Accurate mobile network traffic prediction is crucial for transit infrastructure service optimization in industrial informatization. Traditional linear models fail to capture complex non-linear dynamics, while existing deep learning methods struggle with rapid temporal changes, signal fluctuations, and diverse network conditions, limiting real-time applicability. To address these challenges, this paper proposes a hybrid model integrating Convolutional Neural Networks (CNNs) and Transformers, tailored for High-Speed Railway (HSR) environments. The proposed hybrid model is evaluated using both public datasets and a real-world HSR dataset collected through empirical field measurements, it not only achieves state-of-the-art (SOTA) predictive accuracy, reducing root mean square error by 4.7% over strong baselines in the challenging HSR environment, but also delivers this performance with superior computational efficiency, achieving over 3.6 times lower inference latency than leading SOTA models. This establishes an optimal performance-to-cost ratio, demonstrating its practical value for real-time HSR systems.

As sixth-generation (6 G) networks evolve towards AI-native architectures, Federated Learning (FL) is becoming a cornerstone for enabling intelligent services by leveraging distributed data from diverse sources such as Integrated Sensing and Communication (ISAC) devices and edge clients. However, a critical challenge lies in efficiently handling data removal requests, mandated by regulations like the “right to be forgotten”. This problem is significantly exacerbated by the extreme data heterogeneity ( non-IID) inherent across diverse 6 G devices and the communication constraints of wireless networks. To address these challenges, this paper proposes SeFUL, a novel two-stage federated unlearning framework tailored for the security and privacy demands of 6 G systems. SeFUL first proactively mitigates data heterogeneity by partitioning clients into clusters based on their data distribution similarity. Subsequently, a lightweight, information-theoretic unlearning strategy is deployed. This method surgically erases information by optimizing a composite loss function which, in the latent space, pushes the feature representations of forgotten data away from their original class cluster and towards samples from other classes, while reinforcing knowledge from the retain set. Comprehensive experiments on benchmark datasets demonstrate that SeFUL achieves unlearning performance on par with the gold standard of complete model retraining. It successfully reduces forget-set accuracy to nearly random guess levels while preserving high retain-set accuracy, significantly outperforming existing state-of-the-art methods. Furthermore, Membership Inference Attacks (MIAs) confirm that SeFUL effectively reduces privacy risks to a level statistically indistinguishable from a fully retrained model, validating its efficacy as a robust privacy-preserving mechanism.

Accurate traffic forecasting in high-speed railway (HSR) systems is hindered by abrupt signal fluctuations and varied mobility scenarios. Conventional approaches that rely on fixed weighted combinations of local and global features are unable to adjust rapidly to real-time changes, resulting in suboptimal performance. To address this limitation, we propose a novel gated guided serial CNN and Transformer network (GsCT) that employs a dynamic combination mechanism implemented via a multilayer perceptron (MLP). In GsCT, CNNs capture fine-grained local variations while Transformers model longrange dependencies, and the adaptive MLP-based gating module adjusts the contribution of each branch based on time-window statistics. This dynamic fusion improves prediction quality by 6.5% compared to conventional fixed weighting mechanisms. Evaluations on both public and real-world HSR datasets demonstrate that GsCT achieves a 2.4% reduction in RMSE relative to LSTM-based methods, and the learned gating coefficients offer transparent interpretability of the feature fusion process. Overall, GsCT provides an effective solution for real-time railway traffic forecasting, paving the way for next-generation HSR services.

With the rapid development of Internet of Vehi-cles (IoV) technology, vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication have become an important part of intelligent transportation systems (ITS). In V2V commu-nication, clustering is essential to optimize network efficiency and reliability. Conventional V2V clustering prioritizes the physical distance between vehicles, neglecting key attributes information of the vehicles (i.e., computational capacity, energy consumption) and behavior like relative motion states between vehicles. Thus, this paper proposes a clustering algorithm based on vehicle attributes and behavioral characteristics (C-VABC) that utilizes the Fuzzy C-Means (FCM) algorithm. Once clusters are established, intelligent data routing decisions between clusters is facilitated by a Deep Q-Network (DQN) based algorithm. This algorithm selects neighboring vehicular Cluster Heads with similar mobility states and higher computational power as the next hop vehicles for packet forwarding. The experimental results demonstrate that the proposed clustering algorithm possesses strong adaptability and reliability, while enhancing the communication quality and efficiency in vehicle networking situations. This provides invaluable support for intelligent transportation systems, as well as intelligent and autonomous driving systems.

With the development of intelligent edge computing (IEC) in industrial IoT (IIoT), there is a growing number of service providers trying to leverage computing resources in the cloud and at the edge to meet the users‘ demand for low latency and high reliability in diversified applications. This evolving landscape necessitates innovative approaches to manage and process the vast amounts of data generated by IIoT devices. Among these approaches, distributed learning frameworks, such as federated learning (FL), have emerged as popular solutions. However, compared to computing, communication remains the primary bottleneck that constrains the speed of federated model training. Most of the previous solutions have focused on reducing communication overhead. Differently, we propose a transmission-centric approach by designing an efficient communication archi-tecture for FL with cloud-edge collaboration, specifically aimed at enhancing communication capabilities through multi-path transmission. We deploy this FL system in a real environment and conduct extensive testing. The results demonstrate that the new approach can significantly reduce communication time in FL setting, thereby enhancing model aggregation efficiency and shortening the overall training duration. Compared to conventional single-path transmission, the proposed solution improves training efficiency by up to 26.4%.

The increasing number of IoT devices in the network brings new challenges to the network carrying capacity of intelligent edge computing, and the complicated network services make the demand for network resources in industrial production scenarios or ordinary network users often exceed the carrying capacity of the edge computing network. To alleviate this problem, this paper proposes an intelligent edge computing architecture that introduces network service identification, extracts and analyses the data characteristics of network traffic, and designs appropriate algorithms to classify network traffic into six different service types. This enables real-time and computing-requiring tasks to be prioritised in the network. Using two machine learning algorithms, KNN and MLP, a model validation is carried out on the constructed dataset, and the results show the effectiveness of the method, with the correct rate of data validation reaching 85%, which is more than 5% higher than the correct rate of direct classification of the specified applications, and the accuracy can be as high as 97% in certain scenarios.

We consider the problem of scheduling audio data transmissions in data-intensive audio sensor networks for animal tracking where the sensor nodes must use WiFi duty cycling to reduce power consumption. WiFi duty cycling entails a startup cost due to probing, authentication, association, and host configuration which are significant and can reach more than 10 seconds. As such, transmission scheduling is not trivial because a naïve approach of switching on WiFi whenever there is a data to send would result in excessive energy consumption overhead. We model duty cycling after an M/G/1 queue with removable server, formulate the optimization problem considering both energy and latency overheads, and obtain the optimal$N^{\ast}$by which the interface should be switched on to schedule data transmission. We propose Optimal Threshold-based Transmission Scheduling (OTTS), a low-complexity algorithm for determining the optimal threshold and commencing transmission. Experiments and trace-based simulations show that OTTS can yield substantial reduction in power consumption that is controllable through an energy-latency trade-off parameter. Compared with interval-based scheduling, OTTS provides lower delay which is more significant at tighter power consumption constraints.

The rise of technologies such as 6G networks, edge computing, and the Industrial Internet has led to a dramatic increase in the amount of data that needs to be transmitted over heterogeneous integrated networks. The resources of embedded devices limit the ability of the Industrial Internet to transmit data. While the multipath transmission mechanism can mitigate data transmission issues of low reliability and low real-time performance from the network-level perspective. As the complexity of industry applications increases, however, the phenomenon that the high-quality data transmission is subject to the influence of the underlying layer is becoming increasingly apparent. The paper aims to explores the possibility of multipath transmission protocol running on a real-time kernel from the perspective of the operating system, as there is a lack of research and reports in this area. Based on RT-Preempt, a real-time system RT-Linux suitable for the “NXP i.MX6Q” ARM integrated board has been proposed, which replaces the native Linux kernel to optimize and enhance its real-time performance. As described in the experiment part, the original standard Linux system OR-Linux and the new RT-Linux are tested with single-threaded and multi-threaded load experiments, respectively. The results of the analysis show that this paper provides a way of validating the trial data and ensuring its accuracy using the lognormal distribution model, which is a statistical distribution used to model variables that are positive and skewed to the right. The RT-Linux scheme has better real-time performance and is more stable than the OR-Linux scheme after real-time processing, showing the viability of the scheme.

Fault diagnosis in Industrial Internet of Things and Cyber-Physical Systems is essential, especially for the rapidly developing 5G and coming 6G technology. Hardware working status is the basis for fault diagnosis. However, in actual engineering, accurately locating faults is not easy for hardware engineers. In this paper, a prototype based on FPGA is developed to connect with the target field device in data acquisition. This prototype aims to extract the underlying data information in a real-time and unattended manner like the mimicry of natural immunity. The preliminary experiment results from the prototype can be used to guide practice through providing real-time data support for hardware troubleshooting and spontaneous and intelligent analysis.

As the demand for real-time and high-quality social network services in mobile communications continues to grow, the performance defects of single-path transmission networks have become more and more prominent. At the same time, multipath transmission provides people with the possibility of stable and smooth communication in mobile wireless social networks. However, since it is usually difficult to obtain frequently varying delays along each path, packets always appear out of order during the communication of heterogeneous social networks, which will cause additional waiting delays in the receiving process. Therefore, it is still a very challenging task to construct a high-bandwidth and low-latency multipath transmission mechanism for social networks in mobile communications. According to the behavioral features of packets of social networks in mobile scenarios, a social network model BAH that reveals wireless social networks bottlenecks is established. Subsequently, this paper proposes a bottleneck-aware algorithm BFDE, which utilizes the one-way delay of periodically probing to derive the features of the wireless social networks bottleneck, so as to achieve an accurate estimation of the delay of each path. In the analysis and simulation, we compared it with the baseline EDPF and proved that the BFDE algorithm can achieve effective scheduling in complex and changeable mobile wireless social networks, thereby effectively increasing the multipath aggregation bandwidth, and has a strong robustness.

The rapid development of the transportation industry has brought about the demand for massive data transmission. In order to make use of a large number of heterogeneous network resources in vehicular network, the research of applying network coding to multipath transmission has become a hot topic. Network coding can better solve the problems of packet reordering and low aggregation efficiency. The determination of coding scale is the key to network coding scheme. However, the existing research cannot adapt to the different characteristics of network resources in vehicular network, leading to larger decoding time cost and lower bandwidth aggregation efficiency. In this paper, we propose a network coding scheme called Delay Determined Group Size (DDGS), which could adaptively adjust the coding group according to the heterogeneous wireless networks state. The mathematical analysis and process design of the DDGS scheme are discussed in detail. Through a large number of simulations, we proved that the DDGS scheme is significantly superior to other coding group determination schemes in terms of decoding time cost and bandwidth aggregation efficiency.

As the demand of mobile services for high bandwidth grows, aggregation of various wireless network resources in hybrid communication system becomes a trend. Network coding has been widely studied to solve the packet reordering, low aggregation efficiency problems which are brought by network heterogeneity. Even if the determination of coding scale is the core issue of network coding, current research can not adapt to heterogeneous wireless channels, resulting in low bandwidth aggregation and decoding efficiency. Therefore, we propose a cross-layer network coding scheme called Delay Determined Group Size (DDGS) scheme, which adaptively adjusts the coding scale to solve the problem of overall performance degradation caused by heterogeneous characteristics of wireless channels. It can get the utmost out of the performance improvement brought by network coding, to better avoid packet reordering as well as reduce the receiving delay at the receiving end. The simulation results show that DDGS is significantly superior to network coding schemes in existing state-of-the-art solutions.

With the development of communication technologies, vehicular network applications have evolved from basic traffic safety and efficiency applications to information and entertainment applications. The implementation of emerging vehicular applications is based on the efficient dissemination of multimedia data. In view of the dynamic topology changes, severe channel fading and limited spectrum resources of vehicular networks, how to achieve efficient multimedia data dissemination in the harsh network environment is an urgent problem. Based on the hybrid cellular-D2D vehicular network, this paper proposes a cluster-based cooperative multicast scheme. The scheme combines multicast transmission with D2D-assisted relay technology to provide high-quality data dissemination for vehicle users under limited spectrum resources. In this paper, we innovatively present a communication quality index that considers multiple performance factors and formulate the relay selection problem as the anti p-center problem in graph theory. Then we propose a heuristic method to solve the problem. The results show that the proposed scheme can effectively improve the utilization of wireless resources and the success rate of data dissemination.

The multipath transmission scheme in vehicular networks has become a hot topic. It is a great challenge to overcome the unreliability of wireless network in multipath transmission. Recently, scholars propose a lot of network coding schemes to solve this problem. These schemes implement network coding algorithms by bitwise X O R or Galois Field arithmetic. However, these schemes cannot take into account both coding flexibility and computational complexity. Therefore, we propose a BigNum Network Coding (BNNC) scheme. The core idea of the BNNC scheme is to treat a packet as an integer and implement the network coding through linear operations on the integer set. It replaces bitwise XOR and Galois Field arithmetic with integer arithmetic that guarantees high coding flexibility and low computational complexity. In this paper, first, we propose the BNN C scheme that can effectively improve the reliability of multipath transmission in vehicular networks with lower computational complexity than current network coding scheme. Second, we design the Independent Matrix that enables the coding process to improve coding efficiency without independent check. Third, we compare BNNC scheme with Earliest Completion First (ECF) and Galois Field network coding scheme through a lot of simulations and real tests. The results show that the BNN C scheme is significantly superior to the Galois Field network coding schemes in terms of computational performance. And in terms of the network performance, the BNNC scheme can overcome the unreliability of links in multipath transmission.

Greater demands are being placed on the access bandwidth, stability, and delay of network because of the quickening rhythm of life and work, especially in mobile scenario. In order to obtain a stable network with low latency and high bandwidth in mobile scenario, taking advantage of the wireless heterogeneous network in parallel is a good choice. Nowadays, people are increasingly concerned about the network quality under the mobile scenario. Some scholars have done the relevant measurements. However, all of those measurements mainly investigate part of the network parameters or part of mobile scenarios. In this paper, we make the following contributions. Firstly, in high‐speed mobile scenario, the wireless network qualities of different vendors are measured synthetically. Secondly, we analyze the benefits of taking advantage of the different vendors. Thirdly, we deploy the replication link mechanism in high‐speed mobile scenario and propose an algorithm to remove the duplicate packet in high‐speed mobile scenario. And the algorithm can also be used in another multipath schedule algorithm to improve the reliability.

The demand for multipath transmission is prominent in 5G networks with the deployment of multiple hierarchical access technologies. However, multipath schemes are still not widely adopted due to many reasons, such as deployment challenges and performance reduction under the circumstances of path heterogeneity. Thus, TCP is still in the dominant position of the transport layer protocol for now and for the foreseeable future. Link asymmetry, such as different latency and different bandwidth of different links, is considered to be the main reasons leading to packet reordering, and further result in TCP performance reduction. However, to the best of knowledge, no one has yet given a theoretical model to analyze the relationship between link asymmetry and TCP multipath performance. In this paper, we present a performance analysis model for TCP over multiple heterogeneous networks, which reveals the effect of link asymmetry on TCP throughput. Both bandwidth and delay asymmetry are taken into consideration in the proposed model. The evaluated throughput using the proposed model can accurately fit the simulation results.

Due to the complexity and fluctuation of the wireless network state in high-speed mobility scenarios, the existing works related to network selection face a great challenge for selecting the accurate network in terms of the imprecise information and mobility. Therefore, we design a novel dynamic imprecise-aware network selection approach, named FSNS by taking advantage of fuzzy logic and utility function of multiple attributes. Our proposed approach not only copes with imprecise network information but also dynamically adapts to the high-speed mobility scenarios, which are not presented for the existing proposals. In this paper, we compare FSNS approach with an enhanced TOPSIS method through simulation experiments of two types of services. The results demonstrate that FSNS outperforms TOPSIS for a preferable decision to keep relatively stable and reduce abnormal selection. The conclusions of experimental results have some extent pragmatic value because the simulation imitates network state in the high-speed mobile environment by real world data from high-speed railways.

Due to the complexity and fluctuation of the wireless network statuses in the high-speed railway scenario, the existing works of the network selection problem in heterogeneous wireless networks face two major challenges, that is, the imprecise statuses and mobility. In this paper, we propose FSNS, a novel dynamic imprecise-aware network selection approach to solve the problems. The imprecise statuses are inferred by fuzzy rules and the status-awareness feature of the status monitor module enables a dynamic network selection. Through the status monitor and fuzzy processing modules, FSNS is formulated as utility functions for meeting quality-of-service (QoS) requirements. What is more, we carry out plenty of simulation experiments and compare FSNS approach with an enhanced TOPSIS method and fuzzy MADM (FMADM) scheme through simulation experiments of two types of services. The results indicate that FSNS outperforms both TOPSIS and FMADM for a good performance improvement and a preferable decision to keep relative stability and reduce abnormal selections. The conclusion of experimental results has some extent pragmatic value since the network statuses of the simulation are complex and fluctuate by real-world data from high-speed railways.

Several wireless sensor network (WSN) applications will in the future be deployed for process control purposes which typically require timely data delivery. The main drawback with current WSNs is that they cannot guarantee predictable delay which is required in many applications. In this paper, we propose WirArb which is a new medium access control (MAC) protocol suitable for time-critical applications that needs deterministic delay guarantees. Each user is pre-defined arbitration frequency which is used to distinguish in which order the users can access the channel. The arbitration frequency is determined by orthogonal sub-carriers in the physical layer. The proposed mechanism will always guarantee the user with highest priority will gain channel access and users with lower priority needs to wait for their turn. We evaluate and compare the proposed WirArb with IEEE 802.15.4 and IEEE 802.11 in terms of latency and throughput. The obtained results show that for a star network with several users the proposed WirArb is superior.

With the growth of different design goals and application requirements, wireless sensor networks (WSNs) are receiving sustained attentions in the recent low-cost industrial automation systems. Moreover, Cognitive Radio (CR) technology gives us a possibility to maximize the utilization efficiency of the limited spectrum resources. However, because the wireless devices coexist in the same radio environment, there are harmful channel conflicts among users, and the increasing radio systems causes great contribution to the increasing energy consumption. In order to realize the industrial circumstance, we complete three major works in this paper. First of all, we describe a practical model of cumulative interferences from the entire cognitive radio-based industrial wireless sensor networks (CR-IWSNs). Then, based on the interference model and the interference avoidance purpose, we propose a self-configurable power control scheme to address the communication requirements on both interference temperature and secondary network Quality-of-Service. Finally, Nonlinear Programming is used to model the scheme and a distributed algorithm is given to solve the problem. Several simulations are given to verify the effectiveness of the proposed power control algorithm on optimizing the total system throughput and energy consumption. Results show that the throughput could be improved and the energy consumption could be reduced with the guarantee that the users are without interference.

With the success of wireless technologies and the number of wireless devices increasing, the wireless coverage footprint expands and removes the needs of wired networks for industrial applications. Wireless industrial sensor networks (WISNs) are a type of wireless sensor networks (WSNs), which dedicate to industrial applications. The WISNs bring several advantages over traditional wired industrial networks for detection and control, include flexibility, scalability, remote maintenance, and rapid deployment. However, some technical issues and routing design principles need to be noticed in terms of the stringent requirements of industrial applications. In this paper, we firstly introduce the characteristics and some routing issues of WISNs. Then we discuss existing routing protocols for WISNs and analysis their performances. Finally, we outline several future research directions.

Wireless sensor network (WSN) is the connection between the physical world and mankind. Particularly, environmental monitoring and devices automatic control of intelligent building based on wireless sensor network is considered as one of the most crucial applications. It can perceive many kinds of environmental parameters and feedback control information to some devices to provide comfortable environment to people. However, it is difficult to deploy a WSN in the buildings because there usually are many wireless LAN devices used in the buildings, which bring serious frequency interferences. In this paper, we conduct a real intelligent building wireless sensor network (IBWSN) for environmental monitoring and air-conditioning automatic control. In order to ensure the effectiveness of this system, actual spectrum analysis is developed. Performance evaluation proves that the presented IBWSN can satisfy the needs of the proposed applications.