Guided by the principle of “Never Trust, Always Verify”, Zero Trust Architecture (ZTA) mandates continuous monitoring and analysis of users and entities, highlighting the critical role of behavior analytics. However, the growing volume of audit data and its complex contextual information render many existing behavior analytics methods insufficient. Moreover, most approaches rely on high-quality labeled data for supervised training, limiting their effectiveness against previously unseen malicious behaviors. To address these challenges, we propose the Large Language Model for Behavior Analytics (LLMBA) framework. LLMBA leverages a Large Language Model (LLM) to analyze behavioral patterns of internal users and entities, capitalizing on the LLM’s strong ability to model sequential data. We introduce a multi-level behavior encoding scheme to capture both contextual and temporal information from behavior records, producing rich input representations for the LLM-enhanced model. The LLM is fine-tuned using self-supervised learning, enabling the detection of unknown malicious behaviors. To reduce the computational and storage overhead inherent in LLMs, we apply knowledge distillation to compress the model while maintaining high detection performance. Extensive experiments on the CERT Insider Threat dataset demonstrate that LLMBA outperforms state-of-the-art baselines in detection accuracy. Furthermore, the compressed student model achieves superior performance compared with existing methods under comparable runtime constraints, making LLMBA highly suitable for real-world deployment.

High-frequency (HF) communication is known as a pivotal task in military operations, disaster relief efforts and global broadcasting by virtue of its unique advantages of long-distance reach, non-relay capabilities and flexible deployment. Accurate forecasting of foF2 values is critical for optimizing HF communication technologies. However, the complex, nonlinear and dynamic nature of ionospheric variations caused by solar activity and atmospheric conditions makes it a huge challenge. International reference ionosphere relies on semi-empirical formulas with substantial prediction errors, while some data-driven variants focus mainly on time series predictions derived from individual stations, thus ignoring the inter-station correlations that are closely related to the model accuracy and effectiveness. In this paper, we propose a novel deep learning approach based on spatio-temporal graph convolutional networks (STG-foF2) to address the above issues of ionospheric forecasting in the context of HF communication, which leverages a cascade structure of graph convolutional layers and long short-term memory layers to extract the intricate patterns and connections across diverse geographical locations and temporal intervals. Extensive experiments on Global ionospheric radio observatory datasets verify the superior performance in terms of accuracy and time complexity of the proposed STG-foF2 over the existing benchmarks under different geomagnetic activity states.

Extended reality (XR) technologies and applications have grown rapidly in recent years. In addition to providing immersive ultrahigh-definition (UHD) video, XR also allows for a haptic experience where devices can be remotely manipulated to accomplish tasks. However, varying numbers of XR users accessing the communication system can strain limited spectrum resources, posing challenges in resource allocation. Therefore, this article studies resource blocks (RBs) allocation problem in a downlink transmission scenario where real-time cloud XR video and haptic contents need to be transmitted simultaneously. We also consider the random variation in the number of XR users and propose an adaptive distributed multiagent deep reinforcement learning (DRL) combined with Transformer (ADMA-DcT) for dynamic RB allocation method. This method addresses the dynamic change in network input dimensions due to user number variability using a state division module and a self-attention mechanism in the encoder module. To our knowledge, this is the first work to study the RB allocation problem in Cloud XR transmission considering simultaneous transmitting of video and haptic services with a dynamically changing user base. Our extensive simulations show that the ADMA-DcT model, end-to-end trained, outperforms other benchmarks in successfully serving a larger number of XR users under varying user number conditions, demonstrating excellent adaptivity and robustness.

Due to sophisticated deployments of all kinds of wireless networks (e.g., 5G, Wi-Fi, Bluetooth, LEO satellite, etc.), multiband signals distribute in a large bandwidth (e.g., from 70 MHz to 8 GHz). Consequently, for network monitoring and spectrum sharing applications, a sniffer for extracting physical layer information, such as structure of packet, with low sampling rate (especially, sub-Nyquist sampling) can significantly improve their cost- and energy-efficiency. However, to achieve a multiband signals sniffer is really a challenge. To this end, we propose Sums, a system that can sniff and analyze multiband signals in a blind manner. Our Sums takes advantage of hardware and algorithm co-design, multi-coset sub-Nyquist sampling hardware, and a multi-task deep learning framework. The hardware component breaks the Nyquist rule to sample GHz bandwidth, but only pays for a 50 MSPS sampling rate. Our multi-task learning framework directly tackles the sampling data to perform spectrum sensing, physical layer protocol recognition, and demodulation for deep inspection from multiband signals. Extensive experiments demonstrate that Sums achieves higher accuracy than the state-of-the-art baselines in spectrum sensing, modulation classification, and demodulation. As a result, our Sums can help researchers and end-users to diagnose or troubleshoot their problems of wireless infrastructures deployments in practice.

Massive MIMO is a key technology for high quality to improve the spectral efficiency of wireless communications. However, it is limited by high cost and power consumption. Inspired by the low-cost property of Reconfigurable Intelligent Surfaces(RIS), we propose a location-based beamforming de-sign for a multi-layer RIS structure-assisted MIMO to realize high gain at a low cost. Specifically, by leveraging the user's location data, we collaboratively optimize the active precoding at the transmitter and the passive precoding at the RIS to achieve the corresponding beam response output. Given the non- convex nature of the optimization issue, we employ the alternate minimization (AltMin) technique. The limitations of computing power may prevent real-time fulfillment of requirements, thus we propose an offline algorithm for beam rotation and beam superposition based on the reference beam and the location information of the user. The simulation results demonstrate the superiority of our proposed location-based beamforming design in multi-layer RIS-assisted MIMO communications.

Deep learning has achieved success in optical image super-resolution (SR). However, few methods have been proposed for synthetic aperture radar (SAR) images. This is because SAR images inherently suffer from severe speckle noise, and their resolution is typically much lower compared with optical images, making SR more challenging. To simultaneously denoise and preserve texture details in SAR image SR tasks, we propose a joint despeckling and SR network via knowledge distillation (DSRKD). By implementing feature distillation (FD) in the encoding stage, we enable the student network to obtain noise-free latent variables, thereby restoring clean SR images, and simultaneously enhancing effective supervision for the student using target distillation (TD). This distillation strategy does not incur additional computational costs during inference. Experimental results demonstrate that our algorithm effectively suppresses speckle noise while preserving texture details of images on both synthetic and SAR datasets. Compared with other state-of-the-art algorithms, the proposed method achieves superior performance on both visual results and quantitative metrics, such as PSNR, SSIM, and other nonreference indicators under various SR scales and speckle intensities.

This paper presents a network slice design for ultra high definition (UHD) video broadcast/multicast to achieve higher network efficiency and improved quality of experience (QoE). The proposed network slice design consists of a rateless source compression scheme and an analog-coded SoftCast scheme. The rateless Spinal code is adopted to compress the video source at content server and the compressed source is transmitted from content server across wireless core network to the base station. Ana prioriinformation-assisted Spinal decoder is designed to utilize the sparsity of bit planes for compression. In the analog-coded SoftCast scheme, we design a new chaotic function-based analog code with negligible power penalty for the generalized Gaussian-distributed source in SoftCast because the existing chaotic functions designed for uniformly distributed sources suffer from serious power penalty in SoftCast. We also design a maximuma posterioriprobability decoding algorithm for the proposed analog code in order to exploit the statistics of video source asa prioriinformation to improve the performance. The experimental results show that the proposed rateless code-based compression scheme achieves efficient compression and approaches the bound of binary erasure channel. In particular, the 1/2 analog-coded SoftCast has almost 2 dB gain over conventional SoftCast with two repetitions, and the 1/3 analog-coded SoftCast has almost 3 dB gain over conventional SoftCast with three repetitions. The system simulations for the broadcast system show higher network capacity and improved QoE in the proposed UHD slice, because the reconstructed video quality of each user is commensurate with its channel condition.