Traditional automatic modulation classification methods operate under the closed-set assumption, which proves to be impractical in real-world scenarios due to the diverse nature of wireless technologies and the dynamic characteristics of wireless propagation environments. Open-set environments introduce substantial technical challenges, particularly in terms of detection effectiveness and computational complexity. To address the limitations of modulation classification and recognition in open-set scenarios, this paper proposes a semi-supervised open-set recognition approach, termed SOAMC (Semi-Supervised Open-Set Automatic Modulation Classification). The primary objective of SOAMC is to accurately classify unknown modulation types, even when only a limited subset of samples is manually labeled. The proposed method consists of three key stages: (1) A signal recognition pre-training model is constructed using data augmentation and adaptive techniques to enhance robustness. (2) Feature extraction and embedding are performed via a specialized extraction network. (3) Label propagation is executed using a graph convolutional neural network (GCN) to efficiently annotate the unlabeled signal samples. Experimental results demonstrate that SOAMC significantly improves classification accuracy, particularly in challenging scenarios with limited amounts of labeled data and high signal similarity. These findings are critical for the practical identification of complex and diverse modulation signals in real-world wireless communication systems.

An autonomous place recognition system is essential for scenarios where GPS is useless, such as underground tunnels. However, it is difficult to use existing algorithms to fully utilize the small number of effective features in underground tunnel data, and recognition accuracy is difficult to guarantee. In order to solve this challenge, an efficient point cloud position recognition algorithm, named Dual-Attention Transformer Network (DAT-Net), is proposed in this paper. The algorithm firstly adopts the farthest point downsampling module to eliminate the invalid redundant points in the point cloud data and retain the basic shape of the point cloud, which reduces the size of the point cloud and, at the same time, reduces the influence of the invalid point cloud on the data analysis. After that, this paper proposes the dual-attention Transformer module to facilitate local information exchange by utilizing the multi-head self-attention mechanism. It extracts local descriptors and integrates highly discriminative global descriptors based on global context with the help of a feature fusion layer to obtain a more accurate and robust global feature representation. Experimental results show that the method proposed in this paper achieves an average $F_1$ score of 0.841 on the SubT-Tunnel dataset and outperforms many existing state-of-the-art algorithms in recognition accuracy and robustness tests.

In a satellite-denied environment, a swarm of drones is capable of achieving relative positioning and navigation by leveraging the high-precision ranging capabilities of the inter-drone data link. However, because of factors such as high drone mobility, complex and time-varying channel environments, electromagnetic interference, and poor communication link quality, distance errors and even missing distance values between some nodes are inevitable. To address these issues, this paper proposes a low-rank optimization algorithm based on the eigenvalue scaling of the distance matrix. By gradually limiting the eigenvalues of the observed distance matrix, the algorithm reduces the rank of the matrix, bringing the observed distance matrix closer to the true value without errors or missing data. This process filters out distance errors, estimates and completes missing distance elements, and ensures high-precision calculations for subsequent topology perception and relative positioning. Simulation experiments demonstrate that the algorithm exhibits significant error filtering and missing element completion capabilities. Using the F-norm metric to measure the relative deviation from the true value, the algorithm can optimize the relative deviation of the observed distance matrix from 11.18% to 0.25%. Simultaneously, it reduces the relative positioning error from 518.05 m to 35.24 m, achieving robust topology perception and relative positioning for the drone swarm formation.

Precise odor source localization in turbulent and sparse environments plays a vital role in enabling robotic systems for hazardous chemical monitoring and effective disaster response. To address this, we propose Cooperative Gravitational-Rényi Infotaxis (CGRInfotaxis), a distributed decision-optimization framework that combines multi-agent collaboration with hybrid cognitive strategy to improve search efficiency and robustness. The method integrates a gravitational potential field for rapid source convergence and Rényi divergence-based probabilistic exploration to handle sparse detections, dynamically balanced via a regulation factor. Particle filtering optimizes posterior probability estimation to autonomously refine search areas while preserving computational efficiency, alongside a distributed interactive-optimization mechanism for real-time decision updates through agent cooperation. The algorithm’s performance is evaluated in scenarios with fixed and randomized odor source locations, as well as with varying numbers of agents. Results demonstrate that CGRInfotaxis achieves a near-100% success rate with high consistency across diverse conditions, outperforming existing methods in stability and adaptability. Increasing the number of agents further enhances search efficiency without compromising reliability. These findings suggest that CGRInfotaxis significantly advances multi-agent odor source localization in turbulent, sparse environments, offering practical utility for real-world applications.

Traditional automatic modulation classificationmethods operate under the closed-set assumption, which provesto be impractical in real-world scenarios due to the diversenature of wireless technologies and the dynamic characteristicsof wireless propagation environments. Open-set environments introduce substantial technical challenges, particularly in terms ofdetection effectiveness and computational complexity. To addressthe limitations of modulation classification and recognition inopen-set scenarios, this paper proposes a semi-supervised openset recognition approach, termed SOAMC (Semi-SupervisedOpen-Set Automatic Modulation Classification). The primaryobjective of SOAMC is to accurately classify unknown modulation types, even when only a limited subset of samples ismanually labeled. The proposed method consists of three keystages: (1) A signal recognition pre-training model is constructedusing data augmentation and adaptive techniques to enhancerobustness. (2) Feature extraction and embedding are performedvia a specialized extraction network. (3) Label propagation isexecuted using a graph convolutional neural network (GCN) toefficiently annotate the unlabeled signal samples. Experimentalresults demonstrate that SOAMC significantly improves classification accuracy, particularly in challenging scenarios with limitedlabeled data and high signal similarity. These findings are criticalfor the practical identification of complex and diverse modulationsignals in real-world wireless communication systems.

Generalized space-time-frequency index modulation (GSTFIM) inherits the drawbacks of the conventional orthogonal frequency-division multiplex (OFDM), such as being sensitive to carrier frequency offset (CFO). For a robust design against this problem, in this contribution, a novel construction of a message passing (MP)-aided detector is developed for GSTFIM systems to combat the influence of CFO, while offering a flexible tradeoff between transmission performance and computational complexity. Through complexity analysis and simulation results, we demonstrate that, in the context of CFO, with a careful design, the developed MP detector is capable of approaching traditional GSTFIM with maximum likelihood (ML) detection, and of offering better performance at lower complexity compared to its minimum mean-square-error (MMSE)-aided counterpart.