Pansharpening is a powerful technique for generating high-resolution multispectral (HRMS) images by fusing currently available image pairs of low-resolution multispectral (LRMS) and texture-rich panchromatic (PAN) data, effectively addressing the physical constraints of satellite sensors. While recent generative diffusion models have demonstrated impressive performance gains in this domain, their prohibitive computational demands and training costs hinder practicality in resource-constrained remote sensing satellite systems. In this work, we propose NODiff, a novel diffusion framework that replaces the conventional attention-based denoising backbone with a neural operator, seamlessly integrating operator learning and generative modeling into an efficient yet effective solution for pansharpening. In practice, we implement our approach through a two-stage learning paradigm: First, we pretrain the proposed Neural Operator-based diffusion model to learn the high-resolution texture priors essential for pansharpening. Afterward, we freeze the pretrained parameters, and design a lightweight conditional detail guidance adapter to enable efficient fine-tuning for generating desired HRMS images. Meanwhile, a time-aware low-rank adaptation is introduced to dynamically refine high-frequency details potentially affected by spectral mode truncation. Extensive experiments on multiple benchmark datasets demonstrate that NODiff achieves competitive pansharpening performance while significantly reducing training and inference costs. Beyond pansharpening, our method provides new insights into building resource-efficient generative models.

High-mobility and large-bandwidth applications at high frequencies make the acquisition of doubly-selective channels costly and complex, as the fast fading channel causes a surge in pilot overheads and inter-carrier interference. Accurate estimation of the complex channel gain (CG) and the carrier frequency offset (CFO) is required to support subsequent high-accuracy channel prediction that alleviates the heavy pilot burden. The emerging technology of the wireless channel map (WCM) can approximately reproduce the actual propagation environment digitally with customized channel parameters, offering an opportunity to accurately estimate the complex CG and CFO. In this paper, a WCM-based prior distribution construction method and a joint complex CG and CFO estimation algorithm are proposed. Specifically, a parameterized linear estimation problem for the complex CG is generated based on a nonuniform delay-domain off-grid channel representation, and with more realistic prior distributions constructed by the knowledge from the WCM-provided angular-delay power spectrum density, the joint estimation problem is solved under the Bayesian inference framework. Simulation results demonstrate the superiority of the proposed algorithm, with a better performance in terms of estimation accuracy and bit error rates (BER) than existing baselines. It is also verified that the proposed WCM-based algorithm is highly adaptable to different WCM precision and robust to different user speeds.

Traditional geometry-based positioning methods suffer from low accuracy in non-line-of-sight environments, whereas fingerprint-based positioning methods incur high maintenance costs due to the need for continuous database updates. Channel charting-based positioning methods adopt unsupervised learning to infer user positions from channel state information, but often suffer from low accuracy under complex propagation conditions. Channel maps enable the prediction of real-world positions by providing a priori channel information associated with the propagation environment. Building on this concept, channel charting-based positioning methods leverage this prior knowledge to enhance positioning accuracy and reduce reliance on large labeled datasets. In this paper, we propose a novel channel charting-based semi-supervised positioning algorithm with co-training, which utilizes both labeled data from the channel map and unlabeled data from practical communication systems to predict real-world geographical positions. This algorithm leverages a covariance-based channel feature and a corresponding dissimilarity metric to enhance robustness against noise and timing advance interference. Through ray tracing simulations calibrated by real-world measurements, the proposed algorithm is compared with state-of-the-art positioning methods under different noise conditions, demonstrating its effectiveness and superiority.

Conventional wireless channel measurements and modeling typically study channels via discrete spatial sampling. As the sixth-generation (6G) wireless communication places higher demands on the accuracy of channel state information, this discrete approximation becomes insufficient and motivates research on continuous-space channels. In this work, 28 GHz indoor continuous-space channel measurements are conducted, and key channel characteristics are analyzed. Based on the analysis, the necessity of continuous-space channel research is validated, and the role of channel maps is demonstrated. Furthermore, a continuous-space channel map construction method using the Graph SAmple and aggreGatE (GraphSAGE) algorithm is proposed. By the learning on spatial aggregation function rather than performing a passive weighted sum, the proposed GraphSAGE-based map construction method can reduce the performance bias caused by discrete spatial sampling and recover the continuous-space channels accurately. Continuous-space measurements are used as benchmarks to compare the performance of the GraphSAGE-based channel map. Extensive experiments confirm the superiority of the proposed GraphSAGE-based method over existing artificial intelligence (AI) algorithms, providing a robust approach for the 6G continuous-space channel map construction.

To address sparse channel measurement data and inadequate predictive capabilities in conventional channel models, predictive channel modeling employs joint generative and predictive architectures to enhance robustness. In this paper, we propose an enhanced diffusion-driven predictive framework that integrates generative augmentation and prior-aware prediction into a unified learning pipeline. We first introduce a space-time-frequency (STF) coupled diffusion network based on transformers that generates synthetic channel data preserving critical channel statistical properties. Additionally, we compress measured channel state information into a low-dimensional manifold via a latent encoder and introduce an innovative composite training scheme that couples diffusion-driven prior generation with prediction, equipping the predictive module with rich latent features that lift its performance ceiling and markedly improve generalization across diverse scenarios. Extensive experiments confirm the superiority of our algorithm, and its performance is further validated using channel measurement data, thereby demonstrating its robustness for advanced wireless communications in real-world deployment scenarios.

Cell-free (CF) massive multiple-input multiple-output (M-MIMO) provides an almost uniformly high data rate for all user equipment (UE) through multiple access points (APs), with a non-negligible signal processing burden. Angle domain transmission and channel maps promise to alleviate this burden by reducing channel dimensions in the angle domain and providing$a$priori channel information, respectively. In this paper, we propose a channel map-based angle domain multiple access scheme for uplink CF M-MIMO communications. First, we propose a twostage data reception and pilot assignment scheme constituting receive combining and large-scale fading decoding (LSFD) to reduce overall interference and maximize spectral efficiency (SE). Furthermore, we construct two channel map-based transmission mechanisms by wielding different levels of channel information, where a tailored data reception scheme with a newly derived SE upper bound is also proposed for quantitative evaluation. Simulation results show that the proposed schemes outperform both their space domain alternatives and those without using channel maps in terms of SE.

The current propagation scenario classification methods of standardized channel models are rough and the definitions are unclear, despite the diverse propagation scenarios in the sixth-generation (6G) communications technology era. This paper proposes an unsupervised classification algorithm based on the self-organizing map (SOM) to establish a more refined propagation scenario classification for the 6G era. First, the physical environment information of the actual scenario is extracted based on digital maps. Then, the SOM neural network parameters are determined and the propagation scenario classification is implemented. Finally, the effectiveness of the proposed propagation scenario classification is verified by the data collected from the practical communication network in the following three aspects: physical environment characteristics, evaluation indicators, and channel characteristics. The results show that the proposed SOM scenario classification algorithm outperforms K-means, DBSCAN, and the Gaussian mixture model (GMM), among other algorithms, providing an effective solution for the detailed classification of 6G propagation scenarios.

Traditional geometry-based positioning methods (GPMs) suffer from low accuracy in non-line-of-sight (NLOS) environments, while fingerprint-based methods (FPMs) encounter high maintenance costs due to the need for continuous database updates. As an emerging unsupervised technique, channel charting can address these challenges by mapping channel state information (CSI) into a low-dimensional virtual space that represents pseudo-positions of user equipments (UEs). In this paper, a channel charting-based positioning algorithm that leverages a covariance-based channel feature and a corresponding dissimilarity metric is proposed to enhance robustness against noise and timing advance (TA) interference. Additionally, an improved Triplet neural network algorithm is introduced, which enables channel charting to directly and accurately predict real geographical positions. Through ray tracing simulations, the proposed algorithm is compared with state-of-the-art channel charting-based positioning methods, demonstrating its effectiveness and superiority.

Over the past decades, pansharpening has contributed greatly to numerous remote sensing applications, with methods evolving from theoretically grounded models to deep learning approaches and their hybrids. Though promising, existing methods rarely address pansharpening through the lens of underlying physical imaging processes. In this work, we revisit the spectral imaging mechanism and propose a novel physics‐informed neural operator framework for pansharpening, termed PINO, which faithfully models the end‐to‐end electro‐optical sensor process. Specifically, PINO operates as: (1) First, a spatial-spectral encoder pair is introduced to aggregate multi-granularity high-resolution panchromatic (PAN) and low-resolution multispectral (LRMS) features. (2) Subsequently, an iterative neural integral process utilizes these fused spatial-spectral characteristics to learn a continuous radiance field $L_i(x, y, \lambda)$ over spatial coordinates and wavelength, effectively emulating band-wise spectral integration. (3) Finally, the learned radiance field is modulated by the sensor’s spectral responsivity $R_b(\lambda)$ to produce physically consistent spatial–spectral fusion products. This physics-grounded fusion paradigm offers a principled solution for reconstructing high-resolution multispectral and hyperspectral images in accordance with sensor imaging physics, effectively harnessing the unique advantages of spectral data to better uncover real-world characteristics. Experiments on multiple benchmark datasets show that our method surpasses state-of-the-art fusion algorithms, achieving reduced spectral aberrations and finer spatial textures. Furthermore, extension to hyperspectral (HS) data demonstrates its generalizability and universality. The code will be available upon potential acceptance.

The sixth-generation (6G) wireless system will involve large-scale wireless channels, but traditional channel acquisition methods are resource-consuming. Digital twin online channel models have been proposed to facilitate efficient 6G network optimization, which maps physical systems to the virtual world, thereby enabling the prediction of physical data. In this paper, we built a Swin Transformer Aided Urban Digital Twin Online Channel Modeling Platform (SmarTwin-OCMP). The platform begins by importing data from the geographic information system (GIS) to obtain the original static urban models based on Unity, then develops a user-friendly interface functionality, capable of providing the channel characteristics of the corresponding communication scenarios. Swin Transformer is adopted to segment various urban communication scenarios on our remote sensing imagery dataset and achieves satisfactory recognition accuracy. Subsequently, we perform electromagnetic parameter matching and prediction of real-time channel characteristics based on the 6G pervasive channel model (6GPCM) according to the recognition and segmentation results. Finally, we integrate all these functionalities into our Unity platform to enable users to interactively explore urban channel information in a more flexible and intuitive manner.

With the rapid evolution of 6G wireless communication technology, the granular classification of communication scenarios becomes increasingly sophisticated. This necessitates a deeper exploration of the intrinsic relationships between environmental information and channel characteristics. Consequently, the development of efficient and accurate methods for communication scenario reconstruction emerges as a critical imperative. Leveraging comprehensive three-dimensional (3D) spatial information from point cloud data, we propose a two-stage workflow for processing massive unstructured point cloud data to generate triangular mesh models of large indoor communication environments. The first stage implements an enhanced RANdom SAmple Consensus (RANSAC) algorithm with adaptive thresholding for robust wall structure extraction. Subsequently, we employ a hybrid reconstruction method combining template-based deformation for furniture elements with a zero-shot semantic segmentation network for wall opening detection. The geometric information extracted through the aforementioned process is utilized to generate mesh models, and ray-tracing (RT) is adopted to simulate channel characteristics. Finally, the efficiency and accuracy of the proposed scenario reconstruction and channel modeling method is demonstrated by comparing its simulated channel characteristics with those of channel measurements.

The separation of antenna design and channel modeling has made it difficult to capture the intricate interplay between antenna parameters and channel dynamics, resulting in suboptimal performance in real-world environments. To address this challenge, this study introduces a stochastic framework that integrates random antenna array, modeled as one-dimensional Wiener process, with a realistic wireless propagation channel model, i.e., the sixth-generation pervasive channel model (6 GPCM). This framework is grounded in rigorous derivations and incorporates reasonable approximations. It provides a detailed analysis of how antenna mobility, characterized by the increasing variance of the Wiener process over the same time interval, influences the spatial cross-correlation function (SCCF), temporal autocorrelation function (TACF), and channel capacity. Finally, based on the aforementioned derivations, simulations are conducted to validate the theoretical findings, and the results are analyzed to demonstrate their implications for wireless communication system design.

Indoor positioning is an important application of integrated sensing and communication technology in the sixth generation (6G) wireless communications. To address the limitations of existing fingerprint-based positioning methods (FPMs) under severe multipath effects, a novel channel state information (CSI)-based channel fingerprint structure and a novel line-of-sight (LoS)/non-LoS (NLoS) identification-assisted positioning method (IAPM) is proposed for 6G indoor multiple-input multiple-output (MIMO) communications. The proposed channel fingerprint structure uses the proposed maximum received power path to enhance the feature discrimination. The proposed IAPM incorporates a LoS/NLoS identification module and an improved weighted random forest (IWRF) positioning algorithm for accurate positioning. Evaluations on both channel measurement data and channel synthetic data generated by ray tracing demonstrate that the proposed method achieves superior accuracy and robustness compared with widely used FPMs.

Pansharpening is the critical technology for generating high-resolution (HR) multispectral (MS) images by learning the cross-modality complementary representations between the panchromatic (PAN) images and low-resolution (LR) MS images. Though methods based on convolutional neural networks (CNNs) have dominated the pansharpening community, they still suffer from the limited global modeling capability due to the inherent property of the convolutional operator. To remedy this common limitation, the transformer family has recently gained great popularity in this field. However, existing cascaded transformer designs inevitably introduce a heavy memory footprint and computational cost due to the dense dot-product self-attention (SA) computation. More importantly, these paradigms simply ignore the innate sparsity of remote sensing images, leading to information redundancy and a challenging optimization process. To alleviate these issues, we propose the bilateral adaptive evolution transformer (BAEFormer), which is built upon two core mechanisms: bilateral attention computation and adaptive attention evolution. Specifically, we first decompose the conventional quadratic complexity SA into linear-degree height and width computing at the first stage, respectively, which significantly reduces the computational complexity. Given the data-specific properties, furthermore, we devise a novel yet effective neighboring layer-dependent strategy to adaptively update the attention map of two spatial dimensions, thereby avoiding the repetitive SA computation while taking into account the dynamics toward the evolution of attention weights. Our model, called BAEFormer, outperforms other state-of-the-art pansharpening methods on various remote sensing datasets while showing fewer network parameters and computational requirements. The code is available athttps://github.com/coder-JMHou/BAEFormer.

Pan-sharpening essentially to boost the spatial resolution of a multispectral (MS) image guided by its paired panchromatic (PAN) image. In other words, this process intricately integrates the high-frequency components extracted from texture-rich PAN images into the low-resolution (LR) MS images, resulting in texture-rich MS images. Though existing deep learning (DL)-based techniques have made impressive performance compared with traditional algorithms, they still face challenges in accurately restoring high-frequency details in MS images, thus limiting overall pan-sharpening performance. In addition, reference high-resolution (HR) MS images are often underutilized, typically serving only as training labels. In this work, we present a general high-frequency enhancement framework for pan-sharpening, which is implemented through a cooperative optimization strategy using mutual information (MI) maximization and contrastive learning. Specifically, our model comprises two fundamental modules: the high-frequency feature alignment (HFFA) module and the high-frequency detail calibration (HFDC) module. The first employs MI maximization to align the high-frequency semantic statistical distribution between PAN images and reference HRMS images. The latter is designed to calibrate the high-frequency components of MS modality under the guidance of the PAN counterparts through the contrastive learning constraint, thereby producing more accurate high-frequency information on MS modality. By integrating the calibrated high-frequency features of MS modality and those of PAN modality, we can obtain a more comprehensive and precise high-frequency feature representation of these two modalities, facilitating the reconstruction of LRMS images. Our model, incorporating the aforementioned key elements, significantly surpasses other state-of-the-art (SOTA) techniques across multiple satellite datasets in both quantitative and qualitative experiments. Moreover, the real-world full-resolution and cross-sensor assessments testify to its exceptional generalization capabilities. The code is available athttps://github.com/Vcocoi/CONet.

Urban areas serve as one of the most important scenarios in sixth generation (6G) wireless communications, necessitating comprehensive and in-depth wireless channel characteristics studies. The wireless channels in various 6G urban communication scenarios usually show different characteristics. To provide robust identification across a wide range of 6G urban communication scenarios, we propose a novel intelligent scenario identification algorithm. The proposed algorithm leverages an enhanced multi-layer perceptron (MLP) and autoencoder, utilizing easily accessible environmental physical features as inputs. To enhance identification performance, we employ a series of data pre-processing methods and hyperparameter optimization algorithm. Subsequently, experimental results demonstrate the superior performance of our proposed algorithm compared to the benchmark algorithm. Ultimately, the channel characteristics in all domains, including spatial, temporal, frequency, angle, Doppler, and delay, are studied for these scenarios.

The development of sixth-generation (6G) wireless communication systems brings significant challenges in channel modeling. Conducting channel measurements for 6G communications is highly expensive and cannot cover all scenarios and frequency bands. Moreover, existing conventional channel models fail to accurately predict channel characteristics in unknown frequency band. As a result, predictive channel modeling has emerged as a promising solution for addressing these challenges in 6G channel modeling. In this study, we propose a frequency domain predictive channel model that combines an autoencoder with a coupling Convolution Gated Recurrent Unit (Conv-GRU) cells. The proposed model aims to predict channel characteristics in unknown frequency bands. The proposed predictive channel model is validated by using data collected from multiple frequency bands channel measurements. To evaluate its performance, several commonly used prediction networks, i.e., a general LSTM network, a GRU-based predictive network, and a Conv-LSTM-based predictive network, are conducted as benchmarks for comparison. Based on evaluation results, our proposed predictive channel model achieves the highest level of accuracy in predicting channels. Additionally, we provide a performance bound for extrapolation predictability using a Ray Tracing simulator.

Pansharpening is a challenging low-level vision task whose aim is to learn the complementary representation between spectral information and spatial detail. Despite the remarkable progress, existing deep neural network (DNN) based pansharpening algorithms are still confronted with common limitations. 1) These methods rarely consider the local specificity of different spectral bands; 2) They often extract the global detail in the spatial domain, which ignore the task-related degradation, e.g., the down-sampling process of MS image, and also suffer from limited receptive field. In this work, we propose a novel bidomain modeling paradigm for pansharpening problem (dubbed as BiMPan), which takes into both local spectral specificity and global spatial detail. More specifically, we first customize the specialized source-discriminative adaptive convolution (SDAConv) for every spectral band instead of sharing the identical kernels across all bands like prior works. Then, we devise a novel Fourier global modeling module (FGMM), which is capable of embracing global information while benefiting the disentanglement of image degradation. By integrating the band-aware local feature and Fourier global detail from these two functional designs, we can fuse a texture-rich while visually pleasing high-resolution MS image. Extensive experiments demonstrate that the proposed framework achieves favorable performance against current state-of-the-art pansharpening methods. The code is available at https://github.com/coder-qicao/BiMPan.

The research on 6G wireless communications has become a focal point in the global technological competition. For 6G wireless systems, channel modeling is the foundation for the design and deployment of wireless communication systems. With unexplored channel characteristics expected to emerge from new frequency bands and scenarios in 6G wireless channels, predictive channel modeling becomes particularly important. This paper proposes an image-based real-time predictive channel modeling using image processing and machine learning (ML) algorithms, which extracts effective channel information from images and has been validated through channel measurement data and synthetic channel data. The proposed channel model considers large-scale identification and small-scale feature extraction. The former aspect is accomplished by training DeepLabv3+ on our image dataset and classifying the scenarios. For feature extraction, we use VGG-16 as the backbone network and validate it with the 3GPP Uma path loss model for comparison. The results show that predictive channel modeling using segmented images is superior to that of the 3GPP Uma path loss model. Even more, it performs better than the original image-based prediction.

Artificial intelligence (AI) is a promising solution to achieve channel prediction under limited channel data. In this paper, a novel scatterer density-based predictive channel model is proposed to predict channels in multiple scenarios. By exploring the graph attention networks (GAT) and gated recurrent unit (GRU), the proposed model captures multi-domain information in dynamic scenarios. Besides, it extracts highly space-time correlated data characteristics, captures channel dynamic evolutional patterns, and predicts channels in different scenarios. The space-time graph channel datasets are constructed based on the ray tracing (RT) simulation channels. In the prediction experiments, the proposed method is validated on the datasets to predict channels with good performance. Compared with the 3GPP TR 38.901 channel model, the proposed model obtains more accurate channel statistical properties in different scenarios.

Space–air–ground-integrated networks (SAGINs) are deemed as a promising solution to support multifarious Internet of Vehicles (IoV) services with diversified Quality-of-Service (QoS) requirements in future communication networks. Network function virtualization (NFV) and software-defined networking (SDN) are two complementary and promising technologies to reduce the function provisioning cost and coordinate the heterogeneous physical resources in SAGIN. In this article, we investigate the online dynamic virtual network function (VNF) mapping and scheduling in SAGIN, considering the dynamicity of IoV services. The VNF live migration, VNF reinstantiation, and VNF rescheduling are enabled to increase the service acceptance ratio and service provider’s profits. Considering the heterogeneity of space, air, and ground nodes, we first model the migration cost and additional delay incurred by VNF live migration and reinstantiation. We then formulate the dynamic VNF mapping and scheduling jointly as a mixed-integer linear programming (MILP) problem with specified cost and delay models. We propose two Tabu search (TS)-based algorithms, i.e., TS-based VNF remapping and rescheduling (TS-MAPSCH) algorithm and TS-based pure VNF rescheduling (TS-PSCH) algorithm, to obtain suboptimal solutions to the MILP problem efficiently. Simulation results show that the proposed solution is very close to the optimum and that the proposed dynamic algorithms outperform existing works with respect to multiple performance metrics, including the service provider’s profit, service acceptance ratio, and QoS satisfaction level.

In this article, a two-level soft-slicing scheme is proposed for 5G-and-beyond radio access networks to support ultrareliable and low-latency communications (URLLC) and enhanced mobile broadband (eMBB) services with delay/reliability and throughput requirements, respectively. At the network level, we first determine the number of radio resources required for eMBB services and analyze the delay violation probability for URLLC services. Then, an integer nonlinear program is formulated for the network-level resource preallocation. Since the formulated problem is NP-complete, a low-complexity heuristic algorithm is proposed to obtain near-optimal solutions. Given the preallocated resources at each gNodeB (gNB), a gNB-level resource scheduling scheme is designed to enable real-time resource sharing among URLLC services considering the reliability and delay requirements. Simulation results show that the proposed soft-slicing scheme meets stringent quality-of-service requirements for both URLLC and eMBB services and achieves high resource utilization efficiency when compared with conventional hard resource slicing schemes.

The rechargeable battery of a plug-in electric vehicle (PEV) endows the PEV with dual roles in the power grid as power load and mobile energy storage (MES). Owing to the technical advancement of autonomous driving, private PEVs that are parked most of the day can be used as private MESs (PMESs) to autonomously deliver energy for overloaded charging stations (CSs). In this paper, we investigate an energy compensation problem where PMESs are scheduled to deliver energy to overloaded CSs so that the energy balance can be achieved while the energy delivery time can be minimized. Based on the time-variant CS operation status and traffic conditions, we propose a pricing-based scheduling scheme that considers both PMES navigation and incentive price design. First, to navigate PMESs in the energy-capacitated transportation system, a minimum-cost flow problem is formulated to minimize the energy delivery time. Then, the incentive price is determined to encourage PMESs to follow the optimal navigation results for energy delivery. Simulations are conducted based on the traffic data of California highway to validate the effectiveness of the proposed scheduling scheme.

The identification of structural variations (SVs) and viral integrations in circulating tumor DNA (ctDNA) is a key step in precision oncology that may assist clinicians in treatment selection and monitoring. However, due to the short fragment size of ctDNA, it is challenging to accurately detect low-frequency SVs or SVs involving complex junctions in ctDNA sequencing data. Here, we describe Aperture, a new fast SV caller that applies a unique strategy of $k$-mer-based searching, binary label-based breakpoint detection and candidate clustering to detect SVs and viral integrations with high sensitivity, especially when junctions span repetitive regions. Aperture also employs a barcode-based filter to ensure specificity. Compared with existing methods, Aperture exhibits superior sensitivity and specificity in simulated, reference and real data tests, especially at low dilutions. Additionally, Aperture is able to predict sites of viral integration and identify complex SVs involving novel insertions and repetitive sequences in real patient data. Aperture is freely available at https://github.com/liuhc8/Aperture.

Through network function virtualization (NFV), virtual network functions (VNFs) can be mapped onto substrate networks as service function chains (SFCs) to provide customized services with guaranteed Quality of Service (QoS). In this article, we solve a multi-SFC embedding problem by a game-theoretical approach considering the heterogeneity of NFV nodes, the effect of processing-resource sharing among various VNFs, and the capacity constraints of NFV nodes. Specifically, each SFC is treated as a player whose objective is to minimize the overall latency experienced by the supported service flow, while satisfying the capacity constraints of all NFV nodes. Due to processing-resource sharing, additional delay is incurred and incorporated into the overall latency for each SFC. The capacity constraints of NFV nodes are considered by adding a penalty term into the cost function of each player, and are guaranteed by a prioritized admission control mechanism. We prove that the formulated resource-constrained multi-SFC embedding game (RC-MSEG) is an exact potential game admitting at least one pure Nash equilibrium (NE) and has the finite improvement property (FIP). Two iterative algorithms are developed, namely, the best response (BR) algorithm with fast convergence and the spatial adaptive play (SAP) algorithm with great potential to obtain the best NE. Simulations are conducted to demonstrate the effectiveness of the proposed game-theoretical approach.

To efficiently allocate heterogeneous resources for customized services, in this article, we propose a network virtualization (NV)-based network architecture in cybertwin-enabled 6G core networks. In particular, we investigate how to optimize the virtual network (VN) topology (which consists of several virtual nodes and a set of intermediate virtual links) and determine the resultant VN embedding in a joint way over a cybertwin-enabled substrate network. To this end, we formulate an optimization problem whose objective is to minimize the embedding cost, while ensuring that the end-to-end (E2E) packet delay requirements are satisfied. The queueing network theory is utilized to evaluate each service’s E2E packet delay, which is a function of the resources assigned to the virtual nodes and virtual links for the embedded VN. We reveal that the problem under consideration is formally a mixed-integer nonlinear program (MINLP) and propose an improved brute-force search algorithm to find its optimal solutions. To enhance the algorithm’s scalability and reduce the computational complexity, we further propose an adaptively weighted heuristic algorithm to obtain near-optimal solutions to the problem for large-scale networks. Simulations are conducted to show that the proposed algorithms can effectively improve network performance compared to other benchmark algorithms.

Drone cell (DC) is envisioned to enable the dynamic service provisioning for radio access networks (RANs), in response to the spatial and temporal unevenness of user traffic. In this article, we propose a hierarchical deep reinforcement learning (DRL)-based multi-DC trajectory planning and resource allocation (HDRLTPRA) scheme for high-mobility users. The objective is to maximize the accumulative network throughput while satisfying user fairness, DC power consumption, and DC-to-ground link quality constraints. To address the high uncertainties of the environment, we decouple the multi-DC TPRA problem into two hierarchical subproblems, i.e., the higher level global trajectory planning (GTP) subproblem and the lower level local TPRA (LTPRA) subproblem. First, the GTP subproblem is to address trajectory planning for multiple DCs in the RAN over a long time period. To solve the subproblem, we propose a multiagent DRL-based GTP (MARL-GTP) algorithm in which the nonstationary state space caused by the multi-DC environment is addressed by the multiagent fingerprint technique. Second, based on the GTP results, each DC solves the LTPRA subproblem independently to control the movement and transmit power allocation based on the real-time user traffic variations. A deep deterministic policy gradient (DEP)-based LTPRA (DEP-LTPRA) algorithm is then proposed to solve the LTPRA subproblem. With the two algorithms addressing both subproblems at different decision granularities, the multi-DC TPRA problem can be resolved by the HDRLTPRA scheme. Simulation results show that 40% network throughput improvement can be achieved by the proposed HDRLTPRA scheme over the nonlearning-based TPRA scheme.

Unmanned aerial vehicles (UAVs)‐based sensor network is an effective mechanism for recognising and tracking of manoeuvring targets as well as expanding the monitoring coverage in the battlefield. However, there exists redundancy among the spectrum data collected by a UAV monitor within a data collection period, which may waste storage space and reduce the speed of data uploaded to the control centre. The authors assume that each UAV is equipped with an edge computing server and propose a delta compression method, which can save cache space and transfer time. First, they present a cost model and evaluation model for delta compression of the COPY/ADD class. Then an optimisation problem is formulated aiming to obtain the optimal delta encoding. Additionally, a maximal total length of copied fragments (MTLC) algorithm is proposed to find more mutually separated L ‐grams common fragments between the data collected at two adjacent moments. Theoretical analysis proves that the MTLC algorithm can generate a good delta encoding with the maximum total length and then the minimum total number of the COPYs. Moreover, numerical results show that MTLC has better performance in constructing a good delta encoding than the simple greedy and hash suffix array delta algorithms.

The paradigm of network function virtualization (NFV) with the support of software defined networking (SDN) emerges as a promising approach for customizing network services in fifth generation (5G) networks. In this paper, a multicast service orchestration framework is presented, where joint traffic routing and virtual network function (NF) placement are studied for accommodating multicast services over an NFV-enabled physical substrate network. First, we investigate a joint routing and NF placement problem for a single multicast request accommodated over a physical substrate network, with both single-path and multipath traffic routing. The joint problem is formulated as a mixed integer linear programming (MILP) problem to minimize the function and link provisioning costs, under the physical network resource constraints, flow conservation constraints, and NF placement rules; Second, we develop an MILP formulation that jointly handles the static embedding of multiple service requests over the physical substrate network, where we determine the optimal combination of multiple services for embedding and their joint routing and placement configurations, such that the aggregate throughput of the physical substrate is maximized, while the function and link provisioning costs are minimized. Since the presented problem formulations are NP-hard, low complexity heuristic algorithms are proposed to find an efficient solution for both single-path and multipath routing scenarios. Simulation results are presented to demonstrate the effectiveness and accuracy of the proposed heuristic algorithms.

Vehicular networks are expected to support diverse content applications with multi-dimensional quality of service (QoS) requirements, which cannot be realized by the conventional one-fit-all network management method. In this paper, a service-oriented hierarchical soft slicing framework is proposed for the cache-enabled vehicular networks, where each slice supports one service and the resources are logically isolated but opportunistically reused to exploit the multiplexing gain. The performance of the proposed framework is studied in an analytical way considering two typical on-road content services, i.e., the time-critical driving related context information service (CIS) and the bandwidth-consuming infotainment service (IS). Two network slices are constructed to support the CIS and IS, respectively, where the resource is opportunistic reused at both intra- and inter-slice levels. Specifically, the throughput of the IS slice, the content freshness (i.e., age of information) and delay performances of the CIS slice are analyzed theoretically, whereby the multiplexing gain of soft slicing is obtained. Extensive simulations are conducted on the OMNeT++ and MATLAB platforms to validate the analytical results. Numerical results show that the proposed soft slicing method can enhance the IS throughput by 30% while guaranteeing the same level of CIS content freshness and service delay.

Software-defined networking (SDN) and network function virtualization (NFV) together form a promising paradigm that enables the slicing of heterogeneous network resources for agile and efficient service customization. Among other techniques, virtual network function (VNF) mapping and scheduling are crucial to the deployment of SDN/NFV-enabled network services. In this paper, to enhance the performance of service provisioning, dynamic VNF mapping and scheduling are jointly investigated. Specifically, to achieve load balancing with QoS guarantee, we first formulate the VNF mapping and scheduling problem as a mixed integer linear programming (MILP). We then propose a two-stage online algorithm to address the NP-hardness of the MILP. In particular, when new service arrives, we map and schedule the VNFs on a service function chain (SFC) by greedily minimizing the waiting time of VNFs. If the delay requirement cannot be satisfied after the first stage, a delay-aware rescheduling scheme is triggered, in which selected existing VNFs are remapped and rescheduled. The proposed dynamic approach achieves flexible function placement and increases service acceptance ratio. Simulation results are provided to validate the effectiveness of the proposed algorithm.

Drone cell (DC) is an emerging technique to offer flexible and cost-effective wireless connections to collect Internet-of-things (IoT) data in uncovered areas of terrestrial networks. The flying trajectory of DC significantly impacts the data collection performance. However, designing the trajectory is a challenging issue due to the complicated 3D mobility of DC, unique DC-to-ground (D2G) channel features, limited DC-to-BS (D2B) backhaul link quality, etc. In this paper, we propose a 3D DC trajectory design for the DC-assisted IoT data collection where multiple DCs periodically fly over IoT devices and relay the IoT data to the base stations (BSs). The trajectory design is formulated as a mixed integer non-linear programming (MINLP) problem to minimize the average user-to-DC (U2D) pathloss, considering the state-of-the-art practical D2G channel model. We decouple the MINLP problem into multiple quasi-convex or integer linear programming (ILP) sub-problems, which optimizes the user association, user scheduling, horizontal trajectories and DC flying altitudes of DCs, respectively. Then, a 3D multi-DC trajectory design algorithm is developed to solve the MINLP problem, in which the sub-problems are optimized iteratively through the block coordinate descent (BCD) method. Compared with the static DC deployment, the proposed trajectory design can lower the average U2D pathloss by 10-15 dB, and reduce the standard deviation of U2D pathloss by 56%, which indicates the improvements in both link quality and user fairness.

The software defined networking (SDN) enabled network function virtualization (NFV) architecture emerges as a cost-effective solution for service customization in fifth generation (5G) networks. In this paper, a joint traffic routing and virtual network function (VNF) placement problem is studied for a multicast service request accommodated over a physical substrate network, where the multipath traffic routing is considered between embedded VNFs. The joint problem is formulated as a mixed integer linear programming (MILP) problem to minimize the provisioning cost of both VNFs and links, under the physical network resource constraints, flow conservation constraints, and VNF placement rules. Since the problem is NP-hard, low complexity heuristic algorithms, with the consideration of both the single-path and multipath routing cases, are proposed to determine an efficient solution. Simulation results are presented to demonstrate the effectiveness and accuracy of the proposed heuristic algorithms especially for a large-size network.

Network function virtualization (NFV) is one of the enabling technologies for fifth generation (5G) networks. How to allocate physical resources to customized network services both fairly and efficiently remains a challenging research issue in NFV. This paper proposes a two-stage approach to jointly optimize the chaining and embedding of virtual network functions (VNFs), to obtain feasible composition and embedding results with low complexity, while the average embedding cost is minimized and the total revenue is increased. In the first stage, the VNF chaining order is optimized based on the location and functionality of substrate nodes, and the ratio of outgoing data rate over incoming data rate for each required VNF. In the second stage, we allocate the physical resources based on the preliminary VNF ordering under the resource capacity constraints. A node splitting mechanism is also employed to improve the resource allocation fairness and increase the service acceptance ratio for the substrate network. Simulation results are presented to validate the feasibility and effectiveness of the proposed approach.

In this paper, an Air-Ground Integrated VEhicular Network (AGIVEN) architecture is proposed, where the aerial high-altitude platforms (HAPs) proactively push contents to vehicles through large-area broadcast, while the ground roadside units (RSUs) provide high-rate unicast services on demand. To efficiently manage the multi-dimensional heterogeneous resources, a service-oriented network slicing approach is introduced, where the AGIVEN is virtually divided into multiple slices and each slice supports a specific application with guaranteed quality of service (QoS). Specifically, the fundamental problem of multi-resource provisioning in AGIVEN slicing is investigated by taking into account the typical vehicular applications of location-based map and popularity-based content services. For the location-based map service, the capability of HAP-vehicle proactive pushing is derived with respect to the HAP broadcast rate and vehicle cache size, wherein a saddle point exists, indicating the optimal communication-cache resource trading. For the popular contents of common interests, the average on-board content hit ratio is obtained with HAPs pushing newly generated contents to keep on-board cache fresh. Then, the minimal RSU transmission rate is derived to meet the average delay requirements of each slice. The obtained analytical results reveal the service-dependent resource provisioning and trading relationships among RSU transmission rate, HAP broadcast rate, and vehicle cache size, which provides guidelines for multi-resource network slicing in practice. Simulation results demonstrate that the proposed AGIVEN network slicing approach matches the multi-resources across slices, whereby the RSU transmission rate can be saved by 40% while maintaining the same QoS.

Next generation (5G) wireless networks are expected to accommodate proliferation of connected devices and multimedia services. To support multimedia services in an agile, cost-effective, and flexible way, network virtualization is a potential solution. This paper investigates service- oriented network virtualization for 5G wireless networks, to efficiently allocate heterogeneous resources to accommodate multimedia services. Specifically, we study joint resource allocation for virtual network requests (VNRs) and online embedding the resultant VNRs in core networks (CNs). With the deployment of multiple traffic aggregation points (TAPs) in radio access networks (RANs), the end-to- end traffic from heterogeneous access technologies can be aggregated and then grouped based on their destinations. Queueing models are developed in determining the minimal capacity required at each core network element. Virtual network embedding (VNE) in the core network is further proposed to achieve efficient physical resource sharing in CNs. Simulation results validate the VNE process in core networks based on the optimized capacities.

Image saliency detection has been long studied, while several challenging problems are still unsolved, such as detecting saliency inaccurately in complex scenes or suppressing salient objects in the image borders. In this paper, we propose a new saliency detection algorithm in order to solving these problems. We represent the image as a graph with superpixels as nodes. By considering appearance similarity between the boundary and the background, the proposed method chooses non-saliency boundary nodes as background priors to construct the background probability model. The probability that each node belongs to the model is computed, which measures its similarity with backgrounds. Thus we can calculate saliency by the transformed probability as a metric. We compare our algorithm with ten-state-of-the-art salient detection methods on the public database. Experimental results show that our simple and effective approach can attack those challenging problems that had been baffling in image saliency detection.