With the rapid development of the low-altitude economy, Unmanned Aerial Vehicle (UAV) recognition has become increasingly important. However, vision-based methods progressively become ineffective as the UAV’s flight altitude increases. Therefore, radar information has emerged as the predominant approach for UAV recognition. However, existing low-altitude targets recognition methods either depend on long-sequence trajectory information or possess excessively high terminal computing power requirements, thereby rendering them difficult to deploy. In this study, a UAV recognition framework RaLMEN based on radar information is proposed, utilizing an ensemble learning architecture that incorporates BiLSTM and MLP. The framework integrates both temporal and static features extracted from aerial targets, aiming to provide a robust solution for low-altitude traffic surveillance. Experiments demonstrate that the proposed method requires only the temporal RCS information and position coordinates of a few trajectory points to accurately determine whether the target is a UAV. Furthermore, when evaluated with test data sampled from domains different from the training set, the framework demonstrates exceptional generalization performance, outperforming the baseline algorithms of all mainstream temporal neural networks.

The rapid growth of publicly available data has fueled deep learning advancements but also raises concerns about unauthorized data usage. Unlearnable Examples (UEs) have emerged as a data protection strategy that introduces imperceptible perturbations to prevent unauthorized learning. However, most existing UE methods produce perturbations strongly tied to specific training sets, leading to a significant drop in unlearnability when applied to unseen data or tasks. In this paper, we argue that for broad applicability, UEs should maintain their effectiveness across diverse application scenarios. To this end, we conduct the first comprehensive study on the transferability of UEs across diverse and practical yet demanding settings. Specifically, we identify key scenarios that pose significant challenges for existing UE methods, including varying styles, out-of-distribution classes, resolutions, and architectures. Moreover, we propose $\textbf{Versatile Transferable Generator}$ (VTG), a transferable generator designed to safeguard data across various conditions. Specifically, VTG integrates Adversarial Domain Augmentation (ADA) into the generator’s training process to synthesize out-of-distribution samples, thereby improving its generalizability to unseen scenarios. Furthermore, we propose a Perturbation-Label Coupling (PLC) mechanism that leverages contrastive learning to directly align perturbations with class labels. This approach reduces the generator’s reliance on data semantics, allowing VTG to produce unlearnable perturbations in a distribution-agnostic manner. Extensive experiments demonstrate the effectiveness and broad applicability of our approach. Code is available at https://github.com/zhli-cs/VTG.

Semantic segmentation is a critical component of autonomous driving perception systems and has gained increasing attention in recent advancements. Autonomous vehicles frequently encounter diverse environmental conditions, highlighting the significance of research into domain adaptation for semantic segmentation. We established a domain adversarial framework for enhancing the cross-domain perception of autonomous vehicles. However, previous works have shown that the general adversarial training-based methods can lead to indistinguishable features, resulting in a decline in the robustness of perception. In this regard, we adopted contrastive learning to guarantee the proximity of similar samples, and the principle of different classes of samples to a certain extent. This ensures the closeness of similar samples and upholds the feature distinction between different classes of samples to a significant degree. Thus, we proposed a novel domain adversarial training framework incorporating the contrastive learning method to enhance cross-domain feature recognition for autonomous driving systems. We empirically evaluate the proposed method against several recent baselines showing improved benchmark performances, confirming the effectiveness of the proposed method.

Safety is a paramount concern in the realm of autonomous vehicles. Developing precise safety assessment is challenging due to the need to blend qualitative and quantitative analyses of various safety factors. To address this challenge, this paper presents an innovative safety assessment method based on the cloud model. This method employs fundamental cloud model elements like expectation, entropy, and ultra-entropy. It also employs a sophisticated double conditional single rule generator to integrate multiple assessment indicators, resulting in an integrated risk assessment cloud. This cloud dynamically represents varying risk levels based on indicator characteristics. The method evaluates the real-time safety level by assessing the proximity between the integrated risk assessment cloud and the standard cloud. This proximity analysis reveals the prevailing risk level. Empirical validation involves rigorous testing within typical scenarios, demonstrating the utility and potential of the method to assess safety for decision-making of autonomous vehicles. The capacity of the method to monitor and assess autonomous vehicle decision-making systems makes it a significant contribution to the field. Beyond empirical contributions, this paper offers theoretical insights that can shape the future of safety assessment methods for autonomous vehicles. In summary, this paper emphasizes the importance of safety for autonomous vehicles and paves the way for evolving safety assessment methods in this dynamic field.

Network alignment aims at detecting the corresponding entities across multiple networks, which is an essential basis for the fusion and analysis of multiple network information. Moreover, embedding-based network alignment has gradually become one of the promising methods. However, existing methods ignore the confusing selection problem caused by the similarity-orientated principle of network embedding and over-dependence on the hypothesis of structural consistency. In this paper, we propose an end-to-end Graph Alignment Neural Network (GANN) model with graph-to-sequence learning. GANN mainly consists of two modules: Graph encoder and Sequence decoder. In graph encoder module, we present a restricted network embedding method, which can not only capture the local structure and attribute information of nodes but also realize the constraint of node embedding and space reconciliation. In sequence decoder module, we propose a graph-to-sequence learning model to address large graphs' structural consistency hypothesis problem. In this model, an attention-based LSTM mechanism is introduced to infer a node in the source network corresponding to the candidate node sequence in target networks. In this candidate sequence, the correct aligned node is placed at the top. We demonstrate that GANN outperforms the state-of-the-art methods in network alignment tasks on various real-world datasets.

Abstract Several spectrum sensing systems based on sub‐Nyquist sampling have been extensively studied to deal with difficulties of traditional wideband spectrum sensing in cognitive radio (CR) networks. The modulated wideband converter (MWC) is an effective application and has drawn considerably more attention over the past few years. In this study, MWC with a nested array (NA) system is proposed to improve spectrum sensing performance, which is an array sensing system with sub‐Nyquist sampling. The simulations show that the connection of co‐array and MWC obtains the lower minimal system sampling rate than MWC. Second, the proposed system enables frequency and power spectrum estimation with sub‐Nyquist sampling for more sources than sensors. Finally, our alternative spectrum sensing system outperforms other conventional methods in terms of sensing accuracy and design complexity.