Dynamic hierarchical intrusion detection system for internet of vehicle on edge computing platform

Abstract

Abstract In recent days, the Internet of Vehicles (IoV) and its network of connected automobiles have revealed several new security risks. Classical intrusion detection systems face challenges in identifying intrusions due to the growing number of vehicles, the dynamic nature of IoV, and limited resources. A hierarchical clustering method allows dividing the IoV network into clusters. The elements that determine the outcome are the geographical proximity and the traffic density. It is called the Dynamic Hierarchical Intrusion Detection Framework (DHIDF) for the IoV. To protect infrastructure and passengers, an IoV‐specific DHIDF using edge computing has been proposed. Because of this, anomaly detection and localised assessment of danger will become less required. The application of DHIDF on a large scale inside the ecosystem of IoV is not entirely out of the question. The term encompasses several subfields, including intelligent transportation networks (ITNs), smart city infrastructure, fleet management, transportation, and autonomous vehicle systems. The efficacy of DHIDF is assessed through simulations that replicate current and potential future threats, including those related to the Internet of Things. Analysis of key performance parameters, including response time, detection accuracy, asset utilization, and scalability, has been conducted to assess the system's feasibility and durability.

Keywords

• internet of things
• internet of vehicles
• intrusion detection
• learning (artificial intelligence)
• security of data