دانلود کتاب امنیت سایبری در وسایل نقلیه رباتیک خودران – کاربردهای یادگیری ماشین برای تشخیص حملات سایبری

Technology has developed at a very fast rate in the recent past, and this has greatly impacted society in aspects such as communication and transport. One of the most innovative technologies is the Internet of Vehicles (IoV), which can change the world of transportation and build a system of autonomous and intelligent vehicles connected to an interconnected network. However, with the increased connectivity arises a new cybersecurity challenge that stems from the fact that vehicles are now susceptible to cyber threats that can threaten their safety, reliability, and efficiency. These challenges form the core motivation for this book: the urgent need to develop robust cybersecurity solutions for autonomous vehicles (AVs) operating within the IoV ecosystem. Some of the most critical bus systems used in modern vehicles are the controller area network (CAN). CAN bus is the network which controls the flow of information between the electronic control units (ECUs) within the vehicle. Although it is a crucial component in managing the overall functioning of AVs, the CAN bus was not developed with security as a consideration. Its lack of security mechanisms exposes vehicles to serious vulnerabilities, including message injection attacks. In such attacks, hackers can inject malicious commands into the CAN bus, which leads the ECUs to perform certain actions that they were not supposed to, which may lead to malfunctions, accidents, or even fatalities. CAN bus is its reliance on message prioritisation based on CAN IDs. Lower CAN ID values are prioritised for transmission, which presents an opportunity for attackers to manipulate high-priority messages and disrupt the normal functioning of the vehicle. This calls for new strategies to address this threat that cannot be solved by conventional security measures. In response to these challenges, this book introduces a novel intrusion detection system (IDS) specifically designed for AVs, which leverages data prioritisation in CAN IDs to enhance threat detection and mitigation. To be more precise, the IDS analyses the sequence of CAN bus messages and filters the data flow, which increases the efficiency of anomaly and cyber threat detection. We have incorporated machine learning and deep learning frameworks into the IDS in order to analyse CAN bus traffic and identify the presence of any malicious activities in real time with a high level of accuracy. By employing classifiers like support vector machines (SVM), k-nearest neighbours (KNN), and neural networks, the system has been able to maintain an accuracy rate of more than 99% in identifying between the genuine data and the possible threats with minimal false alarms. This book provides a comprehensive examination of the cybersecurity risks faced by AVs, with a particular emphasis on CAN vulnerabilities and the innovative use of data prioritisation within CAN IDs. Through filling the research gap between the theoretical studies and practical implementations, it offers a new method for improving vehicle security. This approach is not only beneficial for researchers but also offers industry professionals helpful meth¬ods that can be used to protect actual AV systems. Readers will find this book invaluable as both a reference guide and a practical manual for improving the cybersecurity of autonomous vehicles. Researchers can use the models and frameworks presented here to further advance the field of vehicular cybersecu¬rity, exploring new directions in data prioritisation and attack detection. These solutions can be used by industry professionals and engineers to improve the safety of current and future autonomous vehicle networks, making the IDS system flexible for use in various aspects of the IoV.