The field of cybersecurity is moving towards developing more effective intrusion detection systems (IDS) for Internet of Things (IoT) and vehicle networks. Researchers are focusing on addressing the challenges of highly imbalanced data, which can lead to missed threats and decreased detection accuracy. Hybrid sampling techniques, machine learning models, and deep learning approaches are being explored to improve detection accuracy and mitigate class imbalance. Additionally, there is a growing interest in developing topology-aware detection methods and hardware-in-the-loop (HIL) frameworks for validating network security approaches. Noteworthy papers include:
- A paper on CSAGC-IDS, a dual-module deep learning network intrusion detection model, which achieves high accuracy and F1-score in detecting complex and imbalanced data.
- A paper on SecCAN, an extended CAN controller with embedded intrusion detection, which shows promising results in detecting attacks with state-of-the-art accuracy and zero software overheads.
- A paper on FAV-NSS, an HIL framework for accelerating validation of automotive network security strategies, which demonstrates a significant reduction in latency compared to traditional coupled accelerators.