The proliferation and advancement of generative AI tools blur the boundaries between authentic and AI-synthesized imagery, sparking widespread public concern about visual content authenticity. While existing AI-generated image detection approaches have demonstrated notable achievements, Multimodal Large Language Models (MLLMs) remain relatively underutilized in this field. As MLLMs show strong abilities in understanding image features and can provide high-quality textual analysis, substantial potential within MLLM itself remains well explored. In this work, we propose a novel triple-feature fusion prompt learning framework to effectively stimulate the potential of the MLLMs in detecting AI-generated images. We build a prompt structure that allows MLLM to analyze the inauthenticity of AI-generated images from different perspectives, both locally and globally, and ask questions on the generated image dataset to obtain open-ended answers. We then design a triple fusion encoder that combines semantic features, structured knowledge representation and fine-grained modifications. Through a hierarchical feature induction mechanism, we were able to construct multidimensional feature sets with interpretive properties. Then, we compose a set of prompts based on the obtained feature sets to guide the MLLM to discriminate the input images under global and partial strategies. The test result shows prominent improvement compared to direct inquiry approaches, suggesting that MLLMs demonstrate substantial potential in no need for fine-tuning in AI-generated image detection. This work will drive future research toward prompt-based methods, further expanding the capabilities of MLLMs.

To address the conflict between the limited availability of spectrum resources and the swiftly growing number of frequency equipment in the Internet of Vehicles, this article starts from the solution of dynamic access of radio equipment and studies the problem of power spectrum density prediction of different channels, which is called spectrum prediction. We first propose the graph structure learning problem of electromagnetic spectrum data under the graph contrastive learning (GCL) framework from the causal perspective, and establishes a graph structure representation model between channel signal activity rules. Then, we establish the statistical distribution models of different channels at fine resolution based on the Gaussian mixture model. Then, the statistical model is embedded as prior knowledge using a graph convolutional network (GCN), and the channel association features are mapped to the association knowledge embedding using a graph structure encoder. This article proposes a novel spectrum prediction architecture based on GCL and multiscale feature fusion GCN (CL-MFGCN) to mine the time-frequency implicit knowledge of spectrum data, and the above knowledge embedding is integrated. This article visualizes the channel association relationship in the form of a graph structure. The experimental results indicate that the CL-MFGCN, reduces the average MAE by 14.6% (from 1.436 to 1.226) and the average MAPE by 12.9% (from 0.017 to 0.0148) compared to the second-best Pyraformer model, while maintaining a lower model complexity.

Specific emitter identification (SEI) enhances wireless communication security by identifying specific devices or signals to monitor anomalies effectively. However, data scarcity and heterogeneity challenge traditional centralized methods and few-shot learning (FSL), which depend on centralized data. We propose a personalized decentralized federated FSL method (FedDePF) for SEI. FedDePF organizes edge devices into clusters, enabling local aggregation within clusters and global collaboration between cluster centers. This reduces server communication overhead and addresses data heterogeneity in distributed environments. Experiments show FedDePF significantly improves SEI performance in data-scarce scenarios, outperforming traditional decentralized methods, providing a secure and efficient solution.

With the rapid development of ubiquitous networks and unmanned devices, distributed unmanned systems have emerged. Applying distributed unmanned systems to AIS signal recognition helps address the challenges of identifying AIS signals in extreme environments such as remote seas and isolated regions. Nevertheless, the direct sharing of multi-domain sample data not only entails substantial costs but also poses a high risk of privacy information disclosure, significantly limiting the advancement of distributed unmanned systems in the area of AIS signal recognition. To address the above problems, this paper proposes a knowledge-driven federated semi-supervised learning based on similarity Consistency (FSLBSC). In detail, firstly, this approach achieves many-to-many cross-domain knowledge sharing while ensuring sample privacy and security, relying on knowledge-sharing federated learning. It resolves the issue of unbalanced sample data labels within a single domain and enables effective annotation of AIS signals through the secure sharing of multi-source knowledge. Results of simulation experiments demonstrate that the proposed method exhibits higher recognition accuracy for AIS signals when faced with unbalanced multi-domain data labeling. In the experiment using real AIS data gathered from four distinct sea surfaces, this approach obtained a global accuracy rate of 99.53%.

Specific Emitter Identification (SEI) is a critical component of the Industrial Internet of Things (IIoT), enabling effective identification and validation of unauthorized communication devices, thereby preventing malicious interference and signal spoofing. However, SEI methods based on deep learning involve significant computational overhead, and SEI methods based on feature engineering require specialized expertise for feature design, limiting their ability to capture complex patterns. In this paper, we propose a data-knowledge dual-driven adaptive feature fusion approach for specific emitter recognition. Specifically, we present an adaptive feature fusion strategy that integrates domain experts’ prior knowledge with the complex feature recognition capability of deep learning models to achieve efficient SEI classifiers. The approach is evaluated using Automatic Dependent Surveillance-Broadcast (ADS-B) data. The experimental results demonstrate that the proposed method achieves a higher level of identification accuracy and lower time complexity.

Automatic Modulation Open Set Recognition (AMOSR) is a crucial technological approach for cognitive radio communications, wireless spectrum management, and interference monitoring within wireless networks. Numerous studies have shown that AMR is highly susceptible to minimal perturbations carefully designed by malicious attackers, leading to misclassification of signals. However, the adversarial security issue of AMOSR has not yet been explored. This paper adopts the perspective of attackers and proposes an Open Set Adversarial Attack (OSAttack), aiming at investigating the adversarial vulnerabilities of various AMOSR methods. Initially, an adversarial threat model for AMOSR scenarios is established. Subsequently, by analyzing the decision criteria of both discriminative and generative open set recognition, OSFGSM and OSPGD are proposed to reduce the performance of AMOSR. Finally, the influence of OSAttack on AMOSR is evaluated utilizing a range of qualitative and quantitative indicators. The results indicate that despite the increased resistance of AMOSR models to conventional interference signals, they remain vulnerable to attacks by adversarial examples.

With the increasingly serious shortage of spectrum resources, spectrum dynamic access based on spectrum prediction technology is widely recognized. Due to the high burstiness and complex intrinsic correlation of spectrum monitoring data, high-precision multi-channel spectrum prediction is challenging. This paper constructs spectrum monitoring data as a kind of graph structure data based on the correlation of spectrum itself, and designs a graph network model combining Graph convolution network(GCN) and Long-short term memory network(LSTM) for multi-channel spectrum prediction. This paper creatively introduces the method of graph network. And GCN is used instead of CNN to extract the correlation of channels, so as to improve the accuracy of multi-channel prediction. Experiments are conducted based on a real-world spectrum measurement dataset. The results show that the model proposed in this paper has better predictive performance compared with other methods.

With the development of the Internet of Things (IoT), the IoT devices are increasing day by day, resulting in increasingly scarce spectrum resources. At the same time, many IoT devices are facing inevitable malicious attacks. The cognitive Radio-enabled IoT (CR-IoT) is proposed as an effective method for spectrum resource allocation and risk monitoring in the IoT. The signal detection and modulation recognition are the key technologies for CR-IoT, addressing the problem of multisignal detection and automatic modulation classification (AMC) is one of the prerequisites for realizing secure dynamic spectrum access. Based on sliding window and deep learning (DL), this study proposes a multisignal frequency domain detection and recognition method. The frequency spectrum of the time-domain overlapping signal is obtained through the fast Fourier transform (FFT), and the frequency spectrum is segmented based on the signal energy detection method. Finally a complex convolutional neural network (CNN) is constructed for the identification of signal spectrum information. The proposed method can recognize 264 time-domain aliasing and frequency-closed signals with an accuracy of 97.3% under the influence of −2 dB corresponding to the noise of the calibration signal. In addition, the proposed method eliminates the influence of bandwidth, which can effectively detect and recognize the signal types of each component in the frequency band. This method has wide applicability and provides an effective scheme for the IoT cognitive technology.

In this article, interferencequality assessment is of great significance to reflect the communication environment and improve speech communication performance. However, most traditional assessment approaches aimed at the degraded speech produced in the communication telephone network, but lacking of methods in extreme communication environment with ultralow SNR. Therefore, in this article, we proposed a convolutional neural network (CNN) model evaluation method based on Log-Mel spectrogram to evaluate interfered speech quality. In this method, the Mel frequency cepstrum coefficients of interference speech are converted into images, which are used as input of CNN. In order to verify the performance of this method, we collected a speech dataset in real interfered communication scenarios and finished manual annotation. Experiments are carried out on this dataset to evaluate the interference speech, and the performance of this method is compared with that of machine learning evaluation method under different features. Experimental results show that the proposed method gives the better evaluation accuracy. Compared with the previous machine learning methods, the accuracy is improved by 12.5% from 75% to 87.5%.

In the coming 6G communications, network densification, high throughput, positioning accuracy, energy efficiency, and many other key performance indicator requirements are becoming increasingly strict. In the future, how to improve work efficiency while saving costs is one of the foremost research directions in wireless communications. Being able to learn from experience is an important way to approach this vision. Transfer learning (TL) encourages new tasks/domains to learn from experienced tasks/domains for helping new tasks become faster and more efficient. TL can help save energy and improve efficiency with the correlation and similarity information between different tasks in many fields of wireless communications. Therefore, applying TL to future 6G communications is a very valuable topic. TL has achieved some good results in wireless communications. In order to improve the development of TL applied in 6G communications, this article performs a comprehensive review of the TL algorithms used in different wireless communication fields, such as base stations/access points switching, indoor wireless localization and intrusion detection in wireless networks, etc. Moreover, the future research directions of mutual relationship between TL and 6G communications are discussed in detail. Challenges and future issues about integrate TL into 6G are proposed at the end. This article is intended to help readers understand the past, present, and future between TL and wireless communications.

To meet the future needs of increasingly crowded airspace, the International Civil Aviation Organization (ICAO ) proposed to use the Automatic Dependent Surveillance-Broadcast (ADS-B ) to provide navigation and surveillance technology to solve the problems of security and capacity in the airspace. But ADS-B does not offer any authentication and encryption. So it is vulnerable to attacks by various illegal devices. A novel radiofrequency fingerprint (RFF ) recognition method of aircraft identity verification based on deep learning is proposed. The ADS-B signal captured by RTL-SDR is used for confirmation. The experimental results show that the fingerprint is called the Contour Stellar Images with a better recognition effect under different networks and different SNR.

Summary GRAPES is a new generation of numerical weather prediction system developed and used by Chinese researchers. As the accuracy requirement of weather prediction system is increasing, the grid resolution of the global model is greatly increasing; therefore, the computing power becomes one of the important factors restricting the performance of the numerical weather prediction system. In order to solve the problem, the paper introduces the geometric multi‐grid solution to solve the GRAPES Helmholtz equation. The core idea of multi‐grid solution is to eliminate the swing component of the residual by iterative algorithm and eliminate the smooth components of the residuals by using the coarse grid interpolation to correct the fine grid solution. This paper designs a parallel solution for GRAPES multi‐grid, including grid coarsening operation, smooth operation, and the coarsest grid accurate solution operation. Finally, the actual data with the resolution of 1 is tested, achieving the better acceleration effect. At the same time, the analysis and explanations of the test results have some useful conclusions.

Nowadays, the serious security threat of industrial control system and sensors has become a major challenge with the rapid development of Industrial Internet of Things (IIoT). Man-in-the-middle (MITM) attack is a very common intrusion method, which will make a great security threat in the application of IIoT. In IIoT scenario, the lightweight safety certification can play a very important role in the development of data-intensive and decentralized applications running on billions of sensors and devices, preserving their security. Therefore, in this paper, a low-latency high-reliability security mechanism is proposed to avoid the MITM attack in IIoT scenario. First, combining the radio frequency fingerprint (RFF) technology with IIoT applications, a lightweight IIoT security architecture is proposed. Based on the proposed IIoT security architecture, the process of device access authentication and communication service is illustrated. Second, according to the requirement of IIoT identification method, an access authentication method of the device is proposed based on the RFF. The method of feature extraction, classifier designing, and the access authentication process is discussed in detail. Finally, the simulation results show that the identification rate of devices can reach 95% under SNR = 6 dB, and can nearly reach 100% under SNR = 15 dB. Through the new process of access authentication, the access authentication rate can reach 95% under SNR = 15 dB. Therefore, according to the simulation results, the new security mechanisms based on RFF can be used to avoid the MITM attack in IIoT scenario.