Driven by the new digital revolution represented by the Internet of Things (IoT) and cyber-physical system (CPS) technologies, the traditional Internet architecture dominated by a static backbone structure is facing the challenge of emerging network application scenarios such as heterogeneous vehicle-to-everything (V2X) communications and delay-sensitive industrial control system (ICS). Numerous resource-constrained devices (e.g., intelligent connected cars with high moving speed, industrial wireless sensors scattered in harsh production environments) are gradually becoming first-class network entities, rather than personal computers and cloud servers equipped with plenty of computing resources.

Intrusion detection technology is one of the most famous security protection technologies in the traditional Internet area. However, with the advent of a new era of digital society featuring the Internet of everything, it becomes very difficult for mainstream intrusion detection technologies to exert their usual effectiveness as done before. On the one hand, network security boundaries become more blurred and complex, which contradicts the conventional assumption of the boundary-based network security strategy, although it has been used for decades as an established rule. On the other hand, these emerging resource-constrained networking entities generally have a limited computing capacity or inadequate power supply to achieve an acceptable detection performance.

This Special Issue will focus on cutting-edge original research and review articles from academia and solicit implementations from industry, with a particular emphasis on intelligent and lightweight intrusion detection technologies that can directly apply to resource-constrained networking scenarios to achieve a more secure digital society.

Potential topics include but are not limited to the following:

• New opportunities for typical classification techniques in intrusion detection
• Novel intrusion detection architecture with emerging learning models
• Lightweight neural network designs for lightweight intrusion detection
• Accurate feature selection techniques for lightweight intrusion detection
• Efficient intrusion detection technologies with limited labeled datasets
• Intrusion detection for industrial process control with strict sequential logic
• Practical intrusion detection techniques for resource-constrained CPS devices
• Learning-based intrusion detection for delay-sensitive ICS scenarios
• Learning-based intrusion detection for in-vehicle networks
• Learning-based intrusion detection for V2X communications
• New use cases for intrusion detection in ICS or V2X or beyond
• Other methods of security and privacy for securing promising network application scenarios