Deep Contractive Autoencoder-Based Anomaly Detection for In-Vehicle Controller Area Network (CAN)

Lokman, Siti Farhana; Othman, Abu Talib; Musa, Shahrulniza; Bakar, Muhamad Husaini Abu

doi:10.1007/978-3-030-28505-0_16

Cited by 17 publications

(8 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lokman et al [ 26 ] developed an IDS for an in-vehicle network using an unsupervised DL-based model, known as Deep Contractive Auto-encoders (DCAEs). The DCAE model outperformed other regularized auto-encoder variants, with a 91.0% detection rate.…”

Section: Background and Related Workmentioning

confidence: 99%

Deep Transfer Learning Based Intrusion Detection System for Electric Vehicular Networks

Mehedi

Anwar

Rahman

et al. 2021

Sensors

View full text Add to dashboard Cite

The Controller Area Network (CAN) bus works as an important protocol in the real-time In-Vehicle Network (IVN) systems for its simple, suitable, and robust architecture. The risk of IVN devices has still been insecure and vulnerable due to the complex data-intensive architectures which greatly increase the accessibility to unauthorized networks and the possibility of various types of cyberattacks. Therefore, the detection of cyberattacks in IVN devices has become a growing interest. With the rapid development of IVNs and evolving threat types, the traditional machine learning-based IDS has to update to cope with the security requirements of the current environment. Nowadays, the progression of deep learning, deep transfer learning, and its impactful outcome in several areas has guided as an effective solution for network intrusion detection. This manuscript proposes a deep transfer learning-based IDS model for IVN along with improved performance in comparison to several other existing models. The unique contributions include effective attribute selection which is best suited to identify malicious CAN messages and accurately detect the normal and abnormal activities, designing a deep transfer learning-based LeNet model, and evaluating considering real-world data. To this end, an extensive experimental performance evaluation has been conducted. The architecture along with empirical analyses shows that the proposed IDS greatly improves the detection accuracy over the mainstream machine learning, deep learning, and benchmark deep transfer learning models and has demonstrated better performance for real-time IVN security.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

Deep Transfer Learning Based Intrusion Detection System for Electric Vehicular Networks

Mehedi

Anwar

Rahman

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…However, under any circumstances, the blockchain network will be built on top of a protocol that determines how the system works, so all elements of the system and network participants will have to follow the basic protocol rules. If the protocol stipulates what the rules are, the algorithm complies with these rules and instructs the system to go through the procedures to derive the desired results [56,57]. For example, the blockchain consensus algorithm determines the validity of transactions and blocks.…”

Section: Blockchain Consensus Algorithmmentioning

confidence: 99%

Automotive Vulnerability Analysis for Deep Learning Blockchain Consensus Algorithm

Kim

2021

Electronics

View full text Add to dashboard Cite

In this study, future cars are attempting self-driving around the world. However, hacking, such as ECUs in automobiles, creates problems that are directly connected to human life. Therefore, this study wrote a paper that detects anomalies in such cars by field. As a related study, the study investigated the vulnerabilities of the automobile security committee and automobile security standards and investigated the detection of abnormalities in the hacking of geo-train cars using artificial intelligence’s LSTM and blockchain consensus algorithm. In addition, in automobile security, an algorithm was studied to predict normal and abnormal values using LSTM-based anomaly detection techniques on the premise that automobile communication networks are largely divided into internal and external networks. In the methodology, LSTM’s pure propagation malicious code detection technique was used, and it worked with an artificial intelligence consensus algorithm to increase security. In addition, Unity ML conducted an experiment by constructing a virtual environment using the Beta version. The LSTM blockchain consensus node network was composed of 50,000 processes to compare performance. For the first time, 100 Grouped Tx, 500 Channels were tested for performance. For the first time, the malicious code detection rate of the existing system was verified. Accelerator, Multichannel, Sharding, Raiden, Plasma, and Trubit values were verified, and values of approximately 15,000 to 50,000 were obtained. In this paper, we studied to become a paper of great significance on hacking that threatens human life with the development of self-driving cars in the future.

show abstract

“…Machine learning based methods imply the usage of artificial neural networks, clustering and supervised models for classification and regression. In the specific field of CAN bus intrusion detection, popular machine learning approaches include autoencoders (Lokman et al, 2019;Lin et al, 2021;Novikova et al, 2020), recurrent neural networks (RNN) such as Long Short-Term Memory (LSTM) networks (Taylor et al, 2016;Negi et al, 2019;Khan et al, 2020;Hanselmann et al, 2020;Hossain et al, 2020), Gated Recurrent Unit (GRU)based networks (Kukkala et al, 2020), replicator neural networks (Weber et al, 2018), and deep convolutional networks (Song et al, 2020). The scrutinized literature shows that recurrent architectures are often the preferred choice for modeling the time series of CAN bus signals, whilst convolutional networks are used when data is transformed to a two-dimensional grid dataframe to resemble an image format (Song et al, 2020).…”

Section: Related Workmentioning

confidence: 99%

Detecting message modification attacks on the CAN bus with Temporal Convolutional Networks

Chiscop,

Gazdag,

Bosman

et al. 2021

Preprint

View full text Add to dashboard Cite

Multiple attacks have shown that in-vehicle networks have vulnerabilities which can be exploited. Securing the Controller Area Network (CAN) for modern vehicles has become a necessary task for car manufacturers. Some attacks inject potentially large amount of fake messages into the CAN network; however, such attacks are relatively easy to detect. In more sophisticated attacks, the original messages are modified, making the detection a more complex problem. In this paper, we present a novel machine learning based intrusion detection method for CAN networks. We focus on detecting message modification attacks, which do not change the timing patterns of communications. Our proposed temporal convolutional network-based solution can learn the normal behavior of CAN signals and differentiate them from malicious ones. The method is evaluated on multiple CAN-bus message IDs from two public datasets including different types of attacks. Performance results show that our lightweight approach compares favorably to the state-of-the-art unsupervised learning approach, achieving similar or better accuracy for a wide range of scenarios with a significantly lower false positive rate. a https://orcid.

show abstract

Deep Contractive Autoencoder-Based Anomaly Detection for In-Vehicle Controller Area Network (CAN)

Cited by 17 publications

References 8 publications

Deep Transfer Learning Based Intrusion Detection System for Electric Vehicular Networks

Deep Transfer Learning Based Intrusion Detection System for Electric Vehicular Networks

Automotive Vulnerability Analysis for Deep Learning Blockchain Consensus Algorithm

Detecting message modification attacks on the CAN bus with Temporal Convolutional Networks

Contact Info

Product

Resources

About