Multi-beam satellites (MBS) exploit multiple spot beams and frequency reuse to achieve high spectral efficiency and flexible coverage, which are key characteristics of modern satellite communication systems. Among them, beam hopping (BH) leverages phased-array antennas to steer onboard beams and employs time-division multiplexing (TDM) to dynamically schedule illumination patterns in the time domain, with electronic beam position switching enabling rapid reconfiguration. However, existing works exhibit two critical limitations: (1) they rely on global decision strategies that incur high inference complexity, which further increases with the number of beams and ground cells, making it difficult to balance performance and computational complexity; (2) in pursuing high spectral reuse, they overlook beam overlap and co-frequency interference (CFI), and thus cannot effectively mitigate interference without sacrificing spectral efficiency. To address these challenges, this paper proposes a cell clustering-based BH (CCBH) algorithm with low complexity and interference avoidance. Specifically, we propose a region-growing cell-clustering method based on user demand load balancing in which each beam independently serves a cell cluster, and full-frequency reuse across beams maximizes spectral efficiency. In addition, based on the centralized training and decentralized execution (CTDE) paradigm, we propose a cooperative multi-agent soft actor–critic (SAC) framework with parameter sharing and centralized training, called CoopMASAC-PSCT. Among them, an SAC agent is deployed for each beam; Specifically, all actor networks share the same architecture and parameters, and the execution phase remains decentralized, with each actor making decisions solely on its local observations; To prevent inter-cluster interference, the shared global reward integrates system throughput, queueing delay fairness, and an interference-penalty term; Moreover, only a single global critic network is employed, which accesses the joint observations and actions of all agents during training, thus balancing individual beam performance with influence between all beams. Simulation and comparative analyses demonstrate that CCBH delivers better performance while significantly reducing inference complexity.

Massive multiple-input and multiple-output (MIMO) systems with orthogonal frequency division multiplexing (OFDM) are foundational for downlink multiuser (MU) systems in future wireless networks, for their ability to enhance spectral efficiency and support a large number of users. However, high user density intensifies severe inter-user interference (IUI) and pilot overhead. Consequently, existing blind and semi-blind channel estimation (CE) and signal detection (SD) algorithms suffer performance degradation and increased complexity, and are further challenged by frequency-selective channels with high-order modulation demands. To this end, this paper proposes a novel semi-blind joint channel estimation and signal detection (JCESD) method. Specifically, the proposed approach employs a hybrid precoding architecture to suppress IUI. Furthermore, we formulate JCESD as a non-convex optimization exploiting constellation affine invariance. A few pilots are used to achieve coarse estimation for initialization and ambiguity resolution. For high-order modulations, a data augmentation mechanism utilizes the symmetry of quadrature amplitude modulation (QAM) constellations to increase the effective number of samples. To address frequency-selective channels, CE accuracy is then enhanced via an iterative refinement strategy that leverages improved SD results. Simulation results demonstrate an average throughput gain of 11% over pilot-based methods in MU scenarios, highlighting its potential for improving spectral efficiency.

Orthogonal time frequency space (OTFS) waveform has emerged as a promising technology for low-earth orbit satellite (LEO) communications owing to its capacity in mitigating Doppler effects. However, the OTFS modulation in LEO communications with massive multiple-input multiple-output (MIMO) techniques suffer from enormous training overhead due to the large number of satellite antennas. In this paper, we propose a hybrid model-data driven channel estimation scheme for massive MIMO-OTFS LEO satellite communications. We first derive the input-output relationship for massive MIMO-OTFS and formulate the channel estimation as a sparse signal recovery problem. Subsequently, we unroll the orthogonal approximate message passing (OAMP) algorithm into a model-driven deep unrolled network (DUN) to recover the sparse channel. On this basis, we design a side information and attention-enhanced OAMP network (SA-OAMPNet), which incorporates a data-driven self-attention mechanism into each iteration of unrolled OAMP network to further exploit the sparsity across the delay-Doppler-angle domain. Simulation results demonstrate that the proposed DUNs outperform conventional algorithms and deep learning methods. Remarkably, the proposed SA-OAMPNet reduces pilot overhead by about 30% without significant degradation in channel estimation accuracy.

The development of satellite communication is driving the demand for device identification that can respond to security threats from illegal uplink terminals. Specific emitter identification (SEI) is a promising technology of physical layer device identification. Deep learning (DL)-based SEI can effectively learn features for distinguishing emitters through a large number of samples. In the satellite communication scenario, due to the high cost of labeling, wireless signals often face the dilemma of few labeled samples, leading to a decrease in recognition accuracy of existing methods. Therefore, we introduce an innovative FS-SEI method leveraging multi-domain fusion and metric learning (MDF-ML), eliminating the reliance on an auxiliary dataset. Specifically, MDF-ML is proposed to explore additional implicit samples within the sample space, enhancing generalization performance through multi-domain representation and phase rotation. It then uses metric learning to constrain feature distances in the feature space, thereby boosting discriminability. Our simulation results demonstrate that the accuracy of our proposed SEI method exceeds 90%, which outperforms the existing method by 24.21% under 5 sample shots.

In space-air-ground integrated systems, radar signal analysis is crucial for effective spectrum management. In recent years, time-frequency transforms (TFT) have gained significant attention for radar signal detection and identification. However, challenges such as cross-term interference and low signal-to-noise ratio (SNR) limit the effectiveness in multicomponent signal analysis. Therefore, this paper proposes a novel multitask learning-based TFT framework, named One-Stage TFT (OSTFT), which directly generates high-quality time-frequency representations (TFR) from raw in-phase and quadrature signals. OSTFT incorporates a generative network combined with classification and localization tasks to enhance feature extraction and image clarity. Experimental results demonstrate that OSTFT achieves superior performance in TFR quality and radar signal recognition, with an 81.4% detection rate using the You Only Look Once (YOLO) frame-work, outperforming existing TFT methods under various noise conditions. Compared to the best-performing TFR-denoising method, OSTFT improves the detection rate by 2.7%.

The frequency-dependent beamforming technology has demonstrated outstanding performance in two-dimensional beam training for wideband THz systems, owing to its synchronized and high-resolution scanning capabilities. However, extending this technique to three-dimensional beam training poses challenges, as linearly split beams cannot provide comprehensive angular coverage in the candidate area. To this end, we propose a two-stage 3D beam training scheme, consisting of an initial stage and a refinement stage. In the initial stage, the angular coverage of frequency-dependent beamforming is modeled as a controllable rectangular region, and the codebook is designed by arranging this rectangular region within the candidate area. In the refinement stage, frequency-dependent beamforming is densely arranged around the candidate positions estimated in the initial stage to achieve precise localization. Additionally, we design a true-time delay network to facilitate frequencydependent beamforming for uniform planar arrays. Simulation results show that the proposed method approaches near-optimal performance in terms of achievable sum-rate while maintaining acceptable pilot overhead.

As a complement to terrestrial networks, low-Earth orbit (LEO) satellite networks are promising to provide ubiquitous and continuous services. To accommodate the dynamic topology of LEO satellite networks and increasing traffic demands, this article proposes a distributed routing approach to optimize the end-to-end delay relying on the multiagent deep reinforcement learning (MADRL), where each satellite node is deployed with an autonomous agent and makes its real-time next-hop decisions independently with local observation information. To promote the inner cooperation between decentralized agents, a centralized training scheme with a unified load-balancing reward is utilized by adopting a novel multiagent Transformer-MIX architecture. Moreover, to derive a better decision for each agent, we design an attention-involved agent network to capture more hidden information, and a Transformer-based parameter recurrent mechanism to generate the joint action-value function is used to enhance a more stable training. The simulation results indicate that our proposed scheme achieves faster convergence and demonstrates superior performance across several key performance metrics compared to existing benchmark schemes. Specifically, when the intersatellite link (ISL) failure rate in the network reaches 18%, our scheme achieves a reduction in end-to-end delay by 13.6% and an increase in packet successful delivery rate by 5.4% compared to the benchmark schemes.

The collaboration among multiple unmanned aerial vehicles (UAVs) can overcome the limitation of spatial sensing capabilities of individual UAVs has attracted extensive attention in the field of direction-of-arrival (DoA) estimation. Considering the constrained hardware resources in UAV swarm, maintaining coherence among all elements is extremely challenging. In this paper, we consider a collaborative UAV swarm with partly calibrated subarrays and propose a dual-driven joint-sparse pattern-coupled sparse Bayesian learning unfolding network (DD-JPC-SBLNet) for non-coherent DoA estimation. Specifically, we first propose a novel joint-sparse pattern-coupled sparse Bayesian learning (JPC-SBL) algorithm, which introduces a pattern-coupled model and a distributed noise variance estimation module, to improve the non-coherent DoA estimation accuracy. Then, the JPC-SBL algorithm is unfolded into cascaded customized neural networks, each of which learns the optimal coupling parameter configuration based on the pattern-coupled model and learns the current optimal signal hyperparameter update rule with a data-driven customized neural network. By effectively combining the advantages of model-driven and data-driven, the proposed dual-driven unfolding network exhibits superior convergence performance and speed. Simulation results demonstrate that the proposed method not only outperforms existing methods in terms of estimation accuracy and angular resolution, but also reduces computational complexity by more than 50% compared with other SBL-based algorithms.

In space–air–ground integrated networks (SAGINs), efficient task scheduling is critical to ensuring low latency and energy efficiency for computation-intensive applications. This paper proposes a hierarchical reinforcement learning (HRL)–based task scheduling framework for space–air integrated edge computing systems, consisting of low Earth orbit (LEO) satellites and unmanned aerial vehicles (UAVs). The architecture is structured into two layers: UAVs handle task reception and local execution, while satellites assist in offloaded task processing. To enable intelligent and decentralized decision-making, we employ a multi-agent twin delayed deep deterministic policy gradient (MATD3) algorithm for UAVs and a TD3-based controller for satellite-level coordination. A shared reward mechanism is introduced to promote cross-layer optimization. Simulation results demonstrate that the proposed framework significantly reduces task execution delay and energy consumption compared to baseline schemes, achieving faster convergence. These results verify the effectiveness and practicality of the proposed method in dynamic and resource-constrained space–air environments. The proposed method reduces average total cost (ATC) by at least 13% compared to existing methods.

By decoupling substrate resources, network virtualization (NV) is a promising solution for meeting diverse demands and ensuring differentiated Quality of Service (QoS). In particular, virtual network embedding (VNE) is a critical enabling technology that enhances the flexibility and scalability of network deployment by addressing the coupling of Internet processes and services. However, in the existing deep neural networks (DNNs)-based works, the closed-box nature DNNs limits the analysis, development, and improvement of systems. For example, in the Industrial Internet of Things (IIoT), there is a conflict between decision interpretability and the opacity of DNN-based methods. In recent times, interpretable deep learning (DL) represented by deep neuro fuzzy systems (DNFSs) combined with fuzzy inference has shown promising interpretability to further exploit the hidden value in the data. Motivated by this, we propose a DNFS-based VNE algorithm that aims to provide an interpretable NV scheme. Specifically, data-driven convolutional neural networks (CNNs) are used as fuzzy implication operators to compute the embedding probabilities of candidate substrate nodes through entailment operations. And, the identified fuzzy rule patterns are cached into the weights by forward computation and gradient back-propagation (BP). Moreover, the fuzzy rule base is constructed based on Mamdani-type linguistic rules using linguistic labels. In addition, the DNFS-driven five-block structure-based policy network serves as the agent for deep reinforcement learning (DRL), which optimizes VNE decision making through interaction with the environment. Finally, the effectiveness of evaluation indicators and fuzzy rules is verified by simulation experiments.

Adaptive bitrate (ABR) streaming provides an effective way to improve the Quality of Experience (QoE) of video users and is now the de facto standard for video delivery. Meanwhile, mobile edge computing (MEC) has been applied to assist ABR streaming, improving the performance of mobile networks and enabling efficient video delivery. However, smooth ABR streaming relies on the bidirectional adaptation between bitrate selection and bandwidth allocation, as they operate on distinct timescales and have different optimization goals. Moreover, since the constrained wireless resources available within a cell are shared by multiple users, their QoE should be optimized not only jointly but fairly. To this end, we propose a QoE-fairness-aware bandwidth allocation (QFA-BA) method for MEC-assisted ABR video transmission. With a novel perspective on buffer occupancy modeling, the relationship between bitrate selection and bandwidth allocation is studied. An enhanced QoE evaluation model is then proposed to correlate bitrate selection with bandwidth allocation and facilitate QFA-BA. Finally, a soft actor-critic (SAC) framework improving both the QoE and QoE-fairness is presented for QFA-BA. Compared with the state-of-the-art methods, our QFA-BA can perceive fine-grained buffer occupancy and stabilize it near a preset value with relatively more and larger bitrate switchings, exhibiting smoother convergence, better QoE (50.29%) and QoE fairness (54.81%).

Mobile edge computing (MEC) supported by Low Earth Orbit (LEO) satellite communication systems is a promising approach to improve the Quality of Service (QoS) of terrestrial users. Most existing offloading methods neglect the downlink connectivity of satellites during task offloading, which may bring in extra delay and affect the offloading efficiency. This paper considers satellite-terrestrial-MEC networks in which the tasks are generated by source devices, and the computing services are provided by destination devices or the LEO satellites. Furthermore, we propose a joint task offloading and beam hopping optimization method, which aims to minimize the average time delay of all computation tasks. We formulate the joint optimization problem as a zero-one integer programming problem, which is NP-hard, and provide a multi-agent deep reinforcement learning (MADRL) framework for the intelligent task offloading scheme. In this framework, each agent is responsible for either task offloading or beam hopping, with shared rewards to enhance cooperation between agents. Additionally, to reduce the training time of traditional reinforcement learning algorithms, an attention mechanism is incorporated into MADRL. Simulation results demonstrate that our method effectively reduces the average offloading delay compared with other baseline schemes.

With the explosive growth in demands for mobile video services, the focus of cellular networks is evolving from the core network to the edge network to facilitate resource-intensive services and mitigate backhaul burdens. However, the overlapping in coverage and the user similarity in preferences lead to substantial cache redundancy. In addition, the mobility of users poses a significant challenge to smart devices for device-to-device (D2D) communications, which affects the content richness and hit ratio of the edge cache. In this paper, an adaptive cache partitioning and placement (ACPP) strategy is proposed for mobile edge computing (MEC) networks to minimize the average cost of content access. A practical two-layer collaborative caching model is presented, which comprises 5G base station (gNB) clusters and D2D caching. Besides, a public and private cache partitioning method is designed for gNBs to improve the content richness of local cache, and a static and dynamic cache partitioning method is developed for user devices to address the varying mobility patterns. Simulation results demonstrate the effectiveness of the proposed ACPP strategy in delivering content at a lower average cost, and achieve a better hit ratio with a relatively high energy consumption at user ends.

Space-aerial-terrestrial integrated networks (SATINs) have gained widespread recognition. Due to the mobility of wireless access points (APs) and user terminals (UTs), handover plays an essential role in ensuring the continuity and quality of communication in SATINs. However, existing handover strategies only consider the instantaneous communication benefits, which makes it difficult to maintain link stability with lower handover frequency. Given that reasonable handover paths help improve the stability of communication, we propose a handover planning (HP) strategy in SATINs, which fully considers the time varying link quality between two handover in planning the handover paths of mobile UTs. Simulation results demonstrate that the proposed HP strategy can maintain high-quality communication links with the lowest handover delay while performing relatively low in handover frequency.

Intrusion detection system (IDS) is an active defense mechanism implemented by the Internet of Things (IoT), which can identify the intrusion behavior and initiate alarms. However, there are concerns regarding the sustainability and feasibility to existing schemes when facing the increasing of threats in IoT. In particular, these concerns in terms of the increasing levels of adaptive performance and the insufficient levels of detection accuracy. In this paper, we present a novel deep learning method to address the aforementioned concerns. We detail the proposed convolution neural network model based on the developed feature fusion mechanism. Furthermore, we also propose a Symmetric logarithmic loss function based on categorical cross entropy. In addition, the proposed detection framework has been applied to GPU-enabled TensorFlow, and evaluated using the benchmark of NSL-KDD datasets. Extensive experimental results indicate that the developed model outperforms traditional approaches and has great potential to be applied for attacks detection in IoTs.

Transmit antenna selection (TAS) is an emerging technology in physical layer security. To provide new insights into the achievable secrecy performance of TAS in practical communication systems, this paper investigates the average secrecy rate (ASR) and secrecy outage probability (SOP) under practical modulation schemes in TAS aided multiple-input multiple-output multiple-antenna eavesdropper (MIMOME) wiretap channels over Rayleigh fading. Particularly, this research concentrates more on the square M-ary quadrature amplitude modulation (M-QAM). Furthermore, in the considered MIMOME channel, a single antenna is selected to transmit the secret message, and selection combining (SC) or maximal-ratio combining (MRC) is utilized at the legitimate receiver and the eavesdropper. Based on this system model, novel expressions for the ASR and SOP are formulated to characterize the secrecy performance of the finite-alphabet driven MIMOME channel. Besides exact analysis, an asymptotic analysis is performed using the considered performance metrics in high signal-to-noise ratio (SNR) regime. Theoretical analyses suggest that the asymptotic ASR and SOP converge to finite constants in high SNR regime due to the discrete constellation constraint, and we find that the asymptotic behaviour of discrete inputs differs from that of Gaussian inputs. Furthermore, we derive concise expressions to characterize the rate of convergence (ROC) of the ASR and SOP, respectively. To unveil more system design insights, we discuss the relationship between the ROC and several important system parameters such as the antenna number and the modulation order.

Speech consists of vocal sounds resulting from the synchronism of all parts of the human phonatory apparatus. The oral communication between individuals is supposed to sound pleasant, and this characteristic is strongly related to the subjective parameters studied by phonoaudiology, such as roughness, breathiness, and strain. Commonly, these characteristics are evaluated with a noninvasive vocal disorder test. This paper proposes the development of an intelligent computational tool to classify and point out the preponderance of such parameters from audio samples. Moreover, a comparative study was made to evaluate the efficiency of several Wavelet families as feature extraction methods. The features extracted were used with artificial neural networks and an automatic routine was set up to find the best topology of the Multi-Layer Perceptron (MLP) architecture used in the classification of such speech parameters. The results are promising and reliable, with accuracy rates higher than 98.9%.

Increasing marine activities taking place within the exclusive economic zone (EEZ) have made broadband maritime communications very attractive in recent years. In this paper, a coordinated satellite and terrestrial architecture (II-CST) is presented to implement real-time and broadband Internet access at sea. Further, since the mobility of shipborne base stations (S-BSs) may cause complex inter-cell interference (ICI), an S-BS sleeping control scheme is proposed to adapt the system resource allocation corresponding with the user load. When the proportion of blocked users is beyond threshold, the sleeping control process is triggered to reduce ICI and recover broadband access. Then we use the sailing position of accessed users as a constraint to adjust the downlink power allocation of S-BSs. Simulation results with real ship data of China's Yellow Sea show that under the II-CST, our adaptive S-BS sleeping control can contribute to achievable downlink data rate, mobility robustness, and power efficiency of the dynamic broadband maritime communication system.

Increasing marine activities taking place within the exclusive economic zone (EEZ) have made broadband maritime communications very attractive in recent years. In this paper, a coordinated satellite and terrestrial (II-CST) architecture is presented to enable real-time and broadband Internet access within the EEZ. Further, a voyage-based resource allocation scheme is proposed to dynamically adapt the system resource consumption corresponding with the user load. We make switching plans for shipborne base stations (S-BSs) considering the proportion of blocked users and changes in user distribution to recover broadband access and reduce inter-cell interference at sea, and take the sailing position of accessed users as a constraint to cooperatively adjust the downlink power of S-BSs. Simulation results under real ship data of China's Yellow Sea show that under the II-CST architecture, our resource allocation scheme can effectively improve the signal to interference noise ratio (SINR) of users, reduce the number of handover-related link failures, and reduce the power consumption of S-BSs.

In network virtualization environment, if a virtual network (VN) needs to be deployed across multiple infrastructure domains, a splitting scheme of the VN should be found. With the goal of minimizing embedding cost, the existing methods solve VN splitting by linear programing. However, since the VN splitting problem is NP-Hard, these methods will take a lot of computing time when the problem scale gets bigger. In this paper, a max-flow/min-cut theory based VN splitting mechanism is proposed. The proposed method first creates a binary tree of the InPs by system clustering method, based on which the multidomain VN splitting problem is decomposed into several two-domain VN splitting problems. Then the method transforms each two-domain splitting problem into a max-flow/min-cut problem, and solves it by the shortest augmenting path algorithm efficiently. Simulations show that the proposed mechanism can improve the efficiency of VN splitting steadily and save the embedding cost.

Multi-domain virtual network embedding (MVNE) aims to embed a virtual network (VN) across multiple physical domains while minimizing the embedding cost. A key phrase of MVNE is VN partitioning which partitions a VN into multiple physical domains. Since the MVNE problem is NP-hard, we provide a heuristic VN partitioning approach named VNP-PSO based on the Particle Swarm Optimization (PSO) to increase the efficiency of VN partitioning. The VNP-PSO algorithm generates a near-optimal solution of VN partitioning through the evolution process of the particles. The simulation results show that our proposal can increase the efficiency of VN partitioning and decrease the embedding cost of MVNE.

Survivable virtual network mapping (SVNM) has been extensively investigated to guarantee that the mapped virtual network (VN) works normally against substrate failures. The existing studies of SVNM mainly focus on single node or single link failure. Since natural disasters usually cause severe substrate failures in geographic regions, some work addressing SVNM against regional failures has been studied. However, the current approaches only solve the mapping problem against single regional failure. When there are multiple regional failures aroused by natural disasters, such approaches are not effective. In this paper, we first design a regional failure model with the knowledge of risk assessment. Then we propose two effective mapping algorithms based on the disaster-prediction scheme with the regional failure model. One is the minimum link risk prior selection algorithm and the other is the asymmetric parallel flow allocation algorithm. Simulation results show that both approaches can reduce the capacity loss of virtual networks caused by regional failures and can effectively increase the average VN acceptance ratio.

Network virtualization has been regarded as a core attribute of Future Internet. To improve the quality of virtual network, it is important to diagnose the faulty components quickly and accurately. Recently more and more researches focus on end-user fault diagnosis, which can fit incomplete knowledge and dynamic challenges. In this paper, we present a fault diagnosis system called DiaEO in virtual network. It improves the present end-user fault diagnosis methods by screening evidences before analyzing to reduce the time-consuming. Besides that, DiaEO also improves the anti-noise ability of the system. The simulation results show that the proposed method can keep high accuracy and ameliorate time performance.

Survivable virtual network embedding (SVNE) aims at embedding a virtual network (VN) in a way, that after being affected by substrate failures, the VN is still operating. Based on the single node failure assumption, that at any time there can be at most one failed substrate node, the existing studies for the SVNE against substrate node failures back up VNs with a maximum resource sharing. However, multiple node failures do happen in reality, thus those methods are not always effective. In this paper, we propose a topology-aware VN embedding approach to enhancing the survivability against multiple node failures. We make use of the topology attributes to provide each substrate node with multiple potential failover choices, based on which a recoverability-based VN embedding algorithm and a profit-driven VN remapping algorithm are presented. Simulation results show that the proposed approach can achieve rational resource allocation and effectively increase the long term business profit to the infrastructure provider.

Network virtualization has been regarded as a core attribute of the Future Internet. In a Network Virtualization Environment (NVE), heterogeneous virtual networks can share the same physical infrastructure regardless of their different topologies, demands, protocols and so on. In this case, the Survivable Virtual Network Embedding (SVNE) problem becomes increasingly critical to overcome the failure of physical infrastructure. Backup resources needed to provide survivability of virtual network undoubtedly increase the challenge of resources efficiency of SVNE. In this paper, we study the SVNE problem and propose a method of allocating bandwidth resources based on load balancing of the physical resources and a strategy of reconfiguring backup resources. Simulation experiments show that load balancing based method has a higher performance in the long term acceptance ratio, revenues and utilization of substrate links. And the reconfiguration of backup resources is cost-efficient and also helpful to increase the acceptance ratio.