This paper investigates the single agile optical satellite scheduling problem, which has received increasing attention due to the rapid growth in earth observation requirements. Owing to the complicated constraints and considerable solution space of this problem, the conventional exact methods and heuristic methods, which are sensitive to the problem scale, demand high computational expenses. Thus, an efficient approach is demanded to solve this problem, and this paper proposes a deep reinforcement learning algorithm with a local attention mechanism. A mathematical model is first established to describe this problem, which considers a series of complex constraints and takes the profit ratio of completed tasks as the optimization objective. Then, a neural network framework with an encoder–decoder structure is adopted to generate high-quality solutions, and a local attention mechanism is designed to improve the generation of solutions. In addition, an adaptive learning rate strategy is proposed to guide the actor–critic training algorithm to dynamically adjust the learning rate in the training process to enhance the training effectiveness of the proposed network. Finally, extensive experiments verify that the proposed algorithm outperforms the comparison algorithms in terms of solution quality, generalization performance, and computation efficiency.

As the core infrastructure for providing all-weather, full-coverage, high-speed, and diversified information services, spatial information networks (SINs) possess significant social, economic, and military value. However, due to the inherent characteristics of their network architecture, SINs are susceptible to core service paralysis and functional failure under large-scale targeted attacks or random disturbances, posing a critical bottleneck that constrains their stable operation. Current research on SIN vulnerability is predominantly confined to a single network topology perspective, lacking an integrated consideration of the task execution perspective. Consequently, it fails to accommodate the dual requirements of “network topology stability” and “task execution effectiveness”. To address the aforementioned research needs and challenges, this study adopts a “topology-task” dual-perspective fusion approach and proposes a vulnerability analysis framework for SINs that integrates multi-layer networks and hypernetworks. First, a two-layer SIN topology model encompassing the user layer and the satellite layer is constructed. Leveraging hypernetwork theory, information tasks involving multiple network entities are formally defined, and an integrated multi-layer hypernetwork model is established. Second, based on distinct task types, three categories of task efficiency evaluation metrics are defined, and corresponding quantitative methods for calculating SIN vulnerability are derived. Third, during the vulnerability analysis phase, a novel strategy for identifying and removing overlapping nodes in hypernetworks is introduced to enable precise localization of critical nodes within the network. Concurrently, a pre-attack node hardening strategy is designed to minimize the impact of attacks on network performance. Finally, through systematic analysis of vulnerability performance and critical node characteristics under different node removal strategies, the results demonstrate enhanced network performance. The effectiveness of the proposed method is validated by comparing the defense performance of the hardening strategy across various attack scenarios. To verify the feasibility and superiority of the proposed method, this study designs 5 × 5 groups of simulation experiments with varying network parameters. The results indicate that, compared with traditional methods, the proposed strategy can more accurately identify core nodes affecting the stable operation of SINs, significantly reducing network vulnerability and improving network survivability. In addition, a comprehensive sensitivity analysis of SIN vulnerability is conducted from three key influencing dimensions—mission scale, satellite count, and constellation configuration—clarifying the impact of each dimension on network invulnerability. Thus, this paper provides a reliable theoretical foundation and technical support for the planning, design, optimal deployment, and operation and maintenance management of SINs.

With the rapid growth in satellite types and numbers in space information networks, accurate and fast inter-satellite link prediction has become a core requirement for topology modeling and capability evaluation. However, the current space information networks are characterized by large scales and the coexistence of multi-orbit satellites, posing dual challenges to inter-satellite link prediction. Link state prediction demands higher accuracy with limited computing power, while diverse satellite communication antenna loads require stronger generalization to adapt to different scenarios. To address these issues, this paper proposes a fully connected neural network model based on Bayesian optimization. By introducing a weighted loss function, the model effectively handles data imbalance in the link states. Combined with Bayesian optimization, the neural network hyperparameters and weighted loss function coefficients are fine-tuned, significantly improving the prediction accuracy and scene adaptability. Experimental results show that the BO-FCNN model exhibited an excellent performance on the test dataset, with an F1 score of 0.91 and an average accuracy of 93%. In addition, validation with actual satellite data from CelesTrak confirms the model’s real-world performance and its potential as a reliable solution for inter-satellite link prediction.

Synthetic aperture radars (SARs), characterized by their excellent performance, have found wide use in military applications. SAR-based image recognition is an important research area. Neural networks (NNs) have been widely used for image recognition and have demonstrated high recognition accuracy. However, NNs require a significant amount of data for training. In situations with limited training data, it is difficult to obtain satisfactory recognition accuracy using NNs. One approach toward alleviating the problems associated with small training sets is data augmentation. Here, we propose a novel SAR-based image augmentation method: using CycleGAN, aerial images are transformed into SARformat images, thereby allowing expansion of the SAR dataset. To improve the quality of these transformed images, we designed two filters: 1) a pre-filter for selecting images that are suitable for the transformation and 2) a post-filter for selecting suitable images after the transformation; this allows approximating the transformed images as real SAR images. We validated our method on the SSDD dataset. Our experimental results show that compared with the traditional data augmentation method, our method exhibits better recognition performance. With only 300 transformed images, the mean average precision improved from 66% to 76%, using only a small fraction of the SSDD dataset, thus overcoming the problem of small-sized training sets.

Due to the widespread deployment of large-scale satellite constellations, Spatial Information Networks (SINs) exhibit increasingly high-order node and service coupling characteristics, rendering them vulnerable to node removal and cascading failures. Traditional graph-theoretic models fail to capture the inherent multi-node cooperative relationships within SINs, while conventional centrality metrics overlook service coupling features and cross-community propagation risks, leading to inaccurate key node identification. To address these issues, this paper proposes a key node mining method based on soft community partition and hypergraph interweaving degree. The method first constructs a hypernetwork to characterize the high-order associations among three types of services—communication, remote sensing, and navigation. Second, a probabilistic generative model, Hypergraph-MT, is introduced to detect overlapping communities and capture the multi-role attributes of satellites. Finally, the Interweaving Degree (IW) and its derived Bridge Score (IWBS) are defined to quantify node importance from two dimensions: load capacity and cross-community connectivity. Experimental results demonstrate that, compared with degree centrality and betweenness centrality, the proposed method increases the service interruption rate from 27.4% to 64.8% under a 5% removal ratio, and the service efficiency loss is elevated by more than 60%. In comparison with pure interweaving degree removal, IWBS triggers deeper cascading removal at a 30% removal ratio, reduces the removal cost by 8.9%, and improves the input–output ratio by 8.2% under high-intensity removal. Moreover, the algorithm exhibits linear scalability and strong robustness to parameter variations. The proposed method significantly enhances the accuracy and efficiency of key node identification in SINs, providing theoretical support for constellation deployment optimization and priority protection decision-making.

Cataloging task resource scheduling is a key technology for the efficient utilization of ground-based radar networks and for supporting space situational awareness. This problem is highly challenging due to the large scale of tasks, strict time window constraints, and complex resource-task mapping relationships. It requires algorithms to effectively balance multiple conflicting optimization objectives within a huge and sparse solution space, placing extremely high demands on the convergence, diversity maintenance, and computational efficiency of the algorithms. This paper presents a systematic review of the latest research progress in cataloging resource scheduling methods. First, commonly used optimization objectives and constraint conditions in this field are outlined, and two key subproblems—priority modeling and conflict resolution—are analyzed in depth. Subsequently, following the trajectory of technological evolution, the application paradigms, performance characteristics, and limitations of mainstream algorithms are reviewed. Given the inherent multi-objective optimization nature of the problem, the advantages and challenges of multi-objective optimization algorithms are discussed. Finally, based on a unified problem context, the performance and operational boundaries of existing algorithms are compared and analyzed, and future research directions and core challenges in the field are presented.