Blockchain Based Machine Learning Model for Secure Data Transfer and Route Preservation in UAV Integrated VANET Systems

The rise of driver assistance and automotive telecommunication systems shows great potential for adaptive transport solutions using vehicular ad hoc networks (VANET). Generally, the two main issues in vehicle ad hoc networks that malicious attackers can greatly affect are privacy and safety. Preventing the spread of harmful messages among vehicles is crucial to protecting the private properties of automobiles from potential threats. This research tackles these issues and proposes a new machine-learning-based message authentication method. This method can be integrated with interplanetary file systems and blockchain to ensure secure message distribution. The Inter Planetary File System (IPFS) is utilized by blockchain technology to create tamper-proof records in a distributed environment. This protocol stores events using content addressing. The source metadata from the IPFS is first stored in a smart contract and then in the distributed ledger technology. This framework makes use of the Iterative Import Vector Machine (IIVM) classifier and Non-overlapped K-means clustering in the event authentication process. It will be classified as malicious or not malicious in order to carry out the vehicle clustering. After clustering, the IIVM classifier works to identify harmful event messages. As a result, dropped messages are recognized as such and the secure messages are sent into the network. According to simulation results, the suggested approach increases event spoofing identification precision by 96.21%. This system's trust model of the occurrence does an excellent task of separating genuine instances from fake ones.