Large-scale Internet of Things (IoT) connections in dynamic low Earth orbit (LEO) satellite networks face significant challenges in uplink resource scheduling. This paper proposes a framework for optimizing spectral efficiency. The framework satisfies heterogeneous quality of service (QoS) requirements and dynamic buffer constraints under time-varying IoT traffic bursts. It integrates three critical aspects. First, it considers the spatial geometric relationship between satellites and ground user equipment (UE), which determines the connection duration. Second, it achieves service-specific QoS priorities through an adaptive weighting mechanism. Third, it addresses time-varying traffic patterns. Under time-varying resource constraints, the high-dimensional scheduling optimization problem is modeled as a multi-dimensional multi-choice knapsack problem (MMKP). A satellite selection scheme is proposed to efficiently solve the MMKP with mixed constraints. This scheme simplifies the three-dimensional knapsack problem (KP) into a two-dimensional one by taking connection duration into account. This reduction explicitly accounts for the space and time limitations of satellite-ground links. It also integrates service-specific priorities. Meanwhile, the scheme enables each satellite to handle its own computations and resource allocation independently. A binary split dynamic programming (BSDP) algorithm is developed to solve the two-dimensional KP. To compare performance, two large-scale integer optimization methods—the Lagrangian Relaxation Algorithm (LRA) and Branch and Bound (B&B)—were used to solve the KP. The results were compared with a perception-based greedy resource block (RB) allocation for the original resource allocation problem. Extensive simulations based on Starlink demonstrate the effectiveness of the proposed solution. When serving over 4000 UEs, the MMKP solution achieves a 46% gain in QoS compared to the greedy benchmark. It also achieves a 60.7% throughput gain. Additionally, BSDP performs almost as well as B&B. BSDP has approximately two orders of magnitude lower computational cost than LRA.

Satellite-terrestrial integrated network (STIN) plays a crucial role in achieving 3-dimensional full-area coverage. STIN enables the Internet of Things (IoT) industry to realize the integrated space-air-ground communication. The stability of satellite-terrestrial communication and the quality of service (QoS) in low earth orbit optical satellite networks (LEO-OSNs) need to be improved, especially for satellite-based IoT (SIoT) services. To address these challenges, we propose the heuristics multi-physical channel switching and dual Hamming distance-based routing and wavelength assignment (RWA) scheme (HMPS-DHR). Based on dual HAPs deployment architecture and link conditions-aware signal-to-noise ratio (SNR) thresholds model, the multi-physical channels can be flexibly switched among free-space optical (FSO) laser links, Ka-band and S-band microwave links to ensure the stability of the satellite-terrestrial feedback links (FLs). Meanwhile, the traffic conflict gain-adaptive and load-aware dual Hamming distance RWA (TCG-LDHR) algorithm is proposed to optimize the routing, address the RWA problem, and enhance QoS. Simulation results demonstrate that the proposed HMPS-DHR effectively guarantees the communication success rates between satellite and ground at approximately 98.9% to 99.2%, and improves the QoS metrics involving total delay, average throughput, packet loss rate, and blocking rate, by 15.6% to 56.4% compared with the Dijkstra-FF and the ant colony optimization with adaptive load balance small window strategy under hop number loose constraint (ACO-ALB-SWS-HNLC), respectively. HMPS-DHR shows acceptable robustness to synchronization deviations despite unavoidable millisecond-level timing mismatches. Although the QoS performance of the proposed HMPS-DHR is slightly lower than that of the integrated multipath network coding (IMPNC) scheme, its computational complexity is significantly reduced.

In optical transport networks (OTNs), proactive and accurate prediction of key performance parameters plays a crucial role in identifying potential failure of OTN equipment and guiding timely operational interventions, reducing downtime and improving overall system performance. However, the performance parameters in OTNs are complex and diverse. The reliance of existing models structure design on specific configurations limits generalizability across diverse equipment types. Moreover, the high computational resource consumption and memory footprints of these models may lead to inefficiency while hindering practical application and large-scale deployment. To address these challenges, this paper presents a general model, KD-TimeLLM, a cross-application of TimeLLM into OTN failure management, for performance parameter prediction of multiple equipment types in OTNs. By learning from its teacher model TimeLLM via a knowledge distillation strategy, KD-TimeLLM can achieve generalizability in performance parameter prediction while enhancing efficiency. We conducted evaluations across multiple metrics using data sets from different operators and various board types. Results show that KD-TimeLLM outperforms other models in predictive effects including the lowest MSE and MAE across all types of board data along with a scaled_RMSE value below 0.5, the varying number of performance parameters, and zero-shot prediction capability, highlighting its generalizability. Moreover, compared to its teacher model, KD-TimeLLM achieves comparable predictive effects with a significant reduction 99.99% in model parameters and an average reduction of 99.23% in inference time across eight different types of board data. Furthermore, compared to a multiple-model system, total inference time and memory footprint of KD-TimeLLM decreased by 94.79% and 89.65%, highlighting its effectiveness and efficiency.

Low Earth orbit (LEO) satellite networks, as a fundamental component of 6G networks, are designed to provide full coverage, low latency, and high quality of service (QoS) for satellite-terrestrial integrated networks (STIN). Topology representations and routing computation in dynamic LEO satellite networks have become key research focuses. However, balancing network dynamics with traffic load remains challenging due to inaccurate topology representation and inefficient routing in existing studies. To address this, we propose a timeslot-adaptive and traffic load-aware routing computation (TA-TLARC) scheme for two-layer LEO satellite networks. The two-layer LEO satellite networks consist of communication layer satellites (CLS) and relay and sensing layer satellites (RSLS). TA-TLARC adaptively adjusts timeslots based on traffic variations and utilizes distributed adjacency matrices for routing computation. Simulation results show that TA-TLARC achieves better performance than existing routing schemes in key QoS metrics such as routing success rate, delay, throughput, and packet loss rate. Although routing hops and power consumption increase within acceptable limits, the routing success rate of TA-TLARC remains 99.6% to 100%. The QoS performance, including delay, throughput, and packet loss rate, is improved by 10% to 40% compared to those of the comparative schemes under different traffic scenarios. The robustness of TA-TLARC is further analyzed and demonstrated to be acceptable under various failure conditions. The results demonstrate that the proposed TA-TLARC effectively addresses routing computation challenges and significantly improves QoS performance in two-layer LEO satellite networks.

The digitization and intellectualization have been envisioned as the fundamental basis for future Industrial Internet of things, which integrates sensor technology, industrial control technology, communication technology, and artificial intelligence (AI). Specifically, the collaboration among these above techniques is crucial for the successful implementation of intelligent applications. This article develops an end-to-end visual control framework to accomplish multi-crane collaborative sorting in wireless time sensitive networking (TSN) networks. The design primarily incorporates field devices, data transmission, artificial intelligence (AI), and industrial control. An advanced binocular stereo visual recognition model based on deep learning is investigated to accurately obtain the world coordinates and types. A cooperative control scheduling model that combines a scheduling strategy with an anti-collision strategy is presented to effectively control multiple cranes for sorting tasks. The device data and commands are transmitted through industrial 5G-TSN integrated network for ultra-reliable, low-latency, and deterministic transmission. The proposed visual sorting system is further validated through the establishment of an experimental prototype, demonstrating its exceptional real-time performance while enabling flexible intelligent manufacturing.

In modern optical networks, alarm analysis plays a pivotal role in detecting fault and guiding operators towards timely interventions. Timely and accurate root cause localization can effectively reduce downtime, prevent cascading failures, and enhance overall system performance. However, the increasing scale and complexity of networks have posed challenges to Root Cause Localization (RCL) due to the huge volume of alarms generated during fault occurrences. The development of automatic RCL methods helps reduce the time costs and human errors associated with manual analysis. Most existing RCL methods face challenges such as poor explainability, low adaptability to network changes, high learning costs and lack of interactivity, reducing their credibility and usability in production environments. This paper introduces a graph structure-enhanced large language model (LLM) for optical network fault detection, capable of performing explainable alarm RCL with improved adaptability and interactivity. Graph structures provide an intuitive means of expressing the relationships between topology and alarms, clearly visualizing alarm propagation paths, which aids in explainable RCL; LLM exhibits profound capabilities in semantic comprehension and language generation, which improves both explainability and interactivity. When integrated with algorithms, they hold promise for reducing operators learning costs, adding interactivity, and introducing new possibilities for the visualization and efficiency of complex tasks. We conducted evaluations and validations across multiple metrics using real alarm data collected from optical transport network (OTN). The results show that the proposed algorithm achieves an accuracy of over 86.8% across various complex scenarios; compared to the base model, the fine-tuned model exhibits an accuracy improvement of over 87%. Other results indicate that the proposed method enhances the explainability, adaptability and interactivity of fault detection in optical network, showcasing significant potential for automated and intelligent network operations.

Digital twin (DT) techniques have been proposed for the autonomous operation and lifecycle management of next-generation optical networks. To fully utilize potential capacity and accommodate dynamic services, the DT must dynamically update in sync with deployed optical networks throughout their lifecycle, ensuring low-margin operation. This paper proposes a dynamic-updating DT for the lifecycle management of optical networks, employing a hybrid approach that integrates data-driven and physics-informed techniques for fiber channel modeling. This integration ensures both rapid calculation speed and high physics consistency in optical performance prediction while enabling the dynamic updating of critical physical parameters for DT. The lifecycle management of optical networks, covering accurate performance prediction at the network deployment and dynamic updating during network operation, is demonstrated through simulation in a large-scale network. Up to 100 times speedup in prediction is observed compared to classical numerical methods. In addition, the fiber Raman gain strength, amplifier frequency-dependent gain profile, and connector loss between fiber and amplifier on C and L bands can be simultaneously updated. Moreover, the dynamic-updating DT is verified on a field-trial C+L-band transmission link, achieving a maximum accuracy improvement of 1.4 dB for performance estimation post-device replacement. Overall, the dynamic-updating DT holds promise for driving the next-generation optical networks towards lifecycle autonomous management.

To further improve the transmission efficiency and link stability for free space optical (FSO)-based satellite communication (SatCom) systems when transmitting large-scale remote sensing images, a scheme based on the integration of FSO and semantic communication (FSO-SC) is proposed, which employs a vector quantized variational autoencoder with spatial normalization to extract essential semantic features of images while preserving intricate details. Additionally, theMáalagadistribution model is utilized to simulate FSO channels with diverse turbulence conditions. Moreover, a comparative evaluation between the FSO-SC and traditional systems is conducted through 28 GBaud satellite-ground simulation with three modulation formats considering various effects. Compared to the traditional systems, without incurring additional bits for error corrections, the FSO-SC system achieves a power gain of over 3 dB while enabling transmission at zenith angles over 60°. Moreover, it achieves performance on par with state-of-the-art 4-receiver spatial diversity technology, while offering superior hardware and transmission efficiency. Furthermore, we conduct 10 Gbps real-time satellite-ground equivalent experiments to validate the practicality of the FSO-SC, where it achieves a 60% reduction in communication overhead compared to existing solutions while maintaining comparable received image quality and can reach a minimum receiver sensitivity gain of 4 dB. Simulation and experimental results demonstrate that the proposed FSO-SC scheme achieves high system efficiency and stability, holding promise as a viable solution for future SatCom.

In future 6G, bounded delay, ultra-high-reliability, and dedicated services will require high-performance network modeling techniques for the accuracy pre-validation such as network delay prediction. Recently, graph neural networks (GNNs) have been shown great potential for network delay prediction. GNNs are able to effectively capture complex topologies and node features in graph data by recursively aggregating the neighborhood information of nodes. To improve the ability to learn representations of graph data, GNNs are suitable for a variety of complex network modeling tasks with high flexibility and powerful scalability. However, the current GNN-based Routenet model can not capture the effect of the path on neighboring links, which is ineffective in complex topologies. To model the effects of the network path on neighboring links, this paper proposes a transformer-based GNN model named Transnet. In this model, the Transformer is first introduced to update the path and link states and describe the effects of the path on neighboring links based on the attention mechanism in the Transformer. Simulation results show that the delay prediction accuracy of the proposed Transnet obviously exceeds those of Routenet on multi-node topology in the Nsfnet and Synth50 datasets.

Network slicing (NS) is a key technology to cost-effectively meet diverse service level agreement (SLA) demands of the Internet of Everything communication. Thanks to high-fidelity network modeling capabilities and flexible feedback optimization techniques, digital twins (DTs) and reinforcement learning (RL) have been applied to dynamic NS management. However, most existing DTs lack the ability of predictive uncertainty evaluations, and tend to be overconfident on the unknown network environment. For classical RL, it is exceedingly intractable to maintain high-stable NS performances in dynamic networks. To address those problems, we propose a DT-based low-risk NS (DT-LNS) framework and method using the safe RL. In the safe RL, a DT using deep neural networks with the data-model uncertainty analysis is adopted to predict NS performances and provide predictive uncertainties. Further, the RL is used to select low-risk NS configuration actions by preverifying the SLA violation risk of candidate actions from the RL and the reference action subspace via DTs. The proposed DT-LNS method can keep the high-SLA satisfaction rate (SSR), reduce the performance jitters, and improve the convergence speed. Compared with the six classic NS configuration methods, including round robin, deep Q network, advantage actor-critic, deep deterministic policy gradient, and advanced RL, assisted with the DT-based model pretraining and the state prediction, the average percentage gain of the proposed method is 7.84%, 93.58%, 65.63%, 84.20%, and 90.27%, regarding the performances of the average SSR, SSR jitter, delay jitter, data rate jitter, and the convergence speed, respectively.

We develop a deep-learning-based multicrane visual sorting system with virtualized programmable logic controllers (PLCs) in intelligent manufacturing, which enables the accurate location and suction of the materials on the conveyor belt. First, virtualized PLCs are deployed in the field and the cloud to break data islands for efficient communication between low-level devices. Second, artificial intelligence algorithms are integrated into the physical industrial control system in which cooperation between virtualized PLCs and the visual recognition model is developed to complete the industrial control closed loop. Third, we establish a visual recognition model in which object detection algorithms are used to process the original image and then obtain the position and type of the object in the pixel coordinate system. In addition, a new linear interpolation-based backpropagation neural network is presented to provide the transform relation between the pixel coordinate system and the world coordinate system that the crane needs to precisely suck the material. The whole system is applied in a time-sensitive network environment in a highly reliable and stable manner. The experimental prototype system demonstrates that high recognition accuracy can be achieved for the visual sorting system within an acceptable time frame. The accuracy of the sorting task reaches 96.5% and the average consumption time of each object is approximately 2.317 s when the speed of the conveyor belt is 5.2 m/min.

Abstract Next‐generation 6G networks will fully drive the development of the industrial Internet of Things. Steel surface defect detection as an important application in industrial Internet of Things has recently received increasing attention from the military industry, the aviation industry and other fields, which is closely related to the quality of industrial production products. However, many typical convolutional neural networks‐based methods are insensitive to the problem of unclear boundaries. In this article, the authors develop a region‐based fully convolutional networks with deformable convolution and attention fusion to adaptively learn salient features for steel surface defect detection. Specifically, deformable convolution is applied into selectively replace the standard convolution in the backbone of the region‐based fully convolutional networks, which performs significantly in scenarios with unclear defect boundaries. Moreover, convolutional block attention module is utilised in region proposal network to further enhance detection accuracy. The proposed architecture is demonstrated on two popular steel defect detection benchmarks, including NEU‐DET and GC10‐DET, which can effectively present the performance of steel surface defect detection by abundant experiments. The mean average precision on two datasets reaches 80.9% and 66.2%. The average precision of defect crazing, inclusion, patches, pitted‐surface, rolled‐in scale and scratches on NEU‐DET is 58.2%, 82.3%, 95.7%, 85.6%, 75.9%, and 87.9% respectively.

Person reidentification (re-ID), which is a significant and potential application in the Internet of Things (IoT), aims to retrieve pedestrians of interest given a labeled image in a camera network. Now, it is still existing many challenges that severely influence feature representation in practical scenarios. Many methods adopt the attention mechanism in convolutional neural network (CNN) to improve the ability of feature learning. Although they only apply 1-D attention block in the popular deep learning architecture, the learned features are not discriminative for the feature representation. In this work, we investigate a self-focusing network (SFNet) that considers both the channel-dimensional attention and spatial-dimensional attention to adaptively learn more discriminative features. Namely, we embed the new attention module into the common backbone network, which can focus on the salient region by inhibiting the redundant features. Specifically, we design eight variants of the channel-dimensional attention and spatial-dimensional attention throughout the entire network and explore the most powerful feature representation. The heatmaps of different layers are visualized to intuitively present the performance of SFNet. Furthermore, we compare SFNet with the prior work on three popular person re-ID benchmarks by abundant experiments.

Telecom big data generated by telecom networks have a high economic value. Thus, telecom operators actively explore telecom big data transactions methods to minimize the possibility of leaking users’ privacy. The existing solutions do not allow the data sets to leave the database, instead only allow the buyers to send data mining algorithms to the telecom operator’s platform for training. However, this centralized platform has a high risk of tampering. In addition, the currently existing solutions cannot be used to accurately and quickly trace the information of telecom big data transactions. To address these limitations, we propose a blockchain-based reliable traceability system for telecom big data transactions using smart contracts and the InterPlanetary File System. Two types of smart contracts are developed to store transaction information for tracing. Access control strategies and a reapproval prevention strategy are designed for ensuring the safe operation of the system and avoiding the problem of favoritism and fraud. We use Ethereum as a verification platform to develop and evaluate this system. The implementation of functions, such as purchasing data sets, sending algorithms, obtaining results, and tracing transactions in the smart contract and the implementation of the proposed strategies are verified. The results demonstrate that the performance of the proposed system is better than the existing solutions, and the traceability response time is improved to the order of seconds, so as to realize the safe and efficient traceability of telecom big data transactions. In addition, Ethereum and Hyperledger Fabric v0.6 were discussed to provide insights for future development.

With the continuous addition of an abundant of heterogeneous devices, the limitation of task delay has become an obstacle to the development of the Industrial Internet of Things (IIoT). Task offloading based on edge computing can provide low-latency computing services for these tasks. However, in the actual IIoT scenario, in contrast to cloud computing, edge computing has limited resources and computing capabilities. Resource-constrained edge resources cannot meet the offloading requirements of massive industrial devices. In this article, we propose an optimal joint offloading scheme based on resource occupancy prediction for the problem of computing offloading with limited edge resources. The scheme is divided into two parts, including edge resource occupancy prediction and task offloading. Simultaneously, considering multitask and the limitations of edge resources, gate recurrent unit (GRU) is used to predict the occupancy of edge resources. Formulating an optimal strategy of task offloading by using a reinforcement learning algorithm according to the network state and predicted results. The simulation results show that the scheme can effectively reduce the average delay of tasks, while minimizing the task offloading failure rate.

In recent years, with the rapid development of maritime activities, the demand for high-speed, reliable, low-latency, and full-coverage marine communication network (MCN) has become increasingly urgent. At present, maritime communication services are mainly provided by satellite networks, but suffers from many limitations, such as surge of data volume, complex communication environment, uneven distribution of traffic and user density and different requirements for maritime services. In order to solve these problems, mobile edge computing (MEC), space-air-ground-sea integrated network (SAGSIN), and blockchain are considered to be promising technologies for MCN enhancement. In this paper, the challenges faced by MCN are discussed, and the above three technologies are applied to solve these challenges. Finally, a space-air-ground integrated network (SAGIN) assisted MCN architecture based on edge computing is proposed, and the future research directions are also put forward.

Mobile crowd sensing (MCS) acts as a key component of Internet of Things (IoT), which has attracted much attention. In an MCS system, participants play an important role, since all the data are collected and provided by them. It is challenging but essential to recruit credible participants and motive them to contribute high-quality data. In this article, we propose a learning-based credible participant recruitment strategy (LC-PRS), which aims to maximize the platform and participants' profits at the same time via MCS participation. Specifically, the LC-PRS consists of two mechanisms, that a learning-based reward allocation mechanism (L-RAM) first calculates the maximum offered reward for different locations based on the number of participants in each location. Under a budget constraint, the proposed L-RAM prefers to collect sensing data from locations in which relatively few data have so far been collected. Furthermore, for each location, we develop a credible participant recruitment mechanism (C-PRM), which employs semi-Markov model and game theory to predict the quality of data provided by each participant and to recruit participants based on the predictions and the maximum offered reward calculated by L-RAM. We formally show LC-PRS has the desirable properties of computational efficiency, selection efficiency, individual rationality, and truthfulness. We evaluate the proposed scheme via simulation using three real data sets. Extensive simulation results well justify the effectiveness of the proposed approach in comparison with the other two methods.