With the development of the Air Traffic Management System (ATM), the ATM network has transitioned from a closed physical isolation network to an open Cyber-Physical System (CPS). As a result, the surface of cyber-attacks is constantly expanding, and the degree of automation and correlation of attacks is constantly increasing. Traditional penetration testing attack path planning relies on the expertise of specific field experts, which can be time-consuming and resource-intensive when dealing with large networks. A large number of useful cyber security information are seriously fragmented and not integrated. In this paper, we aim to address these challenges by constructing a novel ATM system cyber threat knowledge graph called ATMCyKG. This knowledge graph is based on ATT&CK attack TTP style templates, including attack tactics, techniques, and processes. It defines entities and attributes, and their relationships. By combining ATT&CK tactics and techniques with the knowledge graph, we propose an attack path planning method based on ATMCyKG and integrate it into a reinforcement learning model. We use a variety of reinforcement learning algorithms to conduct comparative experiments and conduct a panoramic analysis of its attack process. Finally, the experimental results of the three reinforcement algorithms are analyzed and summarized. This paper introduces the ATMCyKG for the first time and utilizes Neo4j for its construction. From the perspective of an attacker, we utilized reinforcement learning to discover vulnerability sequences by selecting effective action sequences to achieve the desired target. By planning automatic attack paths, we can eliminate the reliance on expert experience, save manpower and time, and improve the operability and testing efficiency of automated penetration testing in ATM. This is of significant importance in ensuring aviation transportation safety and maintaining airspace order.

Deep learning has shown significant advantages in Automatic Dependent Surveillance-Broadcast (ADS-B) anomaly detection, but it is known for its susceptibility to adversarial examples which make anomaly detection models non-robust. In this study, we propose Time Neighborhood Accumulation Iteration Fast Gradient Sign Method (TNAI-FGSM) adversarial attacks which fully take into account the temporal correlation of an ADS-B time series, stabilize the update directions of adversarial samples, and escape from poor local optimum during the process of iterating. The experimental results show that TNAI-FGSM adversarial attacks can successfully attack ADS-B anomaly detection models and improve the transferability of ADS-B adversarial examples. Moreover, the TNAI-FGSM is superior to two well-known adversarial attacks called the Fast Gradient Sign Method (FGSM) and Basic Iterative Method (BIM). To the best of our understanding, we demonstrate, for the first time, the vulnerability of deep-learning-based ADS-B time series unsupervised anomaly detection models to adversarial examples, which is a crucial step in safety-critical and cost-critical Air Traffic Management (ATM).

The continuous advancement of fire protection technologies has necessitated the development of comprehensive safety standards, leading to an increasingly diversified and specialized regulatory landscape. This has made it difficult for fire protection professionals to quickly and accurately locate the required fire safety standard information. In addition, the lack of effective integration and knowledge organization concerning fire safety standard entities has led to the severe fragmentation of fire safety standard information and the absence of a comprehensive “one map”. To address this challenge, we introduce FT-FLAT, an innovative CNN–Transformer fusion architecture designed specifically for fire safety standard entity extraction. Unlike traditional methods that rely on rules or single-modality deep learning, our approach integrates TextCNN for local feature extraction and combines it with the Flat-Lattice Transformer for global dependency modeling. The key innovations include the following. (1) Relative Position Embedding (RPE) dynamically encodes the positional relationships between spans in fire safety texts, addressing the limitations of absolute positional encoding in hierarchical structures. (2) The Multi-Branch Prediction Head (MBPH) aggregates the outputs of TextCNN and the Transformer using Einstein summation, enhancing the feature learning capabilities and improving the robustness for domain-specific terminology. (3) Experiments conducted on the newly annotated Fire Safety Standard Entity Recognition Dataset (FSSERD) demonstrate state-of-the-art performance (94.24% accuracy, 83.20% precision). This work provides a scalable solution for constructing fire safety knowledge graphs and supports intelligent information retrieval in emergency situations.