Radio Frequency Fingerprinting (RFF) exploits inherent hardware-level imperfections of wireless transmitters as unclonable identifiers for device identification. These unique signatures, concealed in transmitted signals, inevitably experience complex distortions during wireless propagation (i.e., coupled with ambient noise and channel fading), making it extremely challenging for reliable extraction. Despite substantial research efforts dedicated to advancing effective fingerprint extraction techniques, current approaches still struggle in handling fingerprint robustness under distance variations, leading to severe SNR fluctuations and complex multipath effects. To address this gap, we propose the first unsupervised framework for distance-invariant radio frequency fingerprinting, eliminating dependence on labeled target domain data. Specifically, we first preprocess raw RF samples by confining them within a specified variation range and filtering noisy high-frequency components while avoiding aliasing. For source domain data, we then propose a set of physics-inspired data augmentation techniques designed to emulate realistic wireless signal propagation effects. Building on this, we introduce a dual alignment contrastive learning method to explicitly decouple identity-discriminative features, ensuring the model focuses on device-specific traits. Furthermore, we incorporate a pseudo-labeling-based domain adaptation module to refine the model for the unlabeled target domain, enhancing its generalization to unseen distances. Extensive experiments on public datasets show that our method achieves the identification accuracy outperforming state-of-the-art approaches by 40%, while maintaining computational efficiency suitable for edge deployment.

Physical layer secret key generation has emerged as a promising approach for secret key establishment in wireless networks. Unlike traditional quantization-based methods, recent studies have explored matching the patterns of segmented channel samples of equal length for key agreement. However, equal-length segmentation either suffers from inconsistencies between users for short segments or a reduced key generation rate for long ones. To address these issues, we propose a Variable-length Segment Matching-based Secret Key Generation method, VSM-SKG, which adaptively partitions channel samples into variable-length segments to enhance overall matching accuracy, key generation rate, and encryption strength. Specifically, we introduce a dissimilarity-enhanced segmentation and calibration strategy that partitions channel samples into variable-length segments to enlarge segment-wise dissimilarity. To achieve consistent key recovery between users, we develop a dynamic path-aware key generation method that identifies potential segmentation patterns and generates agreed-upon secret keys using a recursive approach combined with a fast retrieval mechanism. Theoretical analyses and real-world experiments validate the attributes of VSM-SKG in terms of accuracy, efficiency, and security in key generation.

Physical layer authentication (PLA), which leverages device-specific physical layer features to achieve information-theoretic security with low complexity, offers a hardware-rooted security solution for next-generation Internet of Things (IoT) networks. While existing PLA approaches primarily rely on passive extraction of inherent hardware-induced radio frequency features, they remain vulnerable to adversarial spoofing that replicates legitimate radio frequency fingerprints (RFF). To address this critical vulnerability, we propose an active PLA framework that employs a challenge–response protocol to embed session-specific perturbation into each transmission. Upon receiving a nonce from the receiver, the legitimate transmitter generates a hash value and embeds a corresponding in-phase and quadrature (IQ) imbalance-based perturbation into the baseband signal. This design conceals inherent hardware-specific RFF and injects dynamic, unpredictable fingerprints that vary across sessions and are resilient to forgery. At the receiver, authentication is performed using a learning-based method that combines a CNN-based feature extractor with a lightweight logistic regression classifier trained on augmented samples. Extensive simulations demonstrate that the proposed framework achieves high authentication accuracy under both static and dynamic channel conditions, while effectively resisting advanced spoofing attacks, including GAN-based impersonation. These results confirm the robustness, generalization capability, and applicability of the proposed scheme for secure IoT communications.

Achieving autonomous navigation and positioning for UAV through computer vision technology has emerged as a key focus and hot topic in current research. The precise matching of UAV images is crucial for realizing this functionality. However, the intrinsic difference between satellite images and UAV images resulted in the sub-optimal performance of traditional matching algorithms. Although the depth-local feature matching of linear Transformers has demonstrated superior hierarchical matching capabilities in UAV image matching, the lack of fine local interactions between pixel tokens restricts their ability to extract highly accurate and localized correspondences. To address the limitations of existing linear Transformer matching algorithms, this paper proposes a novel method called MixFormer, a Mixed Attention Matching Transformer for UAV image feature matching. The algorithm comprises two main aspects: firstly, it introduces a hierarchical feature matching attention framework that operates attention at different scales, integrating global attention and local attention to achieve global context awareness and fine-grained matching, thereby enabling efficient and compact implementation. Secondly, it introduces a feature transformation module for transitional Convolutional Neural Network (CNN) to extract features with full receptive fields. Experimental results on benchmark tests confirm the effectiveness and efficiency of the proposed technique.

Mobile device operation tasks are increasingly becoming a popular multi-modal AI application scenario. Current Multi-modal Large Language Models (MLLMs), constrained by their training data, lack the capability to function effectively as operation assistants. Instead, MLLM-based agents, which enhance capabilities through tool invocation, are gradually being applied to this scenario. However, the two major navigation challenges in mobile device operation tasks — task progress navigation and focus content navigation — are difficult to effectively solve under the single-agent architecture of existing work. This is due to the overly long token sequences and the interleaved text-image data format, which limit performance. To address these navigation challenges effectively, we propose Mobile-Agent-v2, a multi-agent architecture for mobile device operation assistance. The architecture comprises three agents: planning agent, decision agent, and reflection agent. The planning agent condenses lengthy, interleaved image-text history operations and screens summaries into a pure-text task progress, which is then passed on to the decision agent. This reduction in context length makes it easier for decision agent to navigate the task progress. To retain focus content, we design a memory unit that updates with task progress by decision agent. Additionally, to correct erroneous operations, the reflection agent observes the outcomes of each operation and handles any mistake accordingly. Experimental results indicate that Mobile-Agent-v2 achieves over a 30% improvement in task completion compared to the single-agent architecture of Mobile-Agent. The code is open-sourced at https://github.com/X-PLUG/MobileAgent.

Abstract As the predominant modern version control system, Git has become an indispensable tool for both commercial and open‐source software projects. It substantially improves software development effectiveness and efficiency through its distributed version control system, fostering seamless collaboration among teams and across locations. However, research has found that many developers have doubts about using Git commands, while the official Git documentation is rather scanty, that is, lacking sufficient explanations and examples. To help developers learn and use Git commands, we propose the first approach (Richen) for enriching Git documentation with usage examples and scenarios by leveraging crowd knowledge from Stack Overflow. Richen retrieves Git‐related posts from Stack Overflow, extracts relevant Q&A pairs, and selects representative command usages, including usage examples and scenarios, for different Git commands. Experimental results have shown that Richen can extract informative and concise command usages for Git commands. Compared with alternative methods adapted from API usage mining, the command usages obtained by Richen have significant advantages in terms of relevance, readability, and usability. Furthermore, we have shown through an empirical study that the command usages extracted by Richen can better help developers complete Git command‐related tasks.

This work analyzes the impacts of training sample size on the performance of supervised classification methods when coarse resolution imageries are employed for regional land cover mapping. We utilized FegnYun-3C composite imageries with 1km spatial resolution and random forest (RF) and support vector machine (SVM) algorithms that were trained and tested with five sets of reference datasets: 66/34, 69/31, 73/27, 76/24 and 79/21.The results show that the performance of the two algorithms increases with increasing the size of the training examples until a certain point, and achieves the maximum accuracy (0.86 for RF and 0.84 for SVM) when the ratio was 76/24. However, considering the 79/21 (train/test) ratio made no change in accuracy, implying increasing a training dataset beyond a certain limit has no effect. Moreover, despite the size of training samples employed, the RF outperformed the SVM in contrary to the claim that SVM yields a better accuracy in case of scarce training data by pervious studies.

The West African Craton hosts economic minerals such as diamonds, gold, and bauxite. Yet, it remains underexplored as poor socio-political and environmental conditions limit traditional exploration in the area. This study assesses the performance of the Landsat-9 OLI-2/TIR-2 in mapping lithologies and highlighting potential metallogenic points using remote sensing techniques. Specifically, we perform classification tasks on preprocessed Landsat-9 dataset to classify the lithological features using a hybrid machine learning (ML) module integrating Local Fisher Discriminant Analysis (LFDA) and Support Vector Machine (SVM) against three (3) baseline ML algorithms. The aim is to enhance computing proficiency and improve classification precision. The accuracy results show that the proposed LFDA and SVM algorithm performs better than Random Forest (RF) and KNN. The results distinguish nine lithological units with a 0.9992 overall accuracy. Additionally, Landsat-9 OLI-2/TIR-2 proves efficient in attaining high accuracy in conducting geological exploration work in terrains under unfavorable conditions.

This paper compared multi-temporal composite products of FY-3C VIRR and MODIS for regional land cover (LC) mapping of a part of Africa. LC classification was conducted using a random forest algorithm in the Scikit-Learn library of Python after the model was trained using reference data that were collected by combining three techniques and employed simultaneously i.e. Landsat 8 image interpretation, referring exiting maps, and crosschecking on Google Earth pro/ maps. Based on the overall accuracy (OA) and kappa value (k) the two instruments showed insignificant performance variation although FY-3C VIRR achieved slightly higher OA (.82) and k (.79) which are 1 % higher than MODIS. However, the two instruments exhibited notable performance differences in discerning individual classes. The FY-3C data are generally better for vegetation classification; MODIS, whereas, classified built-up, bare/sparse-vegetation, and water bodies with better accuracy. Moreover, the incorporation of FY-3C thermal bands improved its accuracy significantly, by 3%.

Modern science is increasingly data-driven and collaborative in nature. Comparing to ordinary data processing, big data processing that is mixed with great missing date must be processed rapidly. Considering this requirement this paper proposes a dynamic perception rule acquirement algorithm to implement fast and accurate information decision supporting model for incomplete data. It is inevitable that information contains incomplete data, and huge information being processing require fast algorithm to complete knowledge extraction. The method based on dynamic perception rule can achieve automatic analysis and intelligent cognition for the information decision supporting. Based on direction of maximum entropy at any moment, the perception rule can improve the recognition rate. Furthermore the dynamic perception rule adopts the tolerant relation to accommodate the incomplete data processing capability. The simulative analysis of diesel engine fault shows that the dynamic perception rule can achieve fast information decision supporting and the accuracy is certainly improved even for incomplete data.

At present, application of dense targets is one of the important means of penetration. In many observation conditions, the echoes from the dense targets mixed with many aliasing signals, and conventional radar signal processing algorithms do not take the aliasing signals into account. Therefore it is difficult for conventional algorithms to recognize multi-targets. In this paper, an improved ESPRIT algorithm is proposed which can recognize the multi-targets from the aliasing echoes and greatly reduce the computational complexity without changing the algorithm accuracy, especially can obtain a better estimation in the case of low SNR environment. The proposed algorithm can firstly quickly realize the estimate of scattering center parameters of target echoes, and then based on the estimation, the aliasing targets can be recognized. The Simulation also verifies the improved ESPRIT algorithm has a better identification and recognition capability of aliasing targets in low SNR condition. Moreover because of reduction of the computational complexity, the performance of proposed algorithm is faster than conventional methods, especially in the case of multiple aliasing scattering centers.

Modern science is increasingly data-driven and collaborative in nature. Comparing to ordinary data processing, big data processing that is mixed with great missing date must be processed rapidly. The Rough Set was generated to deal with the large data.In this paper, we proposed animproved algorithm for dynamic Cognitive extractionwhich deals with adaptive fuzzy attribute values and the fuzzy attribute reduction aiming at uncertainty datasuch asdata with diversity or missing character faced by the big data with using Fuzzy Rough Set Theory.At the aspect of information decision, according to the Real-time input information, it deep analyzes the dynamic information entropy of the data itself and chooses the biggest prediction information entropy direction for the cognitive rules to achieve rapid recognitive of data, complete information of quick decision.Because the algorithm is adopted to predict the best direction of information entropy, so the recognitive effect is also improved. At the end of the paper, we have analyzed superiority of the dynamic cognitive algorithm by using breast cancer data as the foundation.