Fault Prediction and Recovery Using Machine Learning Techniques and the HTM Algorithm in Vehicular Network Environment

• Published in: IEEE open journal of intelligent transportation systems Vol. 5; pp. 132 - 145
• Main Authors: Zidi, Salah, Alaya, Bechir, Moulahi, Tarek, Al-Shargabi, Amal, Khediri, Salim El
• Format: Journal Article
• Language: English
• Published: New York IEEE 2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; classification; Classification algorithms; Communications traffic; Fault detection; Fault diagnosis; fault prediction; fault recovery; Hidden Markov models; hierarchical temporal memory (HTM); Internet of Vehicles; Internet of Vehicles (IoV); Machine learning; Machine learning algorithms; Mobile ad hoc networks; Network latency; Recovery; Sensors; Vehicles; Vehicular ad hoc networks; Vehicular network; Wireless sensor networks
• ISSN: 2687-7813; 2687-7813
• DOI: 10.1109/OJITS.2023.3347484

The amount of data available to vehicles has become very large in the vehicular networks' environment. Failures that mislead real-time data from vehicle sensors and other devices have become massive, and the need for automated techniques that can analyze data to detect malicious sources has become paramount. The application of machine learning techniques in the environment of vehicular ad hoc networks (VANET) is very promising and is beginning to show results in terms of applications designed and articles published. These techniques are increasingly accessible and used intensively, as many researchers are working to detect anomalous data. However, there is no universal, effective technique so far that can detect all abnormal data and then recover it. This work is an effort in that direction. We propose a smart model that uses multiple machine-learning classification methods. Our contribution also relates to a study of the attributes of interest for the algorithm used during the detection phase, namely the hierarchical temporal memory algorithm (HTM). The packets exchanged by the vehicle are grouped in instant description windows. These windows are then analyzed to extract a set of attributes. These are linked to the properties of network traffic such as flow or latency. They are subject to the process of detecting anomalies and intrusions carried out thanks to the algorithm with HTM. We propose the performance of fault detection and recovery at the level of the fog layer. The obtained simulation results demonstrate the efficiency of the learning methods and HTM for the detection of defects and errors in the IoV.