in smart cities. These applications rely on various services, such as communication and computation, provided by additional nodes within the smart city environment. However, these services are delivered by a diverse range of cloud computing-based servers that are widely spread and heterogeneous, leading to cybersecurity becoming a crucial challenge among these servers. Numerous machine-learning approaches have been proposed in the literature to address the cybersecurity challenges in heterogeneous transport applications within smart cities. However, the centralized security and scheduling strategies suggested so far have yet to produce optimal results for transport applications. This work aims to present a secure decentralized infrastructure for transporting data in fog cloud networks. This paper introduces Multi-Objectives Reinforcement Federated Learning Blockchain (MORFLB) for Transport Infrastructure. MORFLB aims to minimize processing and transfer delays while maximizing long-term rewards by identifying known and unknown attacks on remote sensing data in-vehicle applications. MORFLB incorporates multi-agent policies, proof-of-work hashing validation, and decentralized deep neural network training to achieve minimal processing and transfer delays. It comprises vehicle applications, decentralized fog, and cloud nodes based on blockchain reinforcement federated learning, which improves rewards through trial and error. The study formulates a combinatorial problem that minimizes and maximizes various factors for vehicle applications. The experimental results demonstrate that MORFLB effectively reduces processing and transfer delays while maximizing rewards compared to existing studies. It provides a promising solution to address the cybersecurity challenges in intelligent transport applications within smart cities. In conclusion, this paper presents MORFLB, a combination of different schemes that ensure the execution of transport data under their constraints and achieve optimal results with the suggested decentralized infrastructure based on blockchain technology.

Keywords: Agents; Blockchain; Cloud; MORFLB; Self-autonomous vehicle; Training.