Vanet FDIA Solutions using Blockchain Based IPFS-Trust Management System with ML SVR Model

Abstract

The Internet of Vehicles (IoV) is the next phase in the evolution of vehicular ad hoc networks (VANETs).Multiple types of Smart Networks exists in our surrounding.i.e., Wireless Sensor Networks (WSNs), Crowd Sensing Networks (CSNs), and Internet of Vehicles, etc A VANET is a collection of mobile nodes (vehicles) that share data through ad hoc on-demand connections. Vehicle Tracking is one of the uses of IOV(Internet of Vehicles) and Vehicle Security is one of the major issues for all vehicle owners. On a vehicle, there are various on-board sensors that sense a vehicle’s motion and the surrounding environment. On-board sensors can also warn drivers about approaching vehicles, speeding, and slippery road conditions. The main aim of the paper is to provide solutions for False Data Injection Attack by Integration of Blockchain Based IPFS-Trust Management System with ML SVR Regression Model. Due to Network Assaults and Threats under Vanet System, the safety of the drivers is under stake and Critical. A rogue node can send out erroneous messages, causing unavoidable scenarios. We first filter the received data from Vehicles creating false traffic jam warning messages using the Machine learning SVR Regression Model where data is created and split into train and test data. We used Machine learning supervised algorithm to find whether the vehicle is a legitimate vehicle or an attacker vehicle and the result is validated using the parameters like Accuracy, Loss Rate, Precision, Recall, and F-Test Score. Algorithm Implementation results show that the FDIA attack strategy achieves a better performance than the without using ML algorithm of SVR Regression Model based attack strategy in Predicting the Vanet Security. Also, we studied the various ways to mitigate the impact of false data injection into the network through a compromised node. Users can access the system through DApp, an Ethereum-distributed application, and manage their vehicle data.