In-vehicle communication cyber security: A comprehensive review of challenges and solutions

The utilization of autonomous vehicles is experiencing a rapid proliferation in contemporary society. Concurrently, with the relentless evolution of technology, the inexorable integration of autonomous vehicles into urban environments, driven by the overarching paradigm of smart cities, becomes increasingly apparent. This escalating reliance on autonomous vehicles concurrently heightens the susceptibility to malevolent actors orchestrating cyber-attacks against these vehicular systems. While previous years have seen a limited corpus of academic research pertaining to cyber-attack and defense methodologies for autonomous vehicles, the relentless progression of technology mandates a more contemporary and exhaustive inquiry. In addition, to the best of our knowledge, there is no article in the literature that provides detailed information and comparisons about in-vehicle sensors, in-vehicle networks, and in-vehicle network datasets by combining them in one article. Also, to our knowledge, very limited studies have been conducted on separately comparative analysis of in-vehicle networks, in-vehicle sensors or data sets in 2024, and therefore, the necessity of conducting a review study on these topics was recognized. To address this deficiency, we compile articles on attacks and defenses on sensors, in-vehicle networks and present detailed information about the latest datasets and provide comparative analysis. In this paper, we have analyzed 108 papers from the last 10 years on in-vehicle networks and sensors. 38 articles on in-vehicle sensors and 70 articles on in-vehicle networks were reviewed and analyzed. We categorize in-vehicle communication attacks into two main groups: sensor-initiated and network-initiated, with a chronological classification to highlight their evolution. We also compare the progress in securing in-vehicle communication and evaluate the most widely used datasets for attack and protection methods. Additionally, we discuss the advantages and disadvantages of these datasets and suggest future research directions. To the best of our knowledge, this work is the first to offer detailed information and comparative analysis of in-vehicle networks, sensors, and the latest datasets. While the study highlights the significant research conducted to protect in-vehicle networks and sensors from cyber attacks, technological advancements continue to introduce new attack vectors. Cars remain particularly susceptible to threats such as DoS, Fuzzy, Spoofing, and Replay attacks. Moreover, current defense mechanisms, including LSTM and CNN, have notable limitations. Future research is needed to address these challenges and enhance vehicle cybersecurity.