Digital twins (DTs) have become a potential technology to perform risk-free simulation of physical entities for deterministic and high-reliability services in diverse scenarios such as autonomous driving and low-altitude economy. In the autonomous driving scenario, traditional DT methods that rely solely on vehicle's real-time state synchronization, however, might lead to unacceptable computing and communication consumption for construction of high-fidelity DT with redundant data. To address this issue, we first propose a query-driven DT architecture to enable the DT to actively request the desired environment data from vehicles based on its simulation result. Then, we formulate an optimization problem whose goal is to minimize autonomous driving position error while accounting for DT fidelity and communication constraints. We also design a cross-time-step progressive query mechanism to further improve communication efficiency. The simulation results show that our proposed method achieves a 24

We present a systematic design that ensures end-to-end service level agreements (SLAs) while minimizing the operational expenditure (OPEX) in an Open Radio Access Network (O-RAN). Unlike existing frameworks, which adopted theoretical models that are often misaligned with industrial standards, we devise our scheme based on standard-compliant, particularly 3GPP-specified models. We develop an algorithm that decouples the complex resource management into three coordinated closed-loop control stages: long-term infrastructure planning, medium-term computing resource allocation, and short-term radio resource scheduling. The algorithm has a short execution time, making it suitable for real-world implementations. We corroborate the efficacy of the proposed scheme through extensive numerical simulations and a prototype built upon an O-RAN testbed called INA-Infra, implemented using Nephio. Experimental results demonstrate that, compared to the baselines, our scheme significantly reduces SLA violations and enhances resource efficiency under non-stationary traffic conditions, confirming its engineering feasibility.

Recent popularization of the Internet of Vehicles (IoV) and vehicle-to-everything (V2X) enables the emergence of real-time vehicular applications, posing challenges for resource-limited vehicles. Toward this end, vehicular edge computing (VEC) has been proposed to alleviate the computational burden on vehicles by leveraging resources from roadside units (RSUs) and VEC servers. While existing works mainly focus on the task requirements for either vehicles or RSUs, the joint task offloading for both V2X and RSU-to-everything (R2X) has not been fully studied. In this paper, we aim to optimize task offloading strategies for both vehicles and RSUs by adopting a multi-hop task offloading manner to fully utilize VEC network resources. This problem introduces a severe state-action space shift issue with varying dimensions and representations, posing challenges for conventional Deep Reinforcement Learning (DRL) approaches. To address this, we propose the Bidirectional Encoder Representations from Transformers (Bert)-based Matching Q-Network (BMQN) algorithm. First, we design the BMQN model to efficiently capture correlations among all vehicles and RSUs through bi-directional attention. Then, we introduce type-embedded grouped attention and available action embedding to mitigate sequence-length overfitting, thereby enhancing generalization capacity. Moreover, we address the state-action space shift through a matching-based manner, which significantly enhances offloading performance by matching states among devices. Simulation results demonstrate that BMQN achieves superior performance compared to existing approaches in scenarios with varying numbers of vehicles and RSUs, and exhibits robust generalization capacity when adapting to unseen scenarios.

With the rapid advancement of vehicular communication facilities and autonomous driving technologies, connected vehicle platooning has emerged as a promising approach to improve traffic efficiency and driving safety. Reliable and low-latency Vehicle-to-Vehicle (V2V) communication is essential for enabling efficient cooperative driving in vehicular networks. However, in the real-world traffic environment, V2V communication may suffer from time-varying delay and packet loss, leading to degraded control performance. To mitigate the adverse effects of non-ideal communication, this paper proposes a Dynamic Communication Topology based Multi-Agent Reinforcement Learning (DCT-MARL) algorithm for robust cooperative platooning. In the proposed framework, each agent dynamically adjusts its communication topology based on causal confidence, which characterizes the decision-relevant importance of other vehicles' behaviors to its control decision under the current driving context. A multi-key gating network is designed to realize the communication topology adaptation mechanism, and a state augmentation scheme is further incorporated to enhance robustness against communication delays. Simulation results demonstrate that the proposed DCT-MARL significantly outperforms state-of-the-art methods in terms of string stability and driving comfort, while improving robustness under unreliable V2V communication conditions, validating its superior effectiveness in realistic vehicular networking environments.

With the rapid development of Generative Artificial Intelligence (GAI) technology, Generative Diffusion Models (GDMs) have shown significant empowerment potential in the field of wireless networks due to advantages, such as noise resistance, training stability, controllability, and multimodal generation. Although there have been multiple studies focusing on GDMs for wireless networks, there is still a lack of comprehensive reviews on their technological evolution. Motivated by this, we systematically explore the application of GDMs in wireless networks. Firstly, we identify the core challenges of wireless networks and argue why GDMs are uniquely suited to address them. We then introduce the mathematical principles of GDMs and representative models. Furthermore, we organize our comprehensive review through a structured taxonomy that categorizes GDM-based schemes into the sensing, transmission, and Applications, complemented by a security plane. For each representative scheme, we analyze its innovative points, the role of GDMs, strengths, and weaknesses. Ultimately, we extract key challenges and provide potential solutions, with the aim of providing directional guidance for future research in this field.

We introduce a new analytical framework, developed based on the spatial network calculus, for performance assessment of Low Earth Orbit (LEO) satellite networks. Specifically, we model the satellites' spatial positions as a strong ball-regulated point process on the sphere. Under this model, proximal points in space exhibit a locally repulsive property, reflecting the fact that intersatellite links are protected by a safety distance and would not be arbitrarily close. Subsequently, we derive analytical lower bounds on the conditional coverage probabilities under Nakagami-m and Rayleigh fading, respectively. These expressions have a low computational complexity, enabling efficient numerical evaluations. We validate the effectiveness of our theoretical model by contrasting the coverage probability obtained from our analysis with that estimated from a Starlink constellation. The results show that our analysis provides a tight lower bound on the actual value and, surprisingly, matches the empirical simulations almost perfectly with a 1 dB shift. This demonstrates our framework as an appropriate theoretical model for LEO satellite networks.

Holographic communication is a key enabler for 6G immersive services, but its massive data volume poses significant challenges to traditional communication systems. Semantic communication offers a promising solution. However, integrating it with HC, designing a joint source–channel coding (JSCC) model for holographic information extraction and reconstruction, and evaluating the semantic-to-holographic user quality of experience (QoE) remain major challenges. To address these challenges, we propose a QoE-driven semantic transmission framework for holographic communication (ESHC) and systematically optimize its three core components: JSCC module, QoE module, and multimodal resource allocation module. We optimized the design of these three core components individually. First, addressing the limitation that existing JSCC models cannot adapt to a wide range of signal-to-noise ratios with a single model, we developed a voxel importance function and designed a 3D Attention Feature (3D-AF) module based on it. Compared to existing methods, the JSCC model integrated with 3D-AF achieved a 30% performance improvement. Second, we established the first holographic QoE mathematical model incorporating semantic communication factors and proposed a joint optimization problem targeting QoE maximization, involving holographic reconstruction density, computational resources, and semantic communication resource allocation. Finally, for the multimodal resource allocation module, we developed an Online Decision Transformer-based resource allocation algorithm (ODT-RA) to efficiently solve this optimization problem, enabling online optimal resource allocation. Simulation results demonstrate that under identical conditions, ODT-RA achieves 30%-40% higher performance than advanced reinforcement learning methods.

The sixth generation (6G) network is expected to deploy larger multiple-input multiple-output (MIMO) arrays to support massive connectivity, which will increase overhead and latency at the physical layer. Meanwhile, emerging 6G demands such as immersive communications and environmental sensing pose challenges to traditional signal processing. To address these issues, we propose the “semantic-aware MIMO” paradigm, which leverages specialist models and large models to perceive, utilize, and fuse the inherent semantics of channels and sources for improved performance. Moreover, for representative MIMO physical-layer tasks, e.g., random access activity detection, channel feedback, and precoding, we design specialist models that exploit channel and source semantics for better performance. Additionally, in view of the more diversified functions of 6G MIMO, we further explore large models as a scalable solution for multi-task semantic-aware MIMO and review recent advances along with their advantages and limitations. Finally, we discuss the challenges, insights, and prospects of the evolution of specialist models and large models empowered semantic-aware MIMO paradigms.

Non-terrestrial networks (NTNs) are advancing with multi-shell dense satellite constellations, offering sufficient capacity and enabling system-level integration with terrestrial networks (TNs). In this context, adding and maintaining secondary satellite links via dual connectivity (DC) in integrated TN-NTN systems enhances service quality for users underserved by TN coverage. However, satellites operating in diverse orbits and altitudes exhibit heterogeneous signal characteristics and coverage windows, posing challenges for maintaining effective DC configuration under time-varying TN-NTN coverage overlap. To address this, we propose a mobility-aware DC configuration and continuity management framework for dynamic TN-NTN environments. First, all DC admissions are modeled as an optimal user equipment (UE)-satellite bipartite graph matching problem for spatial NTN link allocation. The graph is then temporally extended to capture satellite mobility and handover (HO) opportunities. A graph attention network (GAT) generates UE-specific mobility-aware edge weights, enabling path-level DC configuration that balances instantaneous throughput gains with long-term service continuity. System-level evaluations with practical satellite constellations demonstrate the proposed framework significantly outperforms existing repeated and snapshot-wise DC configuration methods in both service rate and signaling efficiency. By fully exploiting TN-NTN coverage asymmetry and mobility dynamics, the proposed scheme ensures robust and sustained DC performance across diverse scenarios.

The direct-to-device (D2D) satellite network is an important 6G evolution direction to enable seamless ubiquitous connectivity. However, the network faces critical handover challenges due to high satellite mobility and wide beam footprints. Conventional handover strategies, mostly designed for terrestrial networks, may encounter excessive co-channel interference (CCI) and service degradation in the satellite environments. To address the issue, this paper introduces a novel handover optimization method to perform dynamic adjustment of an important parameter called elevation angle threshold (EAT) from an interference-aware perspective. Explicitly, we first analyze the trade-off between satellite visibility and CCI. Then, we propose a numerical algorithm to determine the optimal EAT that can achieve seamless coverage with CCI. We validate our method using a customized D2D LEO satellite network simulator in the Network Simulator (NS-3). The results demonstrate that the EAT optimization significantly reduces packet loss and hence enhances handover reliability. The work highlights the importance of interference-aware handover design for improving service continuity in future D2D satellite networks.

As the real propagation environment becomes in creasingly complex and dynamic, millimeter wave beam prediction faces huge challenges. However, the powerful cross modal representation capability of vision-language model (VLM) provides a promising approach. The traditional methods that rely on real-time channel state information (CSI) are computationally expensive and often fail to maintain accuracy in such environments. In this paper, we present a VLM-driven contrastive learning based multimodal beam prediction framework that integrates multimodal data via modality-specific encoders. To enforce cross-modal consistency, we adopt a contrastive pretraining strategy to align image and LiDAR features in the latent space. We use location information as text prompts and connect it to the text encoder to introduce language modality, which further improves cross-modal consistency. Experiments on the DeepSense-6G dataset show that our VLM backbone provides additional semantic grounding. Compared with existing methods, the overall distance-based accuracy score (DBA-Score) of 0.9016, corresponding to 1.46

Multi-functional uncrewed aerial vehicles (UAVs) are gaining attention for supporting multi-task missions such as delivery and federated learning (FL). However, asynchronous data uploading hinders synchronous model aggregation, and user behavior often exhibits group-level personalized data. To address these challenges, we propose a joint model training and UAV path optimization framework. We formulate a UAV network system model that enables the UAV to function as both a delivery carrier and an FL parameter server. We design a two-tier training framework to mitigate group heterogeneity and model staleness. Specifically, the inner tier conducts synchronous intra-group personalized updates, whereas the outer tier employs staleness-aware asynchronous global aggregation. To reduce model staleness during cargo delivery, we formulate a path optimization problem incorporating training window duration, and we solve it via deep reinforcement learning. Numerical results show that the proposed framework outperforms the static and random path baselines, reducing model staleness by 20.75% and 42.67% while improving convergence rates by 20.15% and 46.50%, respectively. In a group-level non-independent and identically distributed (non-IID) setting, the training framework achieves a 5.28% test accuracy improvement over an asynchronous, device-level personalized learning baseline.

This paper investigates the receiver design for clipped orthogonal time frequency space (OTFS) systems, where the user devices are equipped with power amplifiers (PAs) with low dynamic range. To improve power efficiency, the PAs have to work near the saturation points, which leads to unknown nonlinear distortions, thus making the signal detection more challenging. To solve this problem, techniques like intentional clipping or pre-distortion are adopted, thus approximating the outputs of the PAs as clipped signals. To further compensate for the unknown time-varying multipath channel and the clipping distortion at the receiver, the channel and clipping amplitude (CA) estimation, channel tracking, and signal detection are studied in this paper. Firstly, a receiver framework is developed for clipped OTFS. Secondly, by adopting the sparsity of the delay-Doppler (DD) domain channel and the piecewise linearized signal model with respect to CA, a novel sparse Bayesian learning (SBL) based joint channel and CA estimation scheme is proposed. Then, to further reduce the estimation error and bit error rate, a Kalman filter (KF) based channel tracking scheme and a minimum mean square error decision feedback blockwise equalization (MMSE-DFBE) based detection scheme are proposed. These two schemes are integrated in an expectation maximization (EM) based iterative tracking and detection algorithm. Finally, numerical simulations are conducted to demonstrate the superiority of the proposed schemes in terms of both estimation error and bit error rate.

Large AI models (LAMs) are primed to take wireless systems beyond bit-level reliability toward genuine meaning-aware operation, a shift that aligns squarely with the vision of 6G semantic communication. In this article, we develop a three-axis lens - technology integration, security assurance, and computational efficiency - that clarifies why and how LAMs strengthen the semantic stack. Guided by these insights, we outline an end-to-end architecture in which a single LAM powers multi-source joint coding and decoding, infuses physical-layer security with generative intelligence, and supports lightweight, distributed training for edge deployment. A sentiment- analysis case study illustrates the design's ability to keep inference robust under noisy channels, while a forward look at drone networks, embodied robots, and immersive media shows its breadth of impact. We close by highlighting open problems in data curation, model transparency, and resource-constrained deployment, setting a concise research agenda for "understand-everything" 6G networks.

With the growing adoption of low-altitude economy (LAE), such as urban air mobility and smart logistics, federated learning (FL) has become a key technology for enabling distributed intelligence while preserving data privacy. However, edge clients in such systems often suffer from limited resources and heterogeneous data, making them especially vulnerable to covert backdoor attacks. These threats can compromise critical functions such as navigation control or data confidentiality. Existing backdoor methods typically target specific clients, but their performance significantly degrades on edge clients due to insufficient data and computation power. To address this challenge, we propose LBKD, a novel backdoor attack strategy that combines large language models (LLMs) with bidirectional knowledge distillation. In our method, the attacker first obtains the target client's model and data, then uses LLMs guided by poisoning and data generation instructions to produce both poisoned and synthetic samples. During training, a more expressive distillation model engages in two-way knowledge exchange with the client model, enhancing its ability to learn and retain the backdoor pattern. The final poisoned model is submitted to the server, successfully embedding the backdoor into the global model. Experimental results show that LBKD significantly improves attack success rates (ASR), resulting in up to a 75.4% improvement over existing methods. Even under defense mechanisms, the ASR remains above 90% in some cases, demonstrating the method's effectiveness and robustness on edge clients.

With the explosive advancement of unmanned aerial vehicles (UAVs), the security of efficient UAV networks has become increasingly critical. Owing to the open nature of its communication environment, illegitimate malicious UAVs (MUs) can infer the position of the source UAV (SU) by analyzing received signals, thus compromising the SU location privacy. To protect the SU location privacy while ensuring efficient communication with legitimate receiving UAVs (RUs), we propose an Active Reconfigurable Intelligent Surface (ARIS)-assisted covert communication scheme based on virtual partitioning and artificial noise (AN). Specifically, we design a novel ARIS architecture integrated with an AN module. This architecture dynamically partitions its reflecting elements into multiple sub-regions: one subset is optimized to enhance communication between the SU and RUs, while the other subset generates AN to interfere with the localization of the SU by MUs. We first derive the Cram & eacute;r-Rao Lower Bound (CRLB) for localization with received signal strength (RSS), based on which, we establish a joint optimization framework for communication enhancement and localization interference. Subsequently, we derive and validate the optimal ARIS partitioning and power allocation under average channel conditions. Finally, tailored optimization methods are proposed for the reflection precoding and AN design of the two partitions. Simulation results validate that, compared to baseline schemes, the proposed scheme significantly increases the localization error of MUs by approximately 37.65% with only a 3.69% reduction in the communication rate between the SU and RUs, thereby effectively protecting the SU location privacy.

As large language models (LLMs) are increasingly deployed at the network edge to provide pervasive generative AI services, decentralized federated learning (DFL) provides a vital mechanism for privacy-preserving, domain-specific fine-tuning through peer-to-peer exchanges of parameter-efficient updates. However, the dynamic nature of practical decentralized edge networks, where devices may dynamically join or leave the collaborative training process, requires the system to continuously adapt to new data while selectively removing prior contributions. This correction process remains a significant bottleneck, as individual device updates become deeply entangled within the global fine-tuned parameters. To address this challenge, we propose a priority-aware learning-unlearning correction framework based on orthogonal LoRA that can enhance the knowledge evaluation through topology adjustment. Specifically, we first design an orthogonal LoRA mechanism that yields post-training contribution coordinates, enabling history-free projection addition and deletion in response to membership changes. We then analyze the correction bottleneck and develop a priority-aware policy that selects among topology refinement, local correction, proximal damping, and synchronization scheduling according to the dominant residual term. A resource allocation algorithm is further developed to allocate limited communication across layer groups, prioritizing the primary bottlenecks within per-round wireless constraints. Experiments demonstrate that the proposed framework achieves robust post-event correction for both device join and leave events and validate that different residual regimes necessitate distinct correction actions.

In Federated Learning (FL), with parameter aggregated by a central node, the communication overhead is a substantial concern. To circumvent this limitation and alleviate the single point of failure within the FL framework, recent studies have introduced Decentralized Federated Learning (DFL) as a viable alternative. Considering the device heterogeneity, and energy cost associated with parameter aggregation, in this paper, the problem on how to efficiently leverage the limited resources available to enhance the model performance is investigated. Specifically, we formulate a problem that minimizes the loss function of DFL while considering energy and latency constraints. The proposed solution involves optimizing the number of local training rounds across diverse devices with varying resource budgets. To make this problem tractable, we first analyze the convergence of DFL with edge devices with different rounds of local training. The derived convergence bound reveals the impact of the rounds of local training on the model performance. Then, based on the derived bound, the closed-form solutions of rounds of local training in different devices are obtained. Meanwhile, since the solutions require the energy cost of aggregation as low as possible, we modify different graph-based aggregation schemes to solve this energy consumption minimization problem, which can be applied to different communication scenarios. Finally, a DFL framework which jointly considers the optimized rounds of local training and the energy-saving aggregation scheme is proposed. Simulation results show that, the proposed algorithm achieves a better performance than the conventional schemes with fixed rounds of local training, and consumes less energy than other traditional aggregation schemes.

Traditional machine learning techniques have achieved great success in improving data-rate performance and reducing latency in millimeter wave (mmWave) communications. However, these methods still face two key challenges: (i) their reliance on large-scale paired data for model training and tuning, which limits performance gains and makes beam predictions outdated, especially in multi-user mmWave systems with large antenna arrays, and (ii) meta-learning (ML)-based beamforming solutions are prone to overfitting when trained on a limited number of tasks. To address these challenges, we first propose a memristor-based meta-learning (M-ML) framework to expedite spatial and temporal domain beam prediction. Notably, the M-ML framework generates optimal initialization parameters during the training phase, providing a strong starting point for adapting to unknown environments during the testing phase. By leveraging memory to store key data, M-ML ensures the predicted beamforming vectors are well-suited to episodically dynamic channel distributions, even when testing and training environments do not align. Afterwards, we propose a Gaussian noise-based regularized meta-learning framework to model the uncertainty in the training data and improve its stability and accuracy in complex environments. Simulation results manifest that our approaches deliver high prediction accuracy in new environments, without relying on large datasets. Moreover, M-ML enhances the model’s generalization ability and adaptability.