In this paper, we propose a policy-guided Monte Carlo Tree Search (MCTS) decoder that achieves near maximum-likelihood decoding (MLD) performance for short block codes. The MCTS decoder searches for test error patterns (TEPs) in the received information bits and obtains codeword candidates through re-encoding. The TEP search is executed on a tree structure, guided by a neural network policy trained via MCTS-based learning. The trained policy guides the decoder to find the correct TEPs with minimal steps from the root node (all-zero TEP). The decoder outputs the codeword with maximum likelihood when the early stopping criterion is satisfied. The proposed method requires no Gaussian elimination (GE) compared to ordered statistics decoding (OSD) and can reduce search complexity by 95\% compared to non-GE OSD. It achieves lower decoding latency than both OSD and non-GE OSD at high SNRs.

Secure beamforming in reconfigurable intelligent surface (RIS)-assisted multiuser downlink systems is challenging due to high computational complexity and complex channel state information (CSI) estimation. This work proposes a pilot-driven beamforming network (PilotBeamNet) that jointly designs base-station (BS) transmit beamforming and quantized RIS phases directly from uplink pilot received signals with legitimate-user location cues to capture geometry. The framework avoids explicit channel estimation and slow iterative algorithms. A convolutional module reads each pilot frame, a long short term memory (LSTM) block with lightweight temporal attention aggregates them, and two simple heads output the beamformers and the discrete RIS phases. The location cues are embedded and fused with the features extracted from the pilot frames by the convolutional, LSTM, and temporal attention modules. Training maximizes ergodic secrecy rate (ESR) through Monte Carlo sampling of unknown eavesdropper channels, enabling robustness without requiring eavesdropper CSI. Once trained, PilotBeamNet performs single-pass inference with latency determined only by network depth. Across all tested conditions, PilotBeamNet achieves 10%–30% ESR improvement depending on the signal-to-noise ratio (SNR), pilot length, and RIS size, while reducing inference latency by more than an order of magnitude compared to alternating optimization (AO) and outperforming multilayer perceptron (MLP) baselines. It also maintains consistent performance under phase quantization and delivers higher secrecy rates across all evaluated configurations. • PilotBeamNet enables end-to-end secure RIS beamforming from uplink pilots without explicit channel estimation. • CNN-LSTM-attention predicts BS beamforming and quantized RIS phases with low latency. • Outperforms AO/MLP across SNRs and RIS sizes and remains robust to quantization.

Integrated sensing and communications (ISAC) systems promise dual use of spectrum and hardware for data transmission and environmental awareness. However, extending ISAC to satellite networks is challenged by high path loss, long delays, and the overhead of channel estimation. To address these challenges, we propose a channel-agnostic predictive beamforming framework for satellite ISAC (S-ISAC) within a crowdsourced bistatic architecture. Unlike conventional bistatic architectures that require a dedicated sensing receiver, our design aggregates echoes from multiple ground internet of things (IoT) devices (GIDs) in a crowdsourced manner to improve sensing performance without introducing any additional sensing equipment. We propose a model termed Historical Geometric-based LLM (HG-LLM) as a realization of the channel-agnostic predictive beamforming framework. HG-LLM learns to map historical geometric information (HGI) of the satellite, sensing target, and GIDs directly to future beamforming matrices, eliminating the need for channel state information (CSI). We propose two key modules in HG-LLM, namely, the Histogeometric Encoder, which transforms spatial-temporal data into LLM-compatible embeddings, and the TokenBeamformer, which translates the LLM outputs into optimized beamforming weights. Moreover, the backbone LLM is fine-tuned using low-rank adaptation for efficient adaptation to predictive beamforming tasks. Extensive simulations demonstrate that HG-LLM achieves performance levels comparable to channel-based methods across diverse settings, despite relying solely on HGI without requiring explicit CSI.

Cell-free massive multiple-input multiple-output (CF-mMIMO) is a promising architecture for 6G wireless networks through distributed access point (AP) cooperation. In large-scale deployments where user demand exceeds system capacity, effective user admission control (UAC) is essential to select users while meeting quality-of-service (QoS) requirements. The UAC problem in CF-mMIMO is inherently challenging, involving both discrete user selection and continuous power allocation variables. To address this challenge, we propose a Graphormer-enhanced Monte Carlo Tree Search (GE-MCTS) framework that integrates a Graphormer-based neural network (NN) with Monte Carlo Tree Search (MCTS). This framework leverages the Graphormer’s capability to model the graph-structured AP–user topology and MCTS’s planning proficiency to efficiently explore the vast decision space. Furthermore, to accommodate users initially unadmitted due to system constraints, we introduce a complementary AP deployment problem. By adapting the GE-MCTS framework, we optimize the placement of additional APs to achieve full user admission with the minimal number of new APs required. Simulation results demonstrate the effectiveness of our proposed framework. For UAC, with low computational complexity, GE-MCTS consistently admits 26.3–41.7% more users compared to baseline methods across various network scales. For AP deployment, our framework requires 45–73% fewer additional APs to achieve full user admission, highlighting its efficiency and scalability.

Emerging immersive services, such as virtual reality (VR), are featured by multi-modal data streams (audio, video, and haptic). The fusion of distinct service characteristics introduces a significant challenge to wireless communications. This paper considers hybrid VR video and haptic services, where VR video segments and haptic data packets are transmitted in two different time scales. To optimize resource utilization, we employ dynamic time division duplexing (TDD) to address the asymmetry in uplink/downlink (UL/DL) traffic. We formulate a dual time-scale optimization problem that minimizes overall bandwidth while satisfying both the round-trip delay of VR service and the reliability and latency requirements of haptic service. To address this problem, we propose a hierarchical deep reinforcement learning (DRL) framework: 1) deep deterministic policy gradient (DDPG) optimizes total bandwidth at the beginning of each video segment’s transmission; 2) a diffusion-based actor-critic (Diffusion-AC) algorithm determines the UL time resource ratio at each time slot, which is significantly shorter than the transmission duration of a video segment. Simulation results show that the proposed algorithm can reduce the overall bandwidth usage by 20% compared with baseline methods. In addition, it provides greater adaptability across diverse scenarios and converges faster than the baseline methods.

As a new communications architecture, the Space-Air-Ground integrated network (SAGIN) integrates satellites, airborne platforms, and terrestrial networks to enhance global connectivity and support robust and flexible communication capabilities. Efficient task offloading and resource allocation are crucial for SAGIN to meet the quality of service (QoS) requirements at low cost. In this paper, we formulate task offloading and resource allocation as a time-sequential decision-making problem, aiming to maximize task completion within available communication and computational resources. We propose an online approach referred to as graph fusion deep reinforcement learning (GF-DRL). GF-DRL incorporates a graph feature extraction network that utilizes a graph convolutional network (GCN) to extract features from both the task graph and user equipment (UE) graph, along with two attention mechanisms (hard and soft) to merge the two graphs. We also propose an action encoding and mapping network to generate both discrete (offloading) and continuous (allocation) decisions in an end-to-end manner. Simulation results validate the effectiveness of our proposed GF-DRL compared to state-of-the-art task offloading resource allocation approaches.

Accurate indoor tracking is a critical component of modern location-based services, fundamentally transforming the way we interact with indoor environments. Traditional state space model (SSM) based tracking often struggles in complex environments due to its reliance on fixed and oversimplified transition and observation functions. In this paper, we propose a novel deep state space model (DSSM) approach for indoor tracking that overcomes these limitations by leveraging trainable neural networks (NNs) in place of fixed transition and observation functions. The proposed DSSM retains the structured representation of SSMs while improving the ability to effectively capture the complex dynamics of both target movements and measurement errors. The proposed model incorporates physics constraints to enable self-supervised learning, eliminating the need for labeled data during training. We evaluate our framework using real-world time of flight (ToF) measurements, demonstrating its superior tracking accuracy compared to conventional methods.

Vital sign monitoring plays a key role in modern healthcare, supporting applications ranging from chronic disease management to more advanced elderly care. While traditional systems rely on contact-based devices, recent advances in WiFi sensing allow contactless monitoring using channel state information (CSI), offering a more convenient and unobtrusive approach. However, most existing WiFi-based methods mainly concentrate on monitoring in static conditions, rendering them unsuitable for real-world scenarios such as measuring respiration rate during walking. To address this gap, in this paper we propose MovBeat, a respiration prediction system designed to deliver high-accuracy respiratory monitoring when the subject is in motion. By integrating a contrastive learning framework with attention-based feature extraction, MovBeat effectively mitigates interference from the environment and body movements. Experimental results demonstrate that MovBeat achieves over 90 % accuracy in monitoring respiration on the move - an improvment of approximately 20 % compared to traditional methods. Comprehensive evaluations in diverse movement states, including both line-of-sight (LoS) and non-line-of-sight (NLoS) environments, demonstrate the robustness and generalizability of MovBeat in real-world scenarios.

Space-air-ground integrated networks (SAGIN), a cornerstone of 6G, face challenges in task offloading and resource allocation (TORA) due to their heterogeneity, dynamic topology, and high mobility. To address these, we formulate a multi-objective TORA problem under dynamic topologies to minimize latency and inter-node power consumption. We propose a dynamic heterogeneous graph neural network (DHGNN) that, combined with reinforcement learning, adaptively updates node features to capture cross-domain dependencies. Simulations demonstrate that our method outperforms existing GDRL approaches in reward, latency, and power efficiency.

Radar imaging provides non-contact, privacy-preserving, and environmentally robust monitoring for continuous human motion recognition (HMR) by leveraging diverse information embedded in various radar signal domains. However, current research has not effectively integrated multi-radar and multi-domain imaging to fully exploit the benefits of distributed radar systems. To bridge this gap, we propose a multi-radar, multi-domain parallel cross-attention model with four key components: intra-domain cross-radar weight sharing encoders specific to each domain for consistent feature extraction and parameter reduction, domain-level parallel cross-attention (DLPCAN) modules to fuse domain-specific features and enhance feature representation robustness in each radar, a source-level attention fusion (SLAF) module to highlight significant features from multiple radar inputs, and two bi-directional gated recurrent unit (BiGRU) modules to capture temporal information. The model is trained using connectionist temporal classification (CTC) loss for effective sequence prediction. By integrating data from multiple radar nodes and domains, our approach significantly improves continuous HMR performance compared to single radar systems and single domain data. Comparative evaluations demonstrate that our model outperforms state-of-the-art radar imaging-based HMR solutions.

In this study, we investigate the use of simultaneous transmitting and reflecting reconfigurable intelligent surfaces (STAR-RISs) mounted on energy-efficient uncrewed aerial vehicles (UAVs) to support satellite Internet of Things (IoT) communications served by low-Earth orbit (LEO) satellites. First, we propose a STAR-RIS-equipped UAV framework termed integrated STAR-RIS and UAV (ISRU). Then, we aim to optimize energy efficiency by jointly adjusting the UAV’s flight path, STAR-RIS phase-shifts, and power allocation among IoT devices, all while maintaining equitable user fairness level. However, solving this problem presents considerable challenges due to the nonconvexity and NP-hardness properties of the objective function and constraints. To address, our work introduces a Dinkelbach-based alternating optimization (AO) procedure termed integrated trajectory, phase-shift, and power allocation (ITPP). Our simulation results show that the integration of ISRU and ITPP can achieve 67% higher sum-rates than non-ISRU schemes and save up to 40% more energy than unoptimized trajectory schemes.

Space-Air-Ground integrated network (SAGIN) is a crucial component of the 6G, enabling global and seamless communication coverage. This multi-layered communication system integrates space, air, and terrestrial segments, each with computational capability, and also serves as a ubiquitous computing platform. An efficient task offloading and resource allocation scheme is key in SAGIN to maximize resource utilization efficiency, meeting the stringent quality of service (QoS) requirements for different service types. In this paper, we introduce a dynamic SAGIN model featuring diverse antenna configurations, two timescale types, different channel models for each segment, and dual service types. We formulate a problem of sequential decision-making task offloading and resource allocation. Our proposed solution is an innovative online approach referred to as graphic deep reinforcement learning (GDRL). This approach utilizes a graph neural network (GNN)-based feature extraction network to identify the inherent dependencies within the graphical structure of the states. We design an action mapping network with an encoding scheme for end-to-end generation of task offloading and resource allocation decisions. Additionally, we incorporate meta-learning into GDRL to swiftly adapt to rapid changes in key parameters of the SAGIN environment, significantly reducing online deployment complexity. Simulation results validate that our proposed GDRL significantly outperforms state-of-the-art DRL approaches by achieving the highest reward and lowest overall latency.

Accurate estimation and detection, along with phase shift optimization, are vital for implementing reconfigurable intelligent surface (RIS)-enabled multi-antenna systems in highly disruptive industrial IoT environments. Motivated by the remarkable capabilities of deep learning (DL) techniques, this paper introduces a pioneering approach to address challenges in channel estimation, channel correlation prediction, and symbol detection for industrial IoT. We develop an optimization framework for large-scale IoT deployments to maximize the signal-to-interference-plus-noise ratio (SINR) while minimizing transmit power. We also propose a transformer-based channel correlation predictor for IoT devices, which enables adaptive pilot retransmissions and reduces training overhead through a co-design approach that integrates communication, computation, and control. Extensive simulations under realistic, time-varying industrial IoT channel conditions demonstrate the superiority of our DL-driven approach, achieving significant improvements in detection accuracy and SINR.

Approximation-based energy-efficient cyber-secured image classification framework

Automatic modulation recognition (AMR) is a critical technology in wireless communications, aiming to achieve high recognition accuracy with low complexity in increasingly intricate electromagnetic environments. To tackle this challenge, in this paper, we propose a dual-branch convolution cascaded transformer network with feature assistance, termed DCTFANet. To enhance the differentiation between samples, we employ the gramian angular field (GAF) to capture potential temporal correlations between each data point. Subsequently, both I/Q sequences and GAF data are input into the model for joint signal feature extraction. The network backbone is constructed using multiple improved depthwise separable convolution (DSC) blocks, which significantly reduce computational complexity. Moreover, the backbone depth is flexibly adjustable to fully exploit local features of different data types. Finally, feature transition and the transformer encoder are used to reduce parameters and extract global feature. Experimental results on RML2016.10b show that the proposed method achieves higher recognition accuracy compared to several state-of-the-art methods, especially at low signal-to-noise ratios (SNRs), with an increase of at least 10.80% at -20dB.

In the context of massive machine-type communications (mMTC) within the Internet of Things (IoT), joint activity detection and channel estimation (JADCE) is a key challenge in enabling massive access due to sporadic device access patterns. In this work, we consider both single-antenna and multiple-antenna base station scenarios and formulate the JADCE problem using the least absolute shrinkage and selection operator (LASSO) framework. To address this problem, we propose a model-driven network that utilizes compressive sensing (CS) and deep learning techniques. Specifically, we first design a prediction-correction alternating direction method of multipliers (PC-ADMM) as the underlying algorithm of the model-driven network. Then, the network is developed based on the PC-ADMM and is designed to be complex-valued. Furthermore, we also model the proximal operator, typically used to generate sparse solutions in LASSO, as a channel attention module within the model-driven network to enhance robustness. Numerical results show that the proposed PC-ADMM framework outperforms existing LASSO-based methods in terms of channel estimation and device activity detection.

In the vehicular edge networks (VEN), task migration is complicated by issues like vehicle movement, diverse resource allocation, and integrating sensing with communication technologies. This paper presents a task migration strategy to optimize task flow under limited resources in PMN-assisted VEN. Vehicles can send public and private tasks to roadside units (RSUs), constrained by bandwidth, computational power, and storage space. Public tasks aim at data collection for road transportation management, while private tasks cover a spectrum of services from work to entertainment. To address the limitations imposed by resource scarcity and meet the demands of task migration, we have developed a dual-layer graph neural network (GNN) that leverages vehicle mobility patterns. In particular, the first layer of GNN acquires vehicle information and the latest surrounding information, and sends it to the nearby RSU. Considering the variety of tasks and multi-dimensional resource constraints, the second GNN layer forecasts RSU resource availability and vehicular trajectories. Subsequently, a task-based maximum flow algorithm (T-MFA) is proposed to refine task migration paths and resource allocation strategies to maximize task flow. Simulation experiments validate the efficacy of the proposed algorithm, demonstrating its capability to achieve optimal task migration by accommodating differences in tasks, resources, and capacities.

With its dual-functional advantages, integrated sensing and communications (ISAC) technologies can be further extended to satellite communications, enhancing global coverage services. However, achieving vast coverage would result in significant delays and considerable path losses. Motivated by this, in this paper, we focus on satellite-based ISAC (S-ISAC) systems and propose a general transceiver design framework incorporating both transmit waveform and receive filter. Unlike existing approaches, our approach uses a predictive joint transmit waveform and receive filter design that eliminates the need of channel estimation, thereby reducing time overhead. Additionally, a versatile weighting mechanism is designed to allow flexible prioritization between communications and sensing. To tackle the intractability of the ISAC transceiver design problem, we adopt a data-driven deep learning-based approach, where the model learns to design the transmit waveform and receive filter from historical channel data. Specifically, we propose a predictive optimization network (PONet), leveraging convolutional layers and a Transformer encoder to capture long-term spatial-temporal features and facilitate the learning capability. Numerical results demonstrate the effectiveness of the proposed PONet in terms of communications and sensing rates in S-ISAC networks in various system settings.

Space-Air-Ground integrated network (SAGIN) is a crucial component of the 6G, enabling global and seamless communication coverage. An efficient task offloading and resource allocation scheme is key in SAGIN to maximize resource utilization efficiency, meeting the stringent quality of service (QoS) requirements for different service types. In this paper, we introduce a dynamic SAGIN model featuring diverse antenna configurations, two timescale types, different channel models for each segment, and dual service types. We formulate a problem of sequential decision-making task offloading and resource allocation. Our proposed solution is an innovative online approach referred to as graphic deep reinforcement learning (GDRL). This approach utilizes a graph neural network (GNN)-based feature extraction network to identify the inherent dependencies within the graphical structure of the states. Simulation results validate that our proposed GDRL significantly outperforms state-of-the-art deep reinforcement learning (DRL) approaches by achieving the highest reward and lowest overall latency.

Mobile edge computing (MEC) enables mobile devices (MDs) to offload their computational tasks to the network edge, significantly reducing transmission delay and energy consumption. In this paper, we develop a novel partial NOMA (PNOMA) based task offloading scheme in a multi-layer mobile computing network (MD-MEC-Cloud). PNOMA combines the high throughput of NOMA with the low interference of OMA for efficient, low-latency transmission. Furthermore, the PNOMA-based multi-layer collaborations enable rapid task processing across various computing requirements. We formulate a non-convex mixed-integer optimization problem aimed at minimizing the average delay across all MDs. To address this challenging problem, we propose an algorithm called reincarnating proximal policy optimization (RPPO), which uses online inference solutions to significantly reduce complexity. In addition, we incorporate accumulated apriori information into RPPO for fast retraining and design both a reward function and an evaluation phase to ensure the communication/computation constraints are met with a high probability. Simulation results demonstrate that the proposed task offloading scheme outperforms existing methods.

WiFi sensing has emerged as a promising technique in the healthcare industry, enabling contact-free monitoring of vital signs by detecting changes in WiFi signals resulting from physiological activities. State-of-the-art WiFi sensing uses channel state information (CSI) to analyze signal characteristics, capturing subtle changes due to heartbeats and breathing. However, existing methods face challenges in concurrently measuring respiration and heart rates, and they exhibit high sensitivity to environmental factors and individual differences, limiting the detection accuracy of a trained model in real-world environments. In this paper, we propose a novel multi-task contrastive learning framework for concurrent detection of respiration and heart rates. We introduce multi-task learning with hard-shared layers to exploit the physiological link between breathing and heartbeat. Additionally, we leverage contrastive learning to improve our model's ability to differentiate and prioritize CSI changes related to respiratory and cardiac activi-ties. The experimental results demonstrate the proposed model's ability to accurately measure respiratory and heart rates in challenging scenarios, including long-distance and non-line-of-sight conditions, even when utilizing omnidirectional antennas.

Given the popularity of flawless telepresence and the resultants explosive growth of wireless video applications, besides handling the traffic surge, satisfying the demanding user requirements for video qualities has become another important goal of network operators. Inspired by this, cooperative edge caching intrinsically amalgamated with scalable video coding is investigated. Explicitly, the concept of a Pareto-optimal semi-distributed multiagent multipolicy deep reinforcement learning (SD-MAMP-DRL) algorithm is conceived for managing the cooperation of heterogeneous network nodes. To elaborate, a multipolicy reinforcement learning algorithm is proposed for finding the Pareto-optimal policies during the training phase, which balances the teletraffic versus the user experience tradeoff. Then the optimal policy/solution can be activated during the execution phase by appropriately selecting the associated weighting coefficient according to the dynamically fluctuating network traffic load. Our experimental results show that the proposed SD-MAMP- acrshort DRL algorithm: 1) achieves better performance than the benchmark algorithms and 2) obtains a near-complete Pareto front in various scenarios and selects the optimal solution by adaptively adjusting the above-mentioned pair of objectives.

Remote sensing image dehazing is crucial for both military and civil applications. However, dehazed remote sensing images often suffer from pronounced artifacts and tend to overestimate the atmospheric light value. We propose a novel dehazing method based on heterogeneous priors. Specifically, superpixels are extracted from the hazy remote sensing image using a depth-based simple linear iterative clustering superpixel segmentation (DSLIC) algorithm. These superpixels serve as cells for transmission and atmospheric light estimation. To improve the robustness of atmospheric light estimation, we develop an atmospheric light value-map fusion estimation (ALFE) model that integrates the heterogeneous priors-guided haze concentration model (HP-HCM) to derive the global atmospheric light value, while utilizing the bright channel value within each superpixel as the local atmospheric light map. We also introduce a dynamic dehazing intensity parameter (DDIP) model, which refine the transmission map based on the HP-HCM. Extensive comparative experiments validate the superior performance of the proposed method. The PSNR and SSIM achieved by our method exceed those of the dark channel prior (DCP) by 22.2% and 37.5%, respectively.

Driven by digital twin (DT) technology, the industrial Internet of Things (IIoT) is expanding to open up new frontiers in industrial applications. However, traditional DT modeling approaches require synchronizing massive amounts of data, resulting in high communications overhead and privacy vulnerability. To address this problem, this article proposes a novel DT architecture for IIoT, where the DT can showcase the real-time operating status of the industrial environment. Swarm learning (SL) is an emerging decentralized federated learning (FL) technique that eliminates the need of a centralized server. We present a novel credibility-weighted SL scheme to construct the DT models, which improves data security while ensuring the fairness of participants as opposed to conventional FL. In addition, we develop a DT-assisted deep reinforcement learning algorithm for simultaneously optimizing the system reliability and energy consumption of IIoT. Simulation comparisons demonstrate that the proposed scheme outperforms some state-of-the-art benchmarks in terms of both reliability and energy consumption.

Network slicing is a key enabler of 5G and beyond networks to satisfy the diverse quality of service (QoS) requirements of different services simultaneously. In network slicing, radio access network (RAN) slicing is essential to establish a functional network slice by connecting mobile devices and mapping virtualized resource units to different slices. This requires a highly efficient resource allocation scheme to maximize resource utilization efficiency and meet the diverse QoS requirements. In this paper, we propose a dynamic RAN slicing model that incorporates multiple distributions to accommodate different user request types and diverse priorities among traffic types in the same slice, where the total available resources are dynamically changing over time. We formulate resource allocation as a time-sequential dynamic optimization problem that takes into account system stability, resource limitation, different timescales, long-term system performance, and user priority. We propose a deep reinforcement learning-based (DRL-based) approach referred to as prediction-aided weighted DRL (PW-DRL) to online infer the power allocation and user acceptance decisions that can maximize a predefined reward function. Additionally, a prediction network is formulated to capture the correlation between current and future states. Simulation results validate that our proposed PW-DRL significantly outperforms state-of-the-art approaches by achieving the highest long-term reward and fastest convergence.

The ever-increasing number of ubiquitous Internet of Things (IoT) applications entails a high demand for scarce communication and network resources. To meet this stringent requirement, mobile edge computing (MEC) is envisioned as a transformative technique to significantly streamline the existing network operations. Recently, device-to-device (D2D) communication has been proposed as a promising technology in 5G and beyond networks with a significantly increased transmission efficiency, especially suitable for small-packet task exchanges. In this paper, we incorporate D2D communication into the multi-layer computing network and propose a full-dimensional task offloading scheme by jointly optimizing task offloading decisions and computation/communication resource allocation. We formulate it as mixed-integer nonlinear programming (MINLP) problem, where the optimal branch-and-bound (B&B) algorithm with the full strong branching (FSB) variable selection policy features an extremely high complexity. To address this challenge, we propose inverse reinforcement learning with graph neural networks (GIRL) to generate a new variable selection policy that closely matches the FSB variable selection. Without sacrificing the global optimality, the GIRL can directly infer the variable selection with a much lower complexity, significantly accelerating the original B&B algorithm. Simulation results show that the GIRL achieves a lower complexity without sacrificing the global optimality. Furthermore, our proposed full-dimensional task offloading scheme achieves better performance than the existing schemes in terms of average delay for all mobile devices (MDs).

Relying on a data-driven methodology, deep learning has emerged as a new approach for dynamic resource allocation in large-scale cellular networks. This paper proposes a knowledge-assisted domain adversarial network to reduce the number of poorly performing base stations (BSs) by dynamically allocating radio resources to meet real-time mobile traffic needs. Firstly, we calculate theoretical inter-cell interference and BS capacity using Voronoi tessellation and stochastic geometry, which are then incorporated into a neural network as key parameters. Secondly, following the practical assessment, a performance classifier evaluates BS performance based on given traffic-resource pairs as either poor or good. Most importantly, we use well-performing BSs as source domain data to reallocate the resources of poorly performing ones through the domain adversarial neural network. Our experimental results demonstrate that the proposed knowledge-assisted domain adversarial resource allocation (KDARA) strategy effectively decreases the number of poorly performing BSs in the cellular network, and in turn, outperforms other benchmark algorithms in terms of both the ratio of poor BSs and radio resource consumption.

Network slicing is key to enabling 6G and beyond networks to simultaneously meet the diverse quality of service (QoS) requirements of various services. In network slicing, radio access network (RAN) slicing is essential to establish a functional network slice by connecting mobile devices and mapping virtualized resource units to different slices. This demands an efficient resource allocation scheme that maximizes resource utilization while meeting diverse QoS requirements. In this paper, we propose a new dynamic resource allocation framework that encompasses three types of services. We formulate a dynamic resource allocation problem that features a mixed action space and has both long-term power and instantaneously available resource unit constraints. We propose a deep reinforcement learning (DRL)-based approach referred to as prediction-aided weighted DRL (PW-DRL), which infers the power allocation and user acceptance decisions to maximize a predefined reward function. Additionally, we propose a prediction network that significantly improves the DRL learning process under limited resources by supplying future state information. Simulation results validate that our proposed PW-DRL significantly outperforms state-of-art DRL approaches by achieving the highest long-term reward and fastest convergence.

Cell-free massive multiple-input multiple-output (MIMO) deploys a large number of distributed access points (APs) without cell edges, offering seamless connectivity with significantly increased spectral efficiency and system capacity, but suffering degraded energy efficiency. In this paper, we develop a green energy scheme by simultaneously optimizing power allocation and AP selection. We formulate it as a non-convex mixed-integer nonlinear programming problem (MINLP), which is NP-hard. To address this challenging problem, we propose a learning-based algorithm that embeds non-convex optimization into contemporary deep reinforcement learning (DRL), referred to as optimization-embedded soft actor-critic with graph transformer networks (OSAC-G). OSAC-G enjoys the benefits of directly online inferring solutions for the non-convex problem with a much lower computational complexity compared to conventional non-convex optimization. Simulation results demonstrate that the green energy scheme significantly decreases energy consumption compared to the existing ones.

The rapid development of Internet of Things (IoT) applications requires efficient computing and communication resource allocation strategies to streamline the existing network operations. These strategies could be formulated as mixed-integer nonlinear programming (MINLP) problems, where the optimal branch-and-bound (B&B) with the full strong branching (FSB) variable selection policy features an extremely high complexity. We propose inverse reinforcement learning with graph neural networks (GNNIRL) to generate a new variable selection policy that closely matches the FSB variable selection. Without sacrificing the optimality, the GNNIRL can directly infer the variable selection with a significantly lower complexity, which is also verified by simulation.

This paper addresses the blind recovery of the parity check matrix of an (n, k) linear block code over noisy channels by proposing a fast recovery scheme consisting of 3 parts. Firstly, this scheme performs initial error position detection among the received codewords and selects the desirable codewords. Then, this scheme conducts Gaussian elimination (GE) on a k-by-k full-rank matrix and uses a threshold and the reliability associated to verify the recovered dual words, aiming to improve the reliability of recovery. Finally, it performs decoding on the received codewords with partially recovered dual words. These three parts can be combined into different schemes for different noise level scenarios. The GEV that combines Gaussian elimination and verification has a significantly lower recovery failure probability and a much lower computational complexity than an existing Canteaut-Chabaud-based algorithm, which relies on GE on n-by-n full-rank matrices. The decoding-aided recovery (DAR) and error-detection-&-codeword-selection-&-decoding-aided recovery (EDCSDAR) schemes can improve the code recovery performance over GEV for high noise level scenarios, and their computational complexities remain much lower than the Canteaut-Chabaud-based algorithm.

General Instance Re-identification is a very important task in computer vision, which can be widely used in many practical applications, such as person/vehicle re-identification, face recognition, wildlife protection, commodity tracing, snapshots, and so on. To meet the increasing application demand for general instance re-identification, we present FastReID as a widely used software system. In FastReID, the highly modular and extensible design makes it easy for the researcher to achieve new research ideas. Friendly manageable system configuration and engineering deployment functions allow practitioners to quickly deploy models into productions. We have implemented some state-of-the-art projects, including person re-id, partial re-id, cross-domain re-id, and vehicle re-id. Moreover, we plan to release these pre-trained models on multiple benchmark datasets. FastReID is by far the most general and high-performance toolbox that supports single and multiple GPU servers, it can reproduce our project results very easily. The source codes and models have been released at https://github.com/JDAI-CV/fast-reid.

Dehazing based on deep learning has attracted a lot of attention recently. Most dehazing networks seldom consider two critical features of real outdoor-scene haze,i.e., depth and haze density, resulting in degraded performance on real hazy images compared with synthetic hazy images. Moreover, the uncertainty problem is crucial in the image restoration field, but it is often ignored. In this letter, we propose a novel multi-scale density-aware network (MSDAN) for single image dehazing, where a key dual feedback module (DFB) is proposed and embedded in the decoder part of MSDAN. Furthermore, the DFB includes a feedforward mechanism and two feedback mechanisms: feature feedback (FF) and transmission feedback (TF). Specifically, the feedforward mechanism predicts a low-scale transmission map ($t$-map), while FF and TF aim to enhance confident features to reduce model uncertainty in the training process and correct features by introducing depth and density information. In addition, two novel modules: confident feature attention module (CFA) and transmission adjustment module (TADJ) are proposed as cores for confident features estimation of FF and TF, respectively. Extensive quantitative and qualitative experiments are conducted on several public datasets, which demonstrate that the proposed algorithm outperforms the state-of-the-art algorithms.

The booming of the industrial Internet of Things (IIoT) brings an exponential increase in industrial devices, calling for more flexible and low-cost communications. The fifth generation and beyond (B5G) communication technologies provide a dedicated solution by supporting two industry-targeted technologies: Massive machine-type communications (mMTC) and ultra reliable low-latency communications (URLLC). In this article, we design a B5G-aided quality test system in advanced manufacturing, where various sensors are connected to the base station (BS) and send contextual information via mMTC. The BS and quality test machine transmit short length commands and small size feedback to each other, respectively, via URLLC. We formulate a long-term optimization problem to improve the product qualification rate by maximizing the expected average reward with limited testing capacity and changing configurations. To address this problem, we develop a novel context-aware combinatorial quality test (CC-QT) algorithm based on bandit learning (BL), which integrates contextual information to predict the product quality, and a combinatorial method to decrease the complexity of the BL process. Furthermore, we derive a performance upper bound of the proposed CC-QT and analyze its computational complexity. Experimental results illustrate the performance of CC-QT and substantiate its superiority over the existing algorithms.

The accelerated development of the Industrial Internet of Things (IIoT) is catalyzing the digitalization of industrial production to achieve Industry 4.0. In this article, we propose a novel digital twin (DT) empowered IIoT (DTEI) architecture, in which DTs capture the properties of industrial devices for real-time processing and intelligent decision making. To alleviate data transmission burden and privacy leakage, we aim to optimize federated learning (FL) to construct the DTEI model. Specifically, to cope with the heterogeneity of IIoT devices, we develop the DTEI-assisted deep reinforcement learning method for the selection process of IIoT devices in FL, especially for selecting IIoT devices with high utility values. Furthermore, we propose an asynchronous FL scheme to address the discrete effects caused by heterogeneous IIoT devices. Experimental results show that our proposed scheme features faster convergence and higher training accuracy compared to the benchmark.

Integrating mobile edge computing (MEC) in the ultra-dense network (UDN) is a key enabler to meet the service demand by allowing smart devices to perform uninterrupted task offloading via densely deployed MEC servers. In most cases, the smart devices randomly move around the whole network. Consequently, the popular ‘`MEC-centralized decision’' offloading approach could be inapplicable, as joint decision-making among multiple MEC servers becomes difficult due to time synchronization and information exchange overhead. In this paper, we take a user-centric approach to minimize a long-term delay for a given task duration under a price budget constraint. To address this problem, we develop a novel contextual sleeping bandit learning (CSBL) algorithm, which integrates contextual information and sleeping characteristic to accelerate the learning convergence and leverage Lyapunov optimization to deal with the price budget constraint. Furthermore, we extend to a multiple offloading scenario where multiple MEC servers can be selected in each offloading round and propose a CSBL-multiple (CSBL-M) algorithm to address the exponential increase of the offloading selections. For both CSBL and CSBL-M, we derive the upper bounds of learning regret and provide rigorous proofs that they asymptotically approach the Oracle algorithm within bounded deviations for finite task duration.

The industrial Internet of Things (IIoT) interconnects an exponential number of industrial devices, and more flexible and low-cost communications are widely in demand. The fifth-generation (5G) communication provides two industrial-target technologies, massive machine-type communications (mMTC) and ultra-reliable low-latency communications (URLLC), to meet the demand. We design a 5G-aided quality test system, where various sensors are connected to the base station (BS) and send contextual information via mMTC. The BS and quality test machine transmit short-length commands and small-size feedback to each other via URLLC. The problem is formulated as a long-term optimization one with the purpose of improving the product qualification rate. We develop a novel contextual combinatorial quality test (CC-QT) algorithm to solve the problem. We further derive a performance upper bound of the proposed CC-QT and analyze its computational complexity. Experimental results illustrate the performance of CC-QT and substantiate its superiority over the existing algorithms.

A tremendous amount of visual data are bing collected by the Internet of Video Things (IoVT) systems in which ubiquitous cameras deployed in cities enable new applications in the domains of smart transportation and public security. However, the limited resources in terms of communication, computing, and caching (3C) in the conventional cellular network make it challenging to adopt centralized artificial intelligence (AI) to conduct real-time video-based data analytics. In this work, based on the 5G network architecture with edge servers, a three-phase resource-effective solution is proposed to perform surveillance operations in a large-scale wireless IoVT. The proposed strategy integrates front-end cameras with simple on-chip neural networks performing real-time object-of-interest segmentation, edge servers, and cloud servers with AI functionality carrying out image-based target recognition and video-based target analytics tasks. More importantly, we design the optimal 3C strategy to achieve the best video analytics performance constrained by computing offload ratio, network resource allocation and video-related parameters. Extensive simulations with deep neural networks implemented both at the front-end cameras and in the cloud server have validated the effectiveness of the proposed solution.

Sparse code multiple access (SCMA) is a promising non-orthogonal multiple access technique to support massive connectivity for future wireless Internet of Things (IoT) networks. As the main feature of SCMA, modulation and spread spectrum is embedded into codebook mapping, offering significant codebook shaping gains to mitigate inter-cell interference. Maximizing the constellation-constrained average mutual information (AMI) is an effective way for SCMA codebook optimization. However, deriving a closed-form expression of the AMI is analytically intractable, while it is computationally costly to estimate the AMI by numerical methods. To address this challenge, this paper first derives a lower bound of the AMI with a closed-form expression. On this basis, we propose a novel codebook optimization method referred to as joint bare bones particle swarm optimization (JBBPSO) through maximizing the AMI lower bound. The proposed low-complexity method jointly optimizes the mother codebook including basic constellation and other non-zero-dimensional constellations, and the rotation angles of multiple users. Numerical results show that our proposed optimized codebooks outperform the state-of-the-art SCMA codebooks in terms of both the lower bound and the error performance.

The integration of mobile edge computing (MEC) into an ultra-dense network (UDN) can provide ubiquitous task offloading services to computation-demanding users leveraging densely deployed micro base stations. The conventional multi-user task offloading strategies are performed centrally, where a central node makes global task offloading decisions on server selection and resource allocation. In practice, the deployment becomes prohibitively complex with the increasing number of users as it involves high communication overhead and complex global optimization operations. In this paper, we develop a novel decentralized task offloading strategy in UDN, enabling users to independently make local task offloading decisions. We formulate the associated optimization problem to minimize the long-term average task delay among all users. On this basis, we develop a novel calibrated contextual bandit learning (CCBL) algorithm, where users can learn the computational delay functions of micro base stations and predict the task offloading decisions of other users in a decentralized manner. The convergence of the proposed CCBL algorithm is verified via the approachability theory. Moreover, we transfer the target of calibrated learning from all micro base stations to a single user and propose a user-oriented CCBL algorithm to further decrease the computational complexity and increase the convergence rate. Simulation results illustrate that our proposed algorithm outperforms the existing decentralized algorithms and approaches the centralized one.

The rapid development of 5G and Internet-of-Things catalyzes ever-increasing computation-intensive and delay-sensitive applications demanding ubiquitous computation services. Integrating mobile edge computing (MEC) in the ultra-dense network (UDN) is a key enabler to meet the service demand by allowing smart devices to perform uninterrupted task offloading via densely deployed MEC servers. In this paper, we take a user-oriented approach to minimize a long-term delay for a given task duration under a price budget constraint. To address this problem, we develop a novel contextual sleeping bandit learning (CSBL) algorithm, which integrates context information and sleeping bandit theory to handle the fast changing environment and leverages Lyapunov optimization to deal with the price budget. We derive the upper bound of learning regret and provide a rigorous proof that CSBL asymptotically approaches the Oracle algorithm within bounded deviations for finite task duration. Simulation results illustrate that CSBL significantly outperforms existing algorithms.

Envisioned as a promising component of the future wireless Internet-of-Things (IoT) networks, the non-orthogonal multiple access (NOMA) technique can support massive connectivity with a significantly increased spectral efficiency. Cooperative NOMA is able to further improve the communication reliability of users under poor channel conditions. However, the conventional system design suffers from several inherent limitations and is not optimized from the bit error rate (BER) perspective. In this article, we develop a novel deep cooperative NOMA scheme, drawing upon the recent advances in deep learning (DL). We develop a novel hybrid-cascaded deep neural network (DNN) architecture such that the entire system can be optimized in a holistic manner. On this basis, we construct multiple loss functions to quantify the BER performance and propose a novel multi-task oriented two-stage training method to solve the end-to-end training problem in a self-supervised manner. The learning mechanism of each DNN module is then analyzed based on information theory, offering insights into the explainable DNN architecture and its corresponding training method. We also adapt the proposed scheme to handle the power allocation (PA) mismatch between training and inference and incorporate it with channel coding to combat signal deterioration. Simulation results verify its advantages over orthogonal multiple access (OMA) and the conventional cooperative NOMA scheme in various scenarios.

Mobile computing network is envisioned as a powerful framework to support the growing computation-intensive applications in the era of the Internet of Things (IoT). In this paper, we exploit the potential of a multi-layer network via a two-dimensional (2-D) task offloading scheme, which enables horizontal cooperations among the edge nodes. To minimize the average task offloading delay for all the mobile users, we formulate a mixed non-linear programming (MINLP) by jointly optimizing the 2-D offloading decisions and communication/computational resource allocation. To address this very challenging problem, we exploit the unique algorithmic structure of the optimal branch-and-bound (B&B) algorithm, and propose a novel Gaussian process imitation learning (GPIL) method to learn how to discover the shortcut for node searching in the B&B enumeration tree and significantly accelerate the B&B algorithm. When the network key parameters change, we further propose a novel recursive GPIL (RGPIL) method to agilely adapt to the new scenario with a fast policy update, where the new posterior distribution can be recursively updated based on a few new training data. Our simulation results show that the proposed method can achieve a near optimal solution with a significantly reduced complexity (e.g., a reduction of 98.7% in the number of searched nodes for a typical case). On this basis, the advantage of 2-D offloading scheme over the conventional schemes is also verified.

The cooperative relay network is a type of multi-terminal communication system. We present in this paper a Neural Network (NN)-based autoencoder (AE) approach to optimize its design. This approach implements a classical three-node cooperative system as one AE model, and uses a two-stage scheme to train this model and minimize the designed losses. We demonstrate that this approach shows performance close to the best baseline in decode-and-forward (DF), and outperforms the best baseline in amplify-and-forward (AF), over a wide range of signal-to-noise-ratio (SNR) values. It is also shown that training at a list of mixed SNR values can improve the error performance compared to training at a fixed SNR value. Moreover, to verify the robustness of the trained AE model, we test it under the effect of impulse-noise.

Mobile-edge computing (MEC) is a promising technology to support computation-intensive and delay-sensitive applications at smart devices by offloading their local tasks to the network edge. In this paper, we propose a novel index based real-time task offloading policy for an asynchronous large-scale MEC system. We first formulate the policy design as a restless multi-armed bandit (RMAB) to capture the stochasticity and criticality in tasks. Based on the Whittle index theory, we then rigorously establish the indexability of our RMAB and derive a closed-form solution, making it scalable to the number of users and extremely simple to implement in practice. Simulation results show that the propose policy can achieve a significant performance improvement in term of the accumulative reward and completion ratio, compared with some existing policies.

Mobile traffic prediction opens a promising avenue to demand-aware large-scale resource allocation with a significant improvement in the spectral efficiency. Various long-term prediction methods have been proposed in the literature. However, when considering the stringent requirement of the real-time and efficient radio resource allocation for future wireless communications, developing short-term prediction methods with high prediction accuracy is more desirable. In this paper, we exploit spatiotemporal correlations among the mobile traffic data and propose a novel machine learning-based short-term prediction method, referred to as spatiotemporal Gaussian Process Kalman filter (ST-GPKL) method, which includes two phases: the model selection and inference. The function of the model selection is to fine-tune the hyperparameters of the designed kernel function, while that of the inference incorporates the Kalman filter to predict the future mobile data traffic. Compared with the conventional methods, the proposed one can significantly improve the prediction accuracy, resulting in much higher efficiency in large-scale resource allocation.

Emerging recently as a novel concept in communication system design, end-to-end learning introduces deep neural networks (NNs) to represent the transmitter and receiver functions. Consequently, the whole system can be interpreted as an autoencoder (AE), which can be optimized from a holistic approach through a data-driven training method. Until now, the AE technique is mainly developed for point-to-point communication scenarios. In this paper, we aim to develop a novel NN-based AE scheme for relay-assisted cooperative communication systems. Specifically, three NN components are constructed to learn the behavior of the transmitter, relay node, and receiver, respectively. As the conventional end-to-end training is inapplicable, a novel two-stage training approach is proposed to indirectly solve the end-to-end training problem. The implicit approximations involved are analytically expressed based on information theory, offering insights on the achievable performance with the proposed training method. The proposed AE model eliminates the need for channel state information and noise variance of any link, and is adaptive to the variation in the input block length. Simulation results verify its advantages over the conventional decode-and-forward (DF) and amplify-and-forward (AF) schemes in various scenarios.

Opportunistic spectrum access (OSA) is envisioned to support the spectrum demand of future- generation wireless networks. In practice, primary channels are usually correlated and network dynamics is unknown a-priori. This entails a great challenge on sensing policy design, and conventional model-based methods are generally inapplicable. In this paper, we propose a novel Gaussian process reinforcement learning (GPRL) based model-free solution to enable the fast sensing policy optimization in OSA. In essence, Gaussian process is embedded in RL framework as a Q-function approximator to efficiently utilize the past learning experience. A novel kernel function is first tailor designed to measure spectrum data correlation. Then a covariance-based exploration strategy is developed to strike a better trade-off between the exploration and exploitation in RL. Our simulation results show that the proposed GPRL can obtain a near-optimal policy with significantly reduced learning period compared with deep reinforcement learning.

Multi-parameter cognition in a cognitive radio network provides a potential avenue to more efficient spectrum usage. In this paper, we propose a two-stage spectrum sharing strategy, where the primary user operates with multiple transmit power levels. Different from the conventional approaches, our method does not require any prior knowledge of the primary transmitter (PT) power characteristics. In the first stage, we use a conditionally conjugate Dirichlet process Gaussian mixture model to capture the multi-level power characteristics inherent in the PT signals, and design a Bayesian inference method to infer the model parameters. In the second stage, we propose a secondary transmitter (ST) prediction-transmission method based on reinforcement learning, which adapts to the PT power variation and strike an excellent tradeoff between the secondary network throughput and the interference to the primary network. The simulation results show the effectiveness of the proposed strategy.

In this paper, we propose a fast beam tracking strategy for mobile millimeter-wave systems, where the temporal variations of the angle of departure (AoD) are considered and modeled as a discrete Markov process. In contrast to most existing works that rely on the slow-fading assumption, we consider a more practical scenario in which the AoD can vary rapidly due to blockage and other environmental obstructions. In this case, the use of narrow training beams becomes inefficient, and therefore we propose to employ multiple radio-frequency chains generating wide beams to reduce the training time. By optimizing the selected training beams, we aim to minimize the average tracking error probability (ATEP). However, since the exact expression for ATEP is difficult to obtain, we derive its upper bound in a closed form, and aim to minimize this upper bound instead. The associated training beam sequence design problem is transformed into the construction of a bipartite graph that does not contain cycles of length 4, which is implemented with the progressive edge-growth algorithm. Numerical results demonstrate significant gains of the proposed beam tracking strategy over the existing benchmark methods.

In this paper, we introduce a wireless cooperative caching system (WCCS) to reduce the cost of both operators and users and also improve the quality of experience (QoE) of users. In this system, an intelligent routing relay (IRR) assigns a list of popular services and user terminals (UTs) can cache these services with their own cellular network traffic. When UTs upload these cached services to the IRR, they can obtain rewards. The other users can then access the IRR to get the cached services at a faster rate and get better QoE. The cellular network operators can save spectrum resource while they just deliver one copy of services. In order to encourage users to upload, we use reverse auction and first-come-first-served (FCFS) to choose the winning bids and allocate rewards to UTs. In this demo, we present displays on PC and mobile phones.

Content caching offers an effective solution to reduce the traffic load and alleviate the burden on backhaul links in future wireless networks. In this paper, we study the converged networks to push and cache the popular services. The popular services are delivered by the broadcasting networks, and cached in the router nodes in a distributed cache network. Due to the limited storage capacity of the router node, we formulate the service scheduling problem as a Markov Decision Process (MDP), aiming to maximize the equivalent throughput. Considering the large state space involved in the distributed cache network, it is great challenge to obtain a tractable solution by the classical optimization algorithm. To tackle this problem, we propose a service scheduling strategy based on deep Q-learning. Simulation results demonstrate that the proposed scheme can significantly improve the equivalent throughput of the converged networks.

Caching the most popular contents at the wireless network edge such as small-cell base stations (SBSs) is a smart way of reducing duplicated content transmissions and offloading the mobile data traffic in the network backhaul. Currently, most small-cell caching strategies are conceived, designed, and optimized based on the global content request probability (GCRP), with very limited consideration of the individual content request probability (ICRP) reflecting personal preferences. To enable more efficient wireless caching, in this paper, we propose a novel localized deterministic caching framework, drawing upon the recent advances in recommendation systems based on machine learning techniques. By introducing the concept of the rating matrix, we first propose a new Bayesian learning method to predict personal preferences and estimate the ICRP. This crucial information is then incorporated into our caching strategy for maximizing the system throughput, or equivalently, minimizing the download latency, where a deterministic caching algorithm based on reinforcement learning is proposed to optimize the content placement. To this end, we extend the framework to enable device-to-device (D2D) connections to further reduce the download delay, and also design a feedback mechanism to improve the accuracy in the ICRP estimation. Our simulation results verified that with the estimated ICRP and the proposed caching strategy, the proposed framework can significantly outperform the existing methods in terms of hit rate and system throughput.

In this paper, we propose a codebook-based beam tracking strategy for mobile millimeter-wave (mmWave) systems, where the temporal variation of the angle of departure (AoD) is considered. A closed-form upper bound of the average tracking error probability (ATEP) is derived and further optimized. We first consider a slow-varying scenario where narrow training beams implemented by single radio-frequency (RF) chain are employed. We show that the ATEP can be reduced by optimizing the power allocation strategy over these training beams, which is formulated and transformed into a second-order cone programming. The fast-varying scenario is further considered where the use of narrow training beams becomes inefficient due to the rapid variations of AoD. In order to reduce the training time, multiple RF chains generating wide beams are employed to track the AoD's variations, and the associated beam pattern design problem is shown to be a 0 - 1 nonlinear optimization problem (NLP). A sequential quadratic programming method is used to solve this binary NLP. To reduce the complexity, a progressive edge-growth algorithm is further introduced by associating the binary NLP with a bipartite graph. Numerical results demonstrate significant gains of the proposed beam tracking strategy over existing benchmarks for both scenarios.

Spectrum sensing in heterogeneous networks is very challenging as it usually requires a large number of static secondary users (SUs) to capture the global spectrum states. In this paper, we tackle the spectrum sensing in heterogeneous networks from a new perspective. We exploit the mobility of multiple SUs to simultaneously collect spatial-temporal spectrum sensing data. Then, we propose a new non-parametric Bayesian learning model, referred to as beta process hidden Markov model to capture the spatio-temporal correlation in the collected spectrum data. Finally, Bayesian inference is carried out to establish the global spectrum picture. Simulation results show that the proposed algorithm can achieve a significant spectrum sensing performance improvement in terms of receiver operating characteristic curve and detection accuracy compared with other existing spectrum sensing algorithm.

Generalized spatial modulation (GSM) activates 'it out of Nt (1 ≤ 'it <; Nt) available transmit antennas, and information is conveyed through 'it modulated symbols as well as the index of the 'it activated antennas. GSM strikes an attractive tradeoff between spectrum efficiency and energy efficiency. Linear precoding that exploits channel state information at the transmitter enhances the system error performance. For GSM with 'it = 1 (the traditional SM), the existing precoding methods suffer from high computational complexity. On the other hand, GSM precoding for 'it ≥ 2 is not thoroughly investigated in the open literature. In this paper, we develop a unified precoding design for GSM systems, which universally works for all 'it values. Based on the maximum minimum Euclidean distance criterion, we find that the precoding design can be formulated as a large-scale nonconvex quadratically constrained quadratic program problem. Then, we transform this challenging problem into a sequence of unconstrained subproblems by leveraging augmented Lagrangian and dual ascent techniques. These subproblems can be solved in an iterative manner efficiently. Numerical results show that the proposed method can substantially improve the system error performance relative to the GSM without precoding and features extremely fast convergence rate with a very low computational complexity. I'idex Terms-

In this paper, we address the challenging problem of vehicle license plate image super-resolution. Different from existing image super-resolution approaches only resorted to one single image, we propose to leverage complementary information from multiple images to recover the license plate numbers. To achieve this goal, we design a principled license plate images super-resolution framework which is composed of two components: progressive vehicle search and Domain Priori GAN (DP-GAN). Particularly, we design a null space based progressive vehicle search approach to retrieve the relevant images captured by different cameras given one vehicle with a low-resolution license plate. To handle the extremely varied license plate images caused by different sensors, times, depths, and viewpoints, we also propose a DP-GAN framework to generate multiple spatial correspondences and high-resolution plate images. In the generator network of DP-GAN, a license plate synthesis pipeline is exploited to generate the nearly canonical license plates. In the discriminator network, a spatial split layer is designed to simultaneously preserve the global and local manufacture standards of the license plate. Finally, a multiple images super-resolution GAN is exploited to combine all the synthetic license plates into one high-resolution image. Different from previous super-resolution criteria mainly focus on pixel-level detail recovery condition, we leverage the downstream tasks, i.e. license plate recognition and vehicle search as criteria. The results on a new collected real-world dataset demonstrate that the proposed method achieves the beyond human-level license plate super-resolution performance for automatic license plate recognition and vehicle search.

Wideband spectrum sensing for cognitive radio is highly challenging because it needs to locate multiple active spectrum subbands (channels) across a large bandwidth. The high-speed Nyquist sampling involved is either technically infeasible or very expensive in implementation. In this paper, we draw on the recent development in Bayesian machine learning, and propose a new high-resolution wideband spectrum sensing method, referred to as sub-Nyquist assisted matrix sparse Bayesian learning (M-SBL). We first use multicoset sampling to significantly reduce the sampling rate. Then we develop a M-SBL method that carries out Bayesian inference from received spectrum data to learn and iteratively reconstruct a latent variable, whose significant peaks can be used to locate multiple active spectrum subbands. Simulation results indicate that the proposed method significantly outperforms conventional ones in sensing accuracy, especially at low signal-to-noise ratios or with a small number of cosets.

In this paper, we investigate linear precoding for spatial modulation (SM) over multiple-input-multiple-output (MIMO) fading channels. With channel state information avail- able at the transmitter, our focus is to maximize the minimum Eu- clidean distance among all candidates of SM symbols. We prove that the precoder design is a large-scale non-convex quadratically constrained quadratic program (QCQP) problem. However, the conventional methods, such as semi- definite relaxation and it- erative concave-convex process, cannot tackle this challenging problem effectively or efficiently. To address this issue, we leverage augmented Lagrangian and dual ascent techniques, and transform the original large-scale non-convex QCQP problem into a sequence of subproblems. These subproblems can be solved in an iterative manner efficiently. Numerical results show that the proposed method can significantly improve the system error performance relative to the SM without precoding, and features extremely fast convergence rate with very low computational complexity.

Carrier-frequency offset (CFO) estimation for uplink orthogonal frequency-division multiplexing access (OFDMA) systems is very challenging as it requires the estimation of multiple CFOs. In this paper, we propose a new framework referred to as sparse blind CFO estimation for interleaved uplink OFDMA. The proposed framework first discretizes the potential frequency offset ranges into discrete grid points, and formulates the original CFO estimation into a sparse signal recovery problem. Then, a novel two-stage matrix Bayesian compressive sensing-based CFO estimation method is proposed to solve the formulated problem. In the first stage, we employ a relatively large grid interval, and iteratively reconstruct a hyperparameter vector to generate a coarse estimation of multiple CFOs. The second stage reduces the grid interval, and the refined CFOs are estimated one by one through a novel low-complexity one-dimension searching algorithm. Numerical results show that the proposed method significantly outperforms conventional ones in terms of estimation accuracy, especially in the scenarios, such as low signal-to-noise ratios, large CFOs, and a large number of users.

Interference from unknown devices makes time-of-arrival (ToA) estimation using conventional signal processing methods unreliable. In this paper, we propose new ToA estimation techniques based on Bayesian compressive sensing (BCS) to improve the accuracy of the ToA estimation under the interference scenario. Our proposed BCS based ToA estimation schemes maximize the posterior probability of the channel impulse response (CIR) with given frequency domain received signals. Simulation results show that proposed BCS based ToA estimations exhibit significantly improved ToA detection accuracy and mean-squared error performance in interference scenarios. We also demonstrate a practical example of the ToA estimation using real measured indoor channels.

Distributed data storage (DDS) provides a promising approach to the reliable recovery of the whole sensor readings in a wireless sensor network (WSN) by visiting a small subset of sensor nodes. Various DDS schemes based on compressive sensing (CS) have been proposed to reduce the number of transmission/receptions to improve network's area energy efficiency. However, these schemes assume that sensor readings are compressible in the discrete cosine transformation (DCT) domain, whereas our experimental results validate that this assumption cannot be established in a real WSN scenario and the performance of the practical sensor readings recovery will be significantly degraded. To address this problem, this paper proposes a novel DDS scheme termed as practical spatiotemporal compressive network coding (P-STCNC). Our idea is to adaptively train the sparse dictionaries to sparsify practical sensor readings as well as optimize corresponding measurement matrices in both spatial and temporal domains to guarantee accurate data recovery. Simulation results based on real datasets demonstrate the effectiveness of the proposed scheme.

The availability of channel state information (CSI) at the transmitter is crucial to multiple-input multiple-output orthogonal frequency-division multiplexing (MIMO-OFDM) systems to suppress interference with precoding techniques. In a MIMO-OFDM frequency division duplex (FDD) system, the amount of CSI required by the transmitter increases linear with the number of subcarriers, and therefore becomes prohibitive for a large bandwidth configuration. Conventional schemes based on broadband analog CSI feedback can reduce the overhead, but has limitations to achieve higher spectral efficiency and feedback robustness. In this paper we first propose a novel broadband analog CSI feedback framework for MIMO-OFDM systems based on the distributed compressive sensing (DCS) theory. Then, we further optimize this framework to maximize its feedback benefits. By exploiting the sparse common support (joint sparsity) inherent in MIMO channels and taking advantage of the high efficiency of DCS in encoding original MIMO channels, the proposed framework is capable of significantly reducing the overhead and enhancing the robustness. The numerical results validate the effectiveness of the proposed scheme.

We study the analog feedback of channel state information (CSI) in the massive multiple-input multiple-output orthogonal frequency-division multiplexing (MIMO-OFDM) transmission. In a massive MIMO-OFDM system, the amount of CSI required by the transmitter increases linearly with the number of transmit antennas and subcarriers. To address this challenge, in this paper we propose a novel CSI feedback method for massive MIMO-OFDM systems based on the multidimensional compressive sensing theory. Taking advantage of a mathematical advance, the Tucker tensor decomposition, and spatial and frequency correlation of the channel, we reveal the connection between the tensor decomposition model and CS involving multidimensional signals (tensors). Then by making use of this connection, we exploit the structure contained in all different dimensions of the original channel matrix and compress it in each dimension simultaneously, thereby resulting in a large reduction in amount of CSI to be fed back and enabling a significant enhancement of spectral efficiency. The numerical results validate the effectiveness of the proposed scheme.

Channel estimation for an orthogonal frequency-division multiplexing (OFDM) broadband system operating with high mobility is very challenging. This is mainly due to the significant Doppler spread, inherent in a time-frequency doubly-selective (DS) channel. Consequently, a large number of channel coefficients must be estimated, forcing the need for allocating a large number of pilot subcarriers. To address this problem, we propose a novel channel estimation method based on basis expansion models (BEMs) and distributed compressive sensing (DCS) theory. To be specific, we develop a two-stage sparse BEM coefficients estimation method, which can effectively combat the Doppler spread and enable accurate channel estimation with dramatically reduced number of pilot subcarriers. The numerical results reveal that, in a typical LTE system configuration, the proposed scheme can increase the spectral efficiency by 40% and achieve a 6 dB gain in terms of normalized mean square error (NMSE), both compared to the conventional scheme.

In this paper, we tackle the energy efficiency (EE) issue in uplink virtual MIMO systems, which requires the optimization of two interlaced parameters: the number of constituent mobile users in the virtual MIMO and their corresponding power allocation. By exploiting the fact that increasing the number of active users can increase the number of contributors to the total EE on one hand but reducing the diversity order for each single user on the other, we can show the existence of an optimal transmission mode and find a simple way for its search. Through in-depth analysis, we show the existence of a unique globally optimal power allocator for the case without power constraints, and further reveal the impact of power constraints upon power allocation, as compared to its global counterpart, aiming to provide a powerful means for power-constrained EE optimization. Finally, we establish theories, for homogeneous networks, to narrow down the search range for possible transmission modes, leading to a significant reduction of computational complexity in optimization. Simulation results are presented to substantiate the proposed schemes and the corresponding theories.

Carrier-frequency offset (CFO) estimation for orthogonal frequency-division multiplexing access (OFDMA) systems operating in multiuser uplink transmission is very challenging due to the presence of a multiple-parameter estimation problem. In this paper, we propose a novel blind CFO estimation method for interleaved OFDMA uplink based on distributed Bayesian compressive sensing (DBCS) theory. Considering the received signal structure, the new method first constructs a measurement matrix associated with a sparse signal matrix weight, which sets up the stage for the application of CS theory in tackling the original estimation problem. Then, the DBCS theory that exploits a common sparse profile of the sparse signal matrix weight is employed to distributively estimate a sparse hyperparameter vector, whose significant peaks are linked to the correct estimation of the multiple CFOs. Compared with the existing subspace theory based methods, the proposed scheme offers a significant enhancement in estimation accuracy, in specific in the low signal-to-noise ratio (SNR) region. The numerical results validate the effectiveness of the proposed scheme.

Limited feedback precoding (LFP) significantly improves multiple-input multiple-output (MIMO) spatial multiplexing link reliability with a small amount of feedback from the receiver back to the transmitter. One of the key problems linked to LFP is how to select an optimal precoder from a pre- determined unitary codebook. We find that the conventional precoder selection criteria are not applicable to the stream maximization transmission (SMT) mode with linear receivers, including zero forcing (ZF) and minimum mean square error (MMSE) decoders. To solve this issue, a novel singular value decomposition (SVD) based precoder selection criterion is proposed in this paper. This criterion features a unified structure for all the linear receivers such as ZF and MMSE decoders, and is shown by simulation to provide significant coding gains in various SMT systems. With the same complexity as the conventional one, the proposed criterion could find its applications in next generation systems employing SMT spatial multiplexing, significantly improving system performance with affordable feedback requirement.

Driven by green communications, energy-efficient transmission is becoming an important design criterion for wireless systems, aiming to extend the life cycle of batteries in mobile devices. In this paper, we tackle the energy efficiency (EE) issue in uplink virtual multiple-input multiple-output (MIMO) systems, which requires the optimization of two interlaced parameters: the number of constituent mobile users in the virtual MIMO and their corresponding power allocation. The former parameter is a structural parameter defining the size of the virtual MIMO (usually known as the transmission mode) and its optimization relies on the method of enumeration. The difficulty is further aggravated by the fact that the EE is a non-convex function of power, even for a given transmission mode. By exploiting the fact that increasing the number of active users can increase the number of contributors to the total EE on one hand but reducing the diversity order for each single user on the other, we can show the existence of an optimal transmission mode and find a simple way for its search. Through in-depth analysis, we show the existence of a unique globally optimal power allocator for the case without power constraints under the assumption of zero-forcing receivers, and further reveal the impact of power constraints upon power allocation, as compared to its global counterpart, aiming to provide a powerful means for power-constrained EE optimization. Finally, we establish theories, for isometric networks, to narrow down the search range for possible transmission modes, leading to a significant reduction of computational complexity in optimization. Simulation results are presented to substantiate the proposed schemes and the corresponding theories.

Channel estimation for an orthogonal frequency-division multiplexing (OFDM) broadband system over a doubly selective channel is very challenging. This is mainly due to the significant Doppler shift, which results in a time-frequency doubly-selective (DS) channel. The DS channel features a large number of channel coefficients, which introduces inter-carrier interference (ICI) and forces the need for allocating a large number of pilot subcarriers. To tackle this problem, in this paper we propose a novel channel estimation scheme based on distributed compressive sensing (DCS) theory. Taking advantage of the basis expansion model (BEM) and the channel sparsity in the delay domain, we transform the original DS channel into a novel two-dimensional channel model, where several jointly sparse BEM coefficient vectors become the estimation goal. Then a special decoupling form originating from a novel sparse pilot pattern is designed for such estimation, which results in an ICI-free structure and enables the DCS application to make joint estimation of these vectors accurately. Combined with a smoothing treatment process, the proposed scheme can achieve significantly higher estimation accuracy than the existing ones, although with a much smaller number of pilot subcarriers. Theoretical analysis and simulation results both confirm its performance merits.

Compressed sensing (CS) has recently emerged as a powerful signal acquisition paradigm. CS enables the recovery of high-dimensional sparse signals from much fewer samples than usually required. Further, quite a few recent channel measurement experiments show that many wireless channels also tend to exhibit sparsity. In this case, CS theory can be applicable to sparse channel estimation and its effectiveness has been validated in point-to-point (P2P) communication. In this work, we study sparse channel estimation for two-way relay networks (TWRN). Unlike P2P systems, applying CS theory to sparse channel estimation in TWRN is much more challenging. One issue is that the equivalent channels (terminal-relay-terminal) may be no longer sparse due to the linear convolutional operation. On this basis, novel schemes are proposed to solve this problem and effectively improve the accuracy of TWRN channel estimation when using CS theory. Extensive numerical results are provided to corroborate the proposed studies.

Energy-efficient transmission is increasing in importance for wireless system design because of limited battery power in mobile devices. In this paper, we consider energy efficiency optimization in uplink virtual MIMO systems. First, accurate closed-form expressions are derived for the achievable energy efficiency for both multi-user (MU) transmission mode and single-user (SU) transmission mode, which are based on the ergodic throughput, transmit power and circuit power. Then, we demonstrate the existence of unique globally optimal energy efficiency for SU and MU mode, respectively. Since users have data rates requirement and peak power limits, we further consider power constraint, and develop joint mode switching with power loading schemes to optimize energy efficiency for both homogeneous and heterogeneous networks. Finally, our simulation results show that the proposed adaptive transmission strategies significantly improve energy efficiency.

We consider how to efficiently employ D2D communications for secondary users (SUs) in a cognitive cellular network. In this network, primary users (PUs) transmit via base station normally, while SUs can employ multiple transmission modes. One is to transmit via base station (BS mode), and the other is to employ D2D communication (D2D mode) due to the scarce idle spectrum. The SUs who have the potential to transmit to each other using D2D mode form a group. Within this group, they can transmit to each other via BS mode or using D2D mode directly. Outside this group, only BS mode is available. To investigate how to employ D2D mode into this network, first we define the utilities of SUs employing BS mode and D2D mode respectively considering achieved data rate, power consumption, price of unit bandwidth and the impact of interference. Then we analyze the optimal power allocation for each mode. To optimize SUs' strategies of mode selection, we adopt replicator dynamics in evolution theory to model the behaviors of SUs. Furthermore, we prove the existence of SUs' evolutionary stable strategy (ESS) of the mode selection process. Based on our model, we finally propose a distributed protocol for SUs within a D2D group to converge to ESS automatically. Numerical results show that our proposed protocol is not only efficient to achieve ESS with improved network performance, but also robust in ESS.

This paper presents an evolution framework of resource allocation by infection among secondary users (SUs) in an OFDMA-based cognitive radio cellular networks. Each primary user (PU) sells his extra sub-channels to SUs in his sensing range to achieve the highest payoff and each SU may come across another in his sensing range to make infection. Two different infection processes among SUs, the infection with and without local knowledge respectively, are considered. We prove the existence and convergence of evolutionary equilibrium (EE) for both cases, and show some interesting properties such as the impact of cheating of SUs in the above infection processes. Besides, we proposed two algorithms for the infection processes to converge to EE in a distributed manner. Simulation results show that the algorithms with local knowledge can equally share the extra resources among SUs efficiently, which is actually the overall optimal solution (OOS). While for the algorithm without local knowledge, we find that though EE exists, OOS cannot always be achieved. Furthermore, we optimize the second algorithm to make EE approximate to OOS.

Cooperative sharing of spectrum among one primary user (PU) and multiple secondary users (SUs) helps to enhance the throughput of the whole system. In this paper, we propose a two-tier game in the sharing of spectrum in which SUs make their decisions on whether to cooperate under replicator dynamics and the PU adjusts its strategy to allocate time slots for the cooperative SUs' transmission. Furthermore, we develop the distributed algorithm to describe the learning process of SUs, and prove that the dynamics can converge to the evolutionary stable strategy (ESS) efficiently, which is actually the optimal strategies of both PU and SUs. Simulation results show that our proposed mechanism converges to ESS automatically, at which point all the SUs will remain their strategies. In addition, it is shown that this mechanism helps the SU to share information and gain higher transmission rate than fully cooperative or non-cooperatively scenario.

As a bridge of connecting orthogonal frequency division multiplexing (OFDM) with single-carrier frequency domain equalization (SC-FDE) techniques, Vector OFDM (V-OFDM) provides significant flexibility in system design. This paper presents an analytical study of V-OFDM over multi-path fading channels. Our goal is to investigate the diversity gain and coding gain of each vector block (VB) in V-OFDM so as to ultimately reveal its performance limits over fading channel. By using algebraic number theory tools, we rigorously prove for the first time that a majority of VBs in V-OFDM can surely realize the diversity gain of min {M,G} , where M is the length of each VB, and G is the total number of channel taps. Furthermore, some specific VBs, whose length equals the total number of channel taps, can not only harvest the maximum diversity gain but also achieve the maximum coding gain. It is further demonstrated that, even though VBs fail to benefit from additional diversity gain when M exceeds G, they can enjoy significantly increased coding gains. Our analysis concludes that it is preferable to choose the length of VBs to be equal to the number of channel taps in consideration of both overall system performance and computational complexity.

In this paper, we propose an agent-based spectrum trading model, where agent plays a third-party role in the trading process. Providing service to secondary users (SUs) with spectrum bought from primary users (PUs), agent makes profit in the spectrum trading process. We address the challenge of finding the most profitable strategy of agent(s) when spectrum demand is uncertain. We first address this challenge for the secondary network where single agent operates, and extend it to a multiple-agent system. To our best knowledge, this is the first solution to agent-based spectrum trading considering demand uncertainty.

In this letter, a low-complexity structure of PAPR reduction scheme, user grouping, in multicarrier code-division multiple access (MC-CDMA) is introduced. The proposed structure is based on joint spreading and inverse fast Fourier transform (S-IFFT) processing, and a class of low-complexity structure of user grouping scheme is proposed on the tradeoff between computational complexity and storage size. Comparison with the conventional PTS schemes, it shows that the proposed low-complexity scheme can achieve similar PAPR reduction with lower complexity.

Partial transmit sequence (PTS) is an attractive distortless peak-to-average power ratio (PAPR) reduction technique for orthogonal frequency division multiplexing (OFDM) system. However, the complexity of PTS increases exponentially with the number of sub-blocks as it requires an exhaustive searching over all phase factor combinations. In this paper, we explored the correlation among the multiple candidate signals of PTS and proposed a modified scheme with lower complexity. The main idea is based on selecting a subset from the whole set of candidate signals by analyzing the correlation of candidate signals. Simulation results show that the performance of the modified scheme is close to the theoretical lower bound and outperforms selected mapping (SLM) scheme with the same computational complexity.

Selected mapping (SLM) is a promising peak-to-average power ratio (PAPR) reduction technique for orthogonal frequency division multiplexing (OFDM) system. In SLM, phase sequences are combined with the data for generating alternative signals, so as to reduce the PAPR. In this paper, a new criterion is developed to examine the effects of different phase sequence sets in SLM-OFDM based on the mathematical correlation analysis among the alternative signals. Furthermore, according to the proposed criteria, the chaotic phase sequence set is introduced for improving the PAPR reduction performance in SLM. Simulation results show that the improved SLM outperforms conventional methods such as proposed in [6].