A deep learning-based IoT-oriented infrastructure for secure smart City

Abstract

In recent years, the Internet of Things (IoT) infrastructures are developing in various industrial applications in sustainable smart cities and societies such as smart manufacturing, smart industries. The Cyber-Physical System (CPS) is also part of IoT-oriented infrastructure. CPS has gained considerable success in industrial applications and critical infrastructure with a distributed environment. This system aims to integrate the physical world to computational facilities as cyberspace. However, there are many challenges, such as security and privacy, centralization, communication latency, scalability in such an environment. To mitigate these challenges, we propose a Deep Learning-based IoT-oriented infrastructure for a secure smart city where Blockchain provides a distributed environment at the communication phase of CPS, and Software-Defined Networking (SDN) establishes the protocols for data forwarding in the network. A deep learning-based cloud is utilized at the application layer of the proposed infrastructure to resolve communication latency and centralization, scalability. It enables cost-effective, high-performance computing resources for smart city applications such as the smart industry, smart transportation. Finally, we evaluated the performance of our proposed infrastructure. We compared it with existing methods using quantitative analysis and security and privacy analysis with different measures such as scalability and latency. The evaluation of our implementation results shows that performance is improved.