The air traffic control (ATC) network’s airspace sector is a crucial component of air traffic management. The increasing demand for air transportation services has made limited airspace a significant challenge to sustainable and efficient air transport operations. To address the issue of traffic congestion and flight delays, improving the operational efficiency of ATC has been identified as a key strategy. A clear understanding of the characteristics of airspace sectors, which are the building blocks of ATC, is essential for optimizing air traffic management. In this research, a novel approach using complex network theory was applied to examine the features and invulnerability of the airspace sector network. We developed a model of the airspace sector network by treating air traffic control sectors as network nodes and the flow of air traffic between these sectors as edges. Network characteristics were analyzed using several metrics including degree, intensity, average path length, betweenness centrality, and clustering coefficient. The static invulnerability of the airspace sector network was evaluated through simulation, and the network efficiency and the size of the connected component were used to assess its invulnerability. A study was conducted in North China based on the ATC sector network. The findings of the study revealed that the sector network did not exhibit the traits of a small-world network model, characterized by short average path lengths and high clustering coefficients. The evaluation of network invulnerability showed that the network’s invulnerability varied depending on the attack strategy used. It was discovered that attacking sectors with high betweenness resulted in the most significant harm to network invulnerability, and betweenness centrality was considered to be a useful indicator for identifying critical sectors that require optimization.

The rise of end-to-end (E2E) speech recognition technology in recent years has overturned the design pattern of cascading multiple subtasks in classical speech recognition and achieved direct mapping of speech input signals to text labels. In this study, a new E2E framework, ResNet–GAU–CTC, is proposed to implement Mandarin speech recognition for air traffic control (ATC). A deep residual network (ResNet) utilizes the translation invariance and local correlation of a convolutional neural network (CNN) to extract the time-frequency domain information of speech signals. A gated attention unit (GAU) utilizes a gated single-head attention mechanism to better capture the long-range dependencies of sequences, thus attaining a larger receptive field and contextual information, as well as a faster training convergence rate. The connectionist temporal classification (CTC) criterion eliminates the need for forced frame-level alignments. To address the problems of scarce data resources and unique pronunciation norms and contexts in the ATC field, transfer learning and data augmentation techniques were applied to enhance the robustness of the network and improve the generalization ability of the model. The character error rate (CER) of our model was 11.1% on the expanded Aishell corpus, and it decreased to 8.0% on the ATC corpus.