Ensuring end-to-end quality of service (QoS) for high-value services in satellite networks is challenging due to dynamic network topologies, varying QoS requirements, and the complex resource allocation across satellite beams and inter-satellite links. To this end, we propose a satellite traffic engineering framework with deterministic QoS (SatTED) by jointly optimizing resource allocation across access and bearer subnets.To tackle the complexity of joint scheduling, SatTED adopts a hierarchical logic-based benders decomposition (LBBD) architecture that coordinates access and bearer subnet resources. The master problem optimizes service admission and satellite selection via binary integer programming, while the subproblem handles routing and bandwidth allocation through linear programming relaxation. Key innovat-ions include scenario-cognizant Benders feasibility cuts to accelerate convergence and a critical constraint link preprocessing (CCLP) mechanism that reduces subproblem complexity by 5.15× in large-scale networks. In simulations on a 220-satellite network with 1 000 flows, SatTED improves total service payoff by 32% and increases high-value flow completion rates by 22%.

Space-terrestrial integrated networks (ST-IN) require a trustworthy and auditable environment for multi-party spectrum sharing under dynamic and heterogeneous conditions. Blockchain, as a decentralized ledger, exhibits promising properties such as transparency and tamper resistance, making it a potential enabler for such scenarios. However, conventional blockchain-based approaches tightly couple strategy execution with transaction consensus, resulting in excessive overhead and poor adaptability to fast-changing spectrum semantics. To address these issues, this paper presents a spectrum-semantics-driven meta-consensus framework built upon a directed acyclic graph (DAG) mainchain architecture. By decoupling policy optimization from on-chain coordination and leveraging semantic representations of spectrum states for meta-level consensus and policy migration, the framework enables agile and scalable spectrum sharing across dynamically clustered network agents. Simulation results verify that the proposed design significantly enhances spectrum utilization and adaptability while maintaining decentralized transparency and auditability in large-scale STIN environments.

As the development of 6G accelerates, non-terrestrial networks (NTN) are emerging as a critical component to ensure seamless global coverage communication. This paper systematically reviews the evolution of the third-generation partnership project (3GPP) NTN standardization, from initial discussions in Release 15 to the in-depth optimizations in Release 19 and beyond. Through a detailed gap analysis, we identify key technical and industrial challenges across the wireless, network, and terminal domains that hinder the realization of a fully integrated space-terrestrial network. To address these challenges, we propose a 6G-oriented holistic architecture composed of three segments and three functional layers. We further outline essential enabling technologies, including networking technologies and intelligent resource management technologies. To validate the feasibility and effectiveness of the proposed architecture, we present a case study on integrated sensing and communication in high-speed mobility scenarios. Simulation results demonstrate significant performance gains in robustness, sensing accuracy, and adaptability compared to conventional approaches. Our findings establish a solid foundation for future research and standardization of 6G integrated networks, aiming to achieve intelligent, ubiquitous, and resilient communication infrastructures across space, air, and ground domains.

With the rapid development of Internet of Things (IoT) service, the provision of network connection is vital for data transmission of widely deployed IoT devices. To provide the global coverage for IoT devices, non-terrestrial networks (NTN) has its inherent advantages with the help of satellite communication. In this paper, we propose a Q-learning based intelligent access strategy to make access control for IoT devices in NTN scenario. With satellite assistance, access control is optimized so that IoT devices can get the connection efficiently. IoT devices interact with the environment and learn gradually to find the proper RACH type, preamble and random access (RA) slot for success access. Besides, a novel RACH procedure is also designed with flexible access type selection between 2-step and 4-step RACH. With the corporation between IoT devices and network, access congestion can be avoided. Simulation results validate that the effectiveness of the proposed access strategy in terms of access efficiency, collision rate and access latency.

Satellite-terrestrial integrated networks (S-TINs) are a key enabler for ubiquitous coverage in 6G communication services. However, the satellite-terrestrial resources exhibit multi-dimensional heterogeneity and inherent conflicts, and the rapid topology variations caused by the high-speed motion of low earth orbit (LEO) satellites lead to the difficulty of maintaining a stable mapping of satellite-terrestrial resources. This dynamic nature ultimately reduces the overall resource utilization efficiency. In this paper, we propose a heterogeneous graph cooperative representation approach for satellite-terrestrial resources and a joint optimization method of transmission-computation resources. Firstly, we construct a heterogeneous graph that achieves mapping between multidimensional resources, dynamic topology, and conflict constraints through typed nodes and edges, where resource cooperativeness is explicitly encoded. Secondly, an STIN transmission-computation model is constructed, and an optimization problem is formulated to jointly resolve conflicts between four objectives. Finally, the proposed many-objective double deep Q-network (DDQN) algorithm achieves the cooperative strategy optimization of task transmission-computation scheduling globally. Simulation experiments show that the proposed algorithm improves the overall resource utilization by up to 11.7% under various access points (APs) and user sizes. Meanwhile, the performance is more stable compared with five algorithms, including deep Q-network (DQN), and a Lyapunov-based optimization method (LyaOpt).

The rapid advancement of 6G communication has drawn significant attention to non-terrestrial networks (NTN), where accurate modeling of clutter loss (CL) is essential for efficient deployment and system optimization. This paper conducts multi-band channel measurements at 6.5 GHz, 10.2 GHz, 11.8 GHz, and 14.2 GHz to investigate the frequency and elevation angle dependencies of CL. To address the limited physical interpretability of existing standard models, a semi-deterministic approach is proposed based on the single knife-edge diffraction theory. The modeling results and analysis show that the proposed model has lower prediction errors than the standard model in complex environments with building and vegetation obstructions, with a simple formula that effectively captures the variations in CL with frequency and elevation angle. This contributes valuable insights for the design and implementation of NTN communication systems.

The rapid development of the low-altitude economy has imposed unprecedented demands on wireless infrastructure to accommodate large-scale drone deployments and facilitate intelligent services in dynamic airspace environments. However, unlocking its full potential in practical applications presents significant challenges. Traditional aerial systems predominantly focus on air-ground communication services, often neglecting the integration of sensing, computation, control, and energy-delivering functions, which hinders the ability to meet diverse mission-critical demands. Besides, the absence of systematic low-altitude airspace planning and management exacerbates issues regarding dynamic interference in three-dimensional space, coverage instability, and scalability. To overcome these challenges, a comprehensive framework, termed low-altitude wireless network (LAWN), has emerged to seamlessly integrate communication, sensing, computation, control, and air traffic management into a unified design. This article provides a comprehensive overview of LAWN systems, introducing LAWN system fundamentals and performance evaluation metrics. Subsequently, we delve into the evolution of functional designs and review critical concerns surrounding privacy and security in the open-air network environment. We survey advanced artificial intelligence techniques that enhance LAWN functionality and enable increasingly autonomous operations. Finally, we present the cutting-edge developments in airspace structuring, air traffic management, and path planning, providing insights to facilitate the practical deployment of LAWNs.

Implementing check node (CN) update ba-sed on the minimum value (MV) and second MV of incoming message magnitudes is crucial for Min-Sum Algorithms (MSAs). In the category of bit-serial implementations, existing schemes suffer from decoding performance degradation, large hardware areas, and/or long latency. In this paper, we propose two efficient CN update functions based on the MV and an approximate second MV, and design bit-serial architectures to implement them. Simulation results show that our functions exhibit the minimum decoding performance degradation compared to the existing functions using approximate second MVs. Moreover, the application-specific integrated circuits (ASIC) implementation results demonstrate the advantages of our architectures in terms of area, latency, etc.

Wideband extremely large-scale multiple-input multiple-output (XL-MIMO) is a promising technology for sixth-generation (6G) communications due to its high spectral efficiency and data rates. However, as the number of antennas increase in XL-MIMO, users are more likely to be in the near-field region, where beam training becomes more complex. Additionally, the expanding bandwidth introduces beam split and grating lobe effects. In this paper, we analyze the conditions for grating lobe generation, demonstrating the relationship between the grating lobe angle and bandwidth. As the bandwidth increases, a large portion of the visible angle range will produce grating lobes. The analysis shows that the grating lobe location always lies on the distance ring of the main lobe, and the grating lobe distance can be uniquely determined by the grating lobe angle on a single distance ring. Then we propose a wideband grating lobe beam training method based on controllable grating lobe generation mechanism, which applies to both the angle and distance dimensions to cover the user range. Simulation results show that this method achieves comparable rate performance with low overhead.

In this paper, we propose a novel cyclic redundancy check (CRC)-aided method to improve the energy efficiency of distribution matching (DM) algorithms based on fixed empirical distribution codebooks. The core design concept is to map a subset of a DM codebook to the entire codebook. The mapping is identified by a binary vector that is convolved during the CRC encoding; thereby avoiding any additional overhead in a CRC-aided system. At the receiver (RX), the CRC check not only performs error detection but also identifies the mapping of the transmitted symbol sequences to the original input of the DM. With the information delivered by a quarter-sized codebook, fewer occurrence of high-energy symbols effectively reduces the average symbol energy and the rate-loss. The proposed method can be seamlessly integrated into any DM algorithm that uses fixed empirical distribution codebooks. We demonstrate its implementation in a polar-coded probabilistic amplitude shaping (PAS) system with CRC-aided successive cancellation list decoding. Using this architecture, we show an energy efficiency improvement of up to $26\%$ and a signal-to-noise ratio improvement of up to $0.8$ dB at a fixed target frame error rate of $10^{-3}$.

Integrated sensing and communication (ISAC) is emerging as a key technology for future cellular networks. This paper focuses on the collaborative ISAC mechanism between base stations (BSs) and users using reference signals (RSs). The main challenges we address are the joint optimization of downlink communication and sensing resources, the selection of users as sensing anchors, as well as the fusion of estimation data between the BS and users. We formulate the collaborative ISAC problem as a multi-objective programming framework, which can balance system performance in sensing and individual benefits in communication. Particularly, to ensure fairness, we propose minimizing the largest sensing age across all users. On this basis, we put forward an efficient solution algorithm that enables a low-complexity computation of the Pareto front when it exists. Simulation results demonstrate that the proposed collaborative ISAC mechanism is capable of efficiently enhancing the system's sensing capacity while ensuring fairness in user scheduling for sensing.

In the 6G environment, addressing challe-nges like missing data, demodulation errors, and off-grid issues during target parameter estimation is a significant hurdle for integrated sensing and communication (ISAC) systems. In the ISAC framework, a commonly used method for parameter estimation is compressive sensing. However, it often struggles with off-grid problems in continuous parameter estimation. In contrast, the atomic norm has been proven effective in overcoming these off-grid issues, making it a more suitable approach for continuous parameter estimation. In this paper, we investigate the application of atomic norm in ISAC and propose an ISAC model based on orthogonal frequency division multiplexing (OFDM) for parameter estimation. We utilize the atomic norm under conditions of incomplete data and demodulation errors. To enhance the convergence speed and accuracy of our algorithm, we implement the alternating direction method of multipliers (ADMM) for iterative processing. We refer to this algorithm as ANMI. Building on this foundation, we develop a deep unfolding network algorithm, ANMI-ADMM-Net, which further mitigates the impact of missing data and demodulation errors on target parameter estimation by training optimal parameters. Experimental results demonstrate that our proposed ANMI and ANMI-ADMM-Net accurately estimate target parameters even in the presence of missing data and demodulation errors, exhibiting superior precision and robustness compared to traditional methods.

This paper presents a circularly polarized (CP) multiple-input multiple-output (MIMO) antenna with wideband and compact size characteristics. The proposed MIMO antenna consists of two wideband dual-CP radiating elements, which are positioned in proximity and decoupled by shorting vias. Accordingly, only two radiating elements are required for a 4-port MIMO array. It distinguishes the proposed design approach from the others, in which 4-port MIMO antennas commonly need four radiating elements. The measured results confirm the wideband performance of the proposed antenna with 10 dB isolation operating bandwidth of 15% (4.68–5.44 GHz), and 15 dB isolation operating bandwidth of 8.2% (4.68–5.08 GHz). Besides, 4-port MIMO antenna can be realized with compact dimensions of 0.84 $\lambda$ $\times$ 0.46 $\lambda$ $\times$ 0.05 $\lambda$ at 4.68 GHz. In comparison with the related CP MIMO antennas, the proposed antenna can work with a higher number of operating ports while achieving smaller overall dimensions. Besides, large operating bandwidth is also another advantage of the proposed work compared to the others.

The integration of hovering unmanned ae-rial vehicles (UAVs) with modulating retro-reflector (MRR) free-space optical (FSO) communication systems offers a promising solution for flexible, high-bandwidth links. This paper presents a comprehensive performance analysis of a UAV-based MRR FSO system employing a multiple-quantum-well (MQW) modulator. We develop a composite channel model that simultaneously accounts for atmospheric turbulence, non-zero boresight pointing errors, and angle-of-arrival fluctuations. Closed-form expressions for key performance metrics—including bit error rate, outage probability, and average channel capacity—are derived. Numerical results validate the analytical framework and provide insights into the effects of various system parameters. This study establishes a theoretical foundation and offers practical guidance for the design of efficient UAV-based MRR FSO systems.

The time-division duplex (TDD) large-scale antenna array system using analog-digital hybrid beamforming precoding technique has a very broad application prospect in the field of 5G/6G mobile communications. However, the gain imbalance between radio frequency (RF) channels introduced by RF devices leads to problems such as decreased channel estimation accuracy, deviation of precoding beam direction, and high side-lobe level. Meanwhile, the RF transceiver gain asymmetry destroys the symmetry of the uplink and downlink channels of the system, leading to a sharp performance degradation of the TDD system that employs the uplink channel information for the downlink beamforming precoding. In this paper, we propose an innovative and low-cost hardware-based self-calibration method to solve the above two problems simultaneously. The calibration is divided into two stages: digital beamforming chain calibration (DBCC) and analog beamforming transceiver calibration (ABTC). Simulation results demonstrate that this approach simultaneously enhances the accuracy of both channel estimation and beamforming precoding, significantly improving the achievable data rate of existing precoding schemes. Furthermore, it offers considerable computational complexity advantages over current over-the-air (OTA) approaches.

In this paper, we explore a secure downlink transmission system and propose a simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) assisted cooperative jamming scheme denoted by STAR-CJ, where a multi-antenna base station (BS) maintains secure communications with a single-antenna user via the STAR-RIS with the help of a jammer in the presence of an eavesdropper (Eve). Notably, since the BS and the jammer are positioned on opposite sides of the STAR-RIS, the STAR-RIS operates in dual-sided mode. We devise a secrecy rate maximization problem by jointly designing the transmission and reflection coefficients of the STAR-RIS and the beamforming vector of the BS and the jammer. In order to solve this problem, we utilise an alternating optimization (AO) algorithm, which effectively decouples the problem into two sub-problems, subsequently employing the successive convex approximation (SCA) method for resolution. Simulation results demonstrate that the proposed STAR-CJ scheme is superior to benchmark schemes including conventional RISs scheme (C-RISs) and without jammer scheme (STAR-WJ) in terms of secrecy rate.

The rapid evolution of wireless networks presents unprecedented challenges in managing diverse wireless tasks. These challenges underscore the need for AI-native solutions in next-generation networks. In this article, we propose WirelessAgent, a novel framework that harnesses large language models (LLMs) to create autonomous AI agents for diverse wireless network tasks. We first define a general framework for WirelessAgent, supported by key components and principles in AI agents. Then, we introduce a basic usage to implement the WirelessAgent based on agentic workflows and the LangGraph architecture. We demonstrate the effectiveness of WirelessAgent through a comprehensive case study on the network slicing task. Our numerical results show that WirelessAgent achieves $44.4\%$ higher bandwidth utilization than the Prompt-based method, while performing only $4.3\%$ below the Rule-based optimality. Notably, WirelessAgent delivers near-optimal network throughput across diverse network scenarios. These underscore the framework's potential for intelligent and autonomous control in next-generation networks. The code is available at https://github.com/jwentong/WirelessAgent_R1.

In vehicular networking applications, Mobile Ad hoc networks (MANETs) enable dynamic, infrastructure-free connectivity for multi-node mobile scenarios. Frequent topology changes, however, challenge routing protocols in delivering quality of service (QoS) for diverse applications. We propose inhanced Ad-hoc on-demand distance vector multipath (I-AOMDV), an enhanced multipath routing protocol using a primary-backup strategy to meet stringent QoS demands in dynamic vehicular environments. Whereas AOMDV relies on hop count, I-AOMDV integrates hop count, bandwidth, and path stability into a QoS-aware framework for optimized path selection. Extensive NS-2 simulations demonstrate that I-AOMDV surpasses AOMDV under high mobility, improving data packet delivery by up to 22% and cutting voice service latency by 8–15%, even in high-mobility scenarios. By addressing vehicular networking needs for latency, reliability, and bandwidth, I-AOMDV delivers a scalable, efficient routing solution.

The increasing complexity of future networks demands intelligent, scalable, and adaptive management solutions. Digital twin network (DTN) provides a high-fidelity replica of the physical network for monitoring and optimization, but faces significant limitations, including complex modeling, high synchronization overhead, and limited scalability. Foundation models (large pre-trained AI models) offer powerful semantic understanding and reasoning abilities, yet suffer from high training costs, risks of generating hallucinations, and limited interpretability. To address these challenges, this paper proposes an integrated architecture that combines DTN with foundation models, leveraging their complementary strengths. DTN ensures fidelity and domain-specific modeling, and acts as a validation platform to help facilitate the training and verification of network foundation models. Foundation models enable data-driven automation, downstream model generation, and adaptive decision-making. Furthermore, we present use cases related to twin network configuration verification and protocol generation, demonstrating enhanced scalability, efficiency, and intelligence for intelligent networks by bridging foundation models and digital twin.

With the increasing adoption of cloud–ed-ge collaborative computing in delay-sensitive applications, sleep control of edge nodes has become a key approach to reducing operational energy consumption. However, existing schemes have not balanced energy efficiency and performance under edge node sleep control and still lack a joint optimization mechanism for computation offloading and resource allocation. This paper tackles the problem of optimizing energy-efficient computation offloading and resource allocation (CORA) in cloud-edge collaborative computing systems, where edge servers can dynamically enter sleep mode to reduce power consumption. We model the problem as a mixed-integer nonlinear programming formulation, with the objective of minimizing a weighted sum of overall task latency and the cumulative energy consumption of all IoT devices and edge servers. To handle the hybrid nature of discrete and continuous decision variables and the complex system dynamics, we reformulate the problem for each device as a Markov Decision Process and develop a deep deterministic policy gradient with multi-agent algorithm tailored for such hybrid action spaces. Simulation results show that the proposed CORA strategy achieves superior performance compared to three benchmark schemes with reduced latency and energy consumption.

This paper aims to improve energy efficiency (EE) of the integrated access and backhaul (IAB) aerial-terrestrial network, facilitating rapid and adjustable network infrastructure deployment. This is challenging,as interference generated by backhaul and access links degrades network throughput, and power imbalance between these links increases overall energy consumption. To this end, we jointly optimize aerial base station (ABS) deployment, user association, and downlink power allocation for both terrestrial base station and ABSs to maximize network EE. Specifically, using fractional programming, the EE maximization problem is transformed into a subtractive-form parametric problem, and then decomposed into ABS deployment and resource allocation subproblems. A hybrid algorithm combining particle swarm optimization and simulated annealing is proposed to solve the ABS deployment subproblem, determining ABS spatial configurations and updating power allocation given fixed user association. Meanwhile, a dynamic power allocation in response to network load is designed to solve the resource allocation subproblem. Furthermore, considering the quality of service requirements of ground users and the transmit power constraints of base stations, a joint EE optimization algorithm is proposed to enhance the network EE. Simulation results validate the effectiveness of the proposed methods in improving network EE, especially in scenarios involving more deployed ABSs.

Integrated sensing and communication (I-SAC) has emerged as a promising technology to address the critical challenges of energy efficiency and spectrum sharing in joint communication and sensing systems. Particularly in fifth-generation (5G) and beyond networks, the deployment of massive antenna arrays at base stations provides sufficient spatial degrees of freedom for the harmonious coexistence of these dual functionalities. This paper investigates the transmit beamforming design for downlink ISAC systems, where a base station equipped with a uniform linear array (ULA) simultaneously transmits multiplexed communication data streams and dedicated sensing probe signals to achieve joint downlink multiuser communication and target sensing. Unlike existing approaches in prior works, we specifically consider the area surveillance scenario where the sensing function is designed for wide-area monitoring. In this context, the sensing performance is optimized by maximizing the number of scanned beam directions, while strictly guaranteeing the individual quality of service (QoS) requirements for communication users. Given that the maximization of scanned beam directions under practical constraints constitutes a nondeterministic polynomial-time hard (NP-hard) problems, we develop a near-optimal convex relaxation approach, accompanied by rigorous performance analysis. Furthermore, we systematically examine the characteristics of sensing-specific beamformers and derive their fundamental relationships with both communication and sensing channels. Simulation results demonstrate that the proposed scheme achieves significant performance with tractable computational complexity.

Integrated satellite-terrestrial edge computing networks (ISTECNs) have been developed to advance the existing wireless communication systems by combining multi-access edge computing (MEC) with radio access network (RAN) slicing. This paper proposes an adaptive slicing resource optimization (ASRAO) method for satellite-terrestrial edge computing networks, aims to deliver the MEC services with high bandwidth, low latency, broad coverage, and low power consumption by creating four end-to-end isolated RAN slices. First, the resource optimization problem for the four RAN slices into two subproblems, the terrestrial layer (TL) and air-space layer (ASL) optimization subproblems, is designed. Then, a parallel traffic prediction model with feature fusion (PTPFF) is developed by integrating an improved convolutional neural network (ICNN) and a long short-term memory (LSTM) model to conduct predictions of network traffic for different service types. Finally, a two-layer traffic adaptive multi-agent reinforcement learning-based (TAMARL) model is introduced. It adaptively adjusts resource allocation for each slice based on the traffic volume and performs an alternating iterative optimization of the TL and ASL subproblems. The experimental results demonstrate that the proposed ASRAO can enhance the throughput by 38%, reduce the average latency by 33%, improve coverage by 20%, and decrease the average energy consumption by 21.5%.