In continuous casting and rolling (CCR) systems, precise billet cutting is critical for ensuring product dimensional accuracy and minimizing material waste. However, conventional rule-based cutting strategies rely on static parameters and heuristic rules, which often fail to adapt to real-time disturbances such as thermal expansion, cross-sectional fluctuations, and trimming losses. These limitations frequently lead to suboptimal control performance and excessive residual lengths. Deep reinforcement learning (DRL), with its dynamic decision-making and self-adaptive capabilities, offers a promising alternative. This study proposes an intelligent billet cutting strategy based on the Proximal Policy Optimization (PPO) algorithm, integrating process disturbances, thermodynamic characteristics, and physical constraints into a unified control framework. Extensive evaluations conducted in a high-fidelity CCR simulation environment demonstrate that the proposed method outperforms both traditional approaches and existing RL-based methods in terms of control accuracy, generalization ability, and robustness to disturbances, highlighting its practical potential in intelligent steel manufacturing systems.

In distribution networks, incipient faults often manifest as faint and transient electrical disturbances before fully developing. Incipient fault detection is challenging due to the weak and non-stationary characteristics of fault signatures. Moreover, fault feeder identification is more difficult, as residuals across feeders tend to appear highly similar. To address these challenges, we present FD-Mamba, a Mamba-based neural state-space model that integrates control-theoretic principles with signal-processing techniques. Specifically, we propose a Kalman-inspired neural correction mechanism that performs residual-driven state updates with learnable gain factors. In addition, we introduce a frequency-momentum updating mechanism that stabilizes frequency tracking under non-stationary perturbations. Experimental results on two datasets show that FD-Mamba outperforms existing methods. It achieves a root mean square error of 0.427 and fault feeder detection accuracy of 98.1% on real-world field dataset.

Unmanned aerial vehicles serving as aerial base stations can rapidly restore connectivity after disasters, yet abrupt changes in user mobility and traffic demands shift the quality of service trade-offs and produce strong non-stationarity. Deep reinforcement learning policies suffer from plasticity loss under such shifts, as representation collapse and neuron dormancy impair adaptation. We propose plasticity-enhanced multi-agent mixture of experts (PE-MAMoE), a centralized training with decentralized execution framework built on multi-agent proximal policy optimization. PE-MAMoE equips each UAV with a sparsely gated mixture of experts actor whose router selects a single specialist per step. A non-parametric Phase Controller injects brief, expert-only stochastic perturbations after phase switches, resets the action log-standard-deviation, anneals entropy and learning rate, and schedules the router temperature, all to restore policy plasticity without destabilizing safe behaviors. We derive a dynamic regret bound showing the tracking error scales with both environment variation and cumulative noise energy. In a phase-driven simulator with mobile users and 3GPP-style channels, PE-MAMoE improves normalized interquartile mean return by 26.3% over the best baseline, increases served-user capacity by 12.8%, and reduces collisions by approximately 75%. Diagnostics confirm persistently higher expert feature rank and periodic dormant-neuron recovery at regime switches.

Satellite edge computing offers promising solutions for extending the coverage of terrestrial networks, particularly in remote and harsh environments. By offloading ground data to satellites for processing, this paradigm enables real-time data handling in non-terrestrial networks (NTNs). However, due to limited onboard resources and dynamic task requirements from Internet of Things (IoT) devices, online energy management becomes a critical challenge that hinders the scalability and deployment of satellite edge computing systems. In this paper, we propose an online energy management framework based on satellite-edge collaboration. A queue-based collaboration scheme is designed to coordinate satellites in handling tasks with random arrival patterns. Building upon this scheme, we formulate a joint optimization problem that integrates transmission resource allocation, computational resource assignment, power control, routing strategies, and collaboration policies, aiming to minimize the energy consumption of the satellite network. Given the dynamic, complex, and distributed nature of the problem, we present a Lyapunov-based multi-agent deep reinforcement learning (MADRL) algorithm. Specifically, we first transform the long-term stochastic optimization problem into a sequence of deterministic subproblems using the Lyapunov theory. Sub-sequently, each subproblem is decomposed into a resource allocation subproblem and an edge collaboration subproblem. Correspondingly, we devise a MADRL-based algorithm for edge collaboration and a Lagrangian multiplier iteration (LMI)-based algorithm for resource allocation. Finally, the overall problem is iteratively optimized using the block coordinate descent (BCD) framework. Simulation results demonstrate that the proposed approach achieves significant performance improvements over both baseline methods and several state-of-the-art pure deep reinforcement learning (DRL) approaches.

In order to meet unprecedented demands for reliable and massive connectivity (MC), sixth-generation (6 G) cellular Radio Access Networks (RANs) require architectural innovations. Conventional cellular RANs' scalability is limited by the tightly coupled control and user planes. Fully-Decoupled RANs (FD-RANs) are a promising architectural innovation that enables flexible plane separation. However, current architectures have limitations due to ineffective multiple access schemes. To this end, we introduce NomaFdRaN, an innovative Non-Orthogonal Multiple Access (NOMA)-optimized FD-RAN architecture in order to optimize reliability and MC. To achieve holistic system optimization, NomaFdRaN applies NOMA on all network planes, control plane and user plane, and transmission paths, uplink and downlink. To improve NOMA efficiency, we develop a user pairing optimization approach that minimizes total transmit power while maintaining linear computing complexity. Based on stochastic geometry, we develop analytical models to analyze NomaFdRaN's performance. Subsequently, we analytically derived closed-form expressions for key performance metrics. Our simulation results demonstrate the effectiveness of the NomaFdRaN architecture and provide insights into the deployment strategies for next-generation FD-RANs.

Cloud–edge collaboration and edge intelligence have greatly driven the growth of the Industrial Internet of Things (IIoT). However, the jittery network delay and limited computational resources of edge servers make it difficult to meet the stringent latency requirements in IIoT, and so far there is no good solution to solve this problem. To this end, we introduce scalable neurocomputing, which provides neural networks with different utilities and computation resource requirements, to be deployed on edge servers of cloud–edge IIoT systems. We then optimize such systems by formulating data scheduling and system computational resource allocation as an infinite horizon optimization problem, considering that the data collection from end devices is an infinite long-term process. To solve this hard problem, we design a rolling prediction-optimization framework that transforms the infinite horizon problem into a truncated finite horizon optimization that maximizes the average system utility while satisfying the stringent delay constraints. We have conducted extensive simulations and built a prototype system, which verify the feasibility and performance of our proposed scheme.

Sensor devices continuously generate large volumes of time series data in the Internet of Things (IoT) environment. These voluminous streams require models that scale to massive data while discerning the intricate, multi-scale patterns embedded in diverse temporal sequences. Transformer models have been widely used for IoT time series analysis due to their strong feature representation and global modeling capability. However, existing architectures struggle to explicitly capture temporal structures and adapt to non-stationary data, limiting classification performance. To address these issues, we propose a novel attention mechanism based on the autocorrelation function, named A2T, which leverages lag characteristics to unify temporal modeling and feature extraction. We further introduce a Parameterized Wavelet Transform Module that learns scale and bandwidth end-to-end and uses an attention gate to fuse multi-resolution coefficients. Building on this, we design a Dual-Channel Time-Frequency Feature Extraction module to improve adaptability to distribution shifts. Integrating these components, we develop A2Tformer for IoT time series classification. Experimental results on the UCR dataset demonstrate that A2Tformer achieves an average accuracy of 84.49% and ranks first on 26 out of all datasets, outperforming state-of-the-art Transformer-based models.

Unmanned aerial vehicle (UAV)-assisted communication networks offer an effective solution for establishing connectivity and enabling data delivery in scenarios where direct communication with terrestrial infrastructure is infeasible, such as in disaster areas. In such networks, unknown user locations, dynamic node mobility, and multi-hop transmission collectively increase the complexity of the communication process. Specifically, the uncertainty of user positions makes it difficult for UAVs to establish and maintain reliable communication links, while the mobility of both mobile users (MUs) and UAVs leads to frequent fluctuations in link quality. Furthermore, the dynamic network topology increases the complexity of routing in multi-hop transmissions, requiring UAVs to continuously adjust their trajectories to ensure stable and efficient data relaying. In this paper, we apply received signal strength indicator (RSSI)-based multilateration and deep reinforcement learning (DRL) to address the associated difficulties. Specifically, UAVs estimate the positions of nearby MUs using RSSI-based multilateration. The joint problem of mobility control and data forwarding is formulated as an Markov decision process (MDP), and a multi-agent proximal policy optimization (MAPPO) algorithm is employed to learn decentralized decision-making policies. Extensive simulations demonstrate that the proposed scheme improves transmission completion ratio and reduces transmission delay in UAV-assisted communication networks.

Radio Access Networks (RAN) are anticipated to gradually transition towards Cloud RAN (C-RAN), leveraging the full advantages of the cloud-native computing model. While this paradigm shift offers a promising architectural evolution to improve scalability, efficiency, and performance, significant challenges remain in managing the massive computing requirements of physical layer (PHY) processing. To address these challenges and meet the stringent Service Level Objectives (SLOs) in 5G networks, hardware acceleration technologies are essential. In this paper, we aim to mitigate this challenge by offloading 5G modulation mapping, a critical yet demanding function to encode bits into IQ symbols, directly onto the switch ASICs. Specifically, we introduce NetMod, a 5G New Radio (NR) standard-compliant in-network modulation mapper accelerator. NetMod leverages the capabilities of new-generation programmable switches within the C-RAN infrastructure to offload and accelerate PHY modulation functions. We implemented a NetMod prototype on a real-world platform using the Intel Tofino programmable switch and commodity servers running the Data Plane Development Kit (DPDK). Through extensive experiments, we demonstrate that NetMod achieves modulation mapping at switch line rate using minimal switch resources, thereby preserving ample space for traditional switching tasks. Furthermore, comparisons with a GPU-based 5G modulation mapper show that NetMod is 2.2$\boldsymbol{\times}$to 3.3$\boldsymbol{\times}$faster using only a single switch port. These results highlight the potential of in-network acceleration to enhance 5G network performance and efficiency.

Federated learning (FL) is a promising artificial intelligence framework that enables clients to collectively train models with data privacy. However, in real-world scenarios, to construct practical FL frameworks, several challenges have to be addressed, including statistical heterogeneity, constrained resources, and fairness. Therefore, we first investigate anaggregation gapcaused by statistical heterogeneity during local model initialization, which not only causes additional computational overhead for clients but also leads to the degradation of fairness. To bridge this gap, we proposepFedCal, a novelpersonalizedfederated learning with lightweight adaptivecalibration strategy that performs calibration compensation through the prior knowledge of clients. Specifically, we introduce compensation for each client at the model initialization, with the compensation derived from the global gradient and the latest gradient bias. To enhance the calibration effect, we introduce a smoothing-based calibration strategy, and we design an adaptive calibration strategy. A representative example demonstrates that the proposed calibration and smoothing strategies improve fairness for clients. The theoretical analysis indicates that with an appropriate learning rate, pFedCal converges to a first-order stationary point for non-convex loss functions. Comprehensive experimental results show that pFedCal achieves faster convergence, higher accuracy, and improved fairness than the state-of-the-art methods.

Supporting mobile edge computing (MEC) in satellite-terrestrial networks (STNs) provides essential offloading services for devices for the Internet of Things (IoT) devices in remote areas. However, when terrestrial demands for computing resources are high, the MEC servers on visible LEO satellites may suffer from insufficient capacity, while those on more distant LEO satellites remain underutilized. To address this issue, this article investigates cooperative task offloading across multiple LEO satellites within an MEC-based STN. We propose a Quality-of-Service (QoS)-aware offloading decision and resource allocation scheme supported by a software-defined network (SDN) for a STN architecture. This architecture integrates the LEO Walker constellation with satellite ground stations (SGSs), with the aim of providing edge computing services to IoT devices in remote areas. To meet the task’s QoS requirements, the tasks can be offloaded to either SGS or LEO satellites within the constellation. To address the challenges of a vast state space and complex action space within the system, we introduce the QOS-aware multihop task offloading in satellite-terrestrial edge networks (OUTSIDE) algorithm, which combines the global search capabilities of genetic algorithms with the local refinement strengths of the Lagrangian multiplier method to minimize the total task computation latency while satisfying QoS demands. Finally, comparative analysis and simulation experiments were conducted. These demonstrate that the OUTSIDE algorithm outperforms other approaches in terms of efficiency and effectiveness.

Federated learning has emerged as a promising approach for fault diagnosis, as its ability to learn from decentralized data while preserving client privacy for industry. Yet, it also brings the problem of nonidentically and independently distributed (Non-IID) data, which can result in model convergence delay and performance degradation. Recent research aims to alleviate the problem caused by cross-domain without considering by cross-location. However, it is common in industrial production to have devices across different monitoring locations. Furthermore, experimental results indicate that the diagnostic models' performance of the latest techniques is significantly affected. To address the cross-location Non-IID data problem, we propose DecFFD, a personalized federated fault diagnosis framework that decouples global and personalized features. In DecFFD, we design a reconstructor for each client that acts as a supervisor and decoupler to disentangle global and personalized features. We then present a client alignment algorithm to eliminate the differences in global features among clients. In addition, we provide a theoretical analysis of fairness and generalization capability, offering a theoretical guarantee for model convergence. Finally, extensive experiments are conducted on two real-world datasets. Experimental results show that the accuracy of DecFFD outperforms the accuracy that of the state-of-the-art approach by 14.67% and converges at a faster rate.

The integration of Roadside Units (RSUs) is vital for the development of autonomous driving technologies. Challenges arise from sinking computing capabilities into RSUs and vehicles in the paradigm of Vehicle Edge Computing (VEC), particularly due to heterogeneous computation and communication capacities of network nodes and multiple sources of computing tasks (node-mounted and offloading tasks). These challenges complicate network stability from the perspective of a long-term optimization evolving over time, considering unpredictable task distribution and environmental states. To tackle these challenges, we approach the problem of partial task offloading to minimize task delay while meeting the demand of system stability over time as a dynamic long-term optimization. Utilizing Lyapunov stochastic optimization tools, we successfully decouple the long-term delay minimization and stability constraint, transforming it into a per-slot scheduling problem. Since the per-slot scheduling problem with complicated Lyapunov drift functions can not be solved by numerical optimization at each time step, our solution leverages a proposed deep reinforcement learning algorithm, leading to extensive simulations that demonstrate the superior effectiveness and efficiency of our proposal compared to existing schemes.

In the context of Internet of Vehicles, Roadside Units (RSUs) play a crucial role in reducing the task delay through their strong computing capabilities. Nevertheless, extended usage of RSUs accelerates the battery degradation and impacts their service capacity. By offloading tasks to the cloud and vehicles, the degradation rate of the RSU battery is reduced at the expense of high task delay. Therefore, the problem of balancing task offloading delay and RSU battery degradation is a challenge. To address this issue, in this paper, we propose a joint optimization model that considers both task offloading delay and RSU battery degradation. RSUs actively perceive information from environment through their sensors. Tasks are processed either through local computation on RSUs, offloading computation to vehicles, or offloading computation to the cloud. By computing the Depth of Discharge (DOD) of the battery, we establish a battery degradation cost model to assess the degree of battery degradation. Due to the complexity of the environment, the optimization problem of jointly considering task offloading delay and RSU battery degradation is a nonconvex problem that is difficult to solve. Thus, this paper proposes transforms the problem into a Markov decision process and applies the Proximal Policy Optimization (PPO) algorithm to strike a balance between task offloading delay and RSU battery degradation. Experimental results have demonstrated the effectiveness of our proposal.

Fuzzing is a popular technique for finding soft-ware vulnerabilities. Despite their success, the state-of-art fuzzers will inevitably produce a large number of low-quality inputs. In recent years, Machine Learning (ML) based selection strategies have reported promising results. However, the existing ML-based fuzzers are limited by the lack of training data. Because the mutation strategy of fuzzing can not effectively generate useful input, it is prohibitively expensive to collect enough inputs to train models. In this paper, propose a generative adversarial networks based solution to generate a large number of inputs to solve the problem of insufficient data. We implement the proposal in the American Fuzzy Lop (AFL), and the experimental results show that it can find more crashes at the same time compared with the original AFL.

Deep Neural Networks (DNNs) have achieved excellent performance in intelligent applications. Nevertheless, it is elusive for devices with limited resources to support computationally intensive DNNs, while employing the cloud may lead to prohibitive latency. Better solutions are exploiting edge computing and reducing unnecessary computation. Multi-exit DNN based on the early exit mechanism has an impressive effect in the latter, and in edge computing paradigm, model partition on multi-exit chain DNNs is proved to accelerate inference effectively. However, despite reducing computations to some extent, multiple exits may lead to instability of performance due to variable sample quality, performance inferior to the original model especially in the worst case. Furthermore, nowadays DNNs are universally characterized by a directed acyclic graph (DAG), complicating the partition of multi-exit DNN exceedingly. To solve the issues, in this paper, considering online exit prediction and model execution optimization for multi-exit DNN, we propose a Dynamic Path based DNN Synergistic inference acceleration framework (DPDS), where exit designators are designed to avoid iterative entry for exits; to further promote computational synergy in the edge, the multi-exit DNN is dynamically partitioned according to network environment to achieve fine-grained computing offloading. Experimental results show that DPDS can significantly accelerate DNN inference by 1.87× to 6.78×.

Abstract Accurate demand prediction of bike-sharing is an important prerequisite to reducing the cost of scheduling and improving the user satisfaction. However, it is a challenging issue due to stochasticity and non-linearity in bike-sharing systems. In this paper, a model called pseudo-double hidden layer feedforward neural networks is proposed to approximately predict actual demands of bike-sharing. Specifically, to overcome limitations in traditional back-propagation learning process, an algorithm, an extreme learning machine with improved particle swarm optimization, is designed to construct learning rules in neural networks. The performance is verified by comparing with other learning algorithms on the dataset of Streeter Dr bike-sharing station in Chicago.

It is an inspiring way to provide emergence communication services in natural disaster areas by deploying wireless routers on the ground and multiple unmanned aerial vehicles (UAVs) in the air. The wireless routers serve as access points. UAVs relay data from routers and themselves to a remote base station in a safe place. Thus, people can communicate with others outside the disaster. The network lifetime is restricted to battery lifetime of routers which are scattered over a complex post-disaster area. There is a potential to prolong the network lifetime by utilizing UAV mobility. We consider the trajectory planning of UAVs with the goal of maximizing the network lifetime, which is modeled as a cooperative game with complete information. However, the time complexity of the problem increases exponentially with the number of UAVs as well as UAV candidate strategies. In our proposal, we boost the game process by excluding some of candidates from the strategy space for each UAV. Specifically, there is no influence for a given UAV, taking the strategies excluded, over the other UAVs. In addition, these strategies are dominated by another strategy at least. Our proposed model is verified through simulations that show its advantage on time complexity over others.

Deploying wireless routers on the ground and unmanned aerial vehicles (UAVs) in the air is believed to be a fast and efficient approach to providing the emergency communication service to disaster areas. The network lifetime is restricted to the lifetime of mesh networks of routers that are scattered across a complex disaster environment. We consider the problem of multi-UAVs relaying and moving with the goal of maximizing the network lifetime. UAVs must learn where to move in order to relay messages from the routers efficiently. However, under the dynamics of the router traffic, that is, the uncertainty of the environment, it is challenging for UAVs to find a movement mechanism that maximizes the network lifetime. By embracing the on-policy reinforcement learning algorithm Sarsa, we are able to demonstrate a greedy movement policy. Specifically, we study the trajectory planning of multiple UAVs in a dense wireless router mesh networks (WMNs) on the ground. UAVs can learn the unknown environment by a little movement of UAVs in each step, in which communication connections between routers with UAVs are retained. A Q-table of movement actions and environment states is formed after multiple attempts of movements. Simulation results show the proposal effectiveness comparing with other methods.

Unmanned aerial vehicles (UAVs) can provide important communication advantages to ground-based wireless mesh networks with multiple connected routers. Their usage in civilian areas has become commonplace, especially more promising recently as emergency communication networks for rescuing victims, dispatching resources and recovering communications to bridge the gap between the post-disaster areas and outsides. First, usually a flash crowd of victims will attempt to make a video or VoIP connection with outsides. Furthermore, it results in a large volume of traffic shift among mesh routers due to high migration of people such as leaving and joining in the refuges. Thus, in this paper we consider the optimization of the location and movement of a single UAV to improve the network throughput by means of high mobility of the UAV to adapt to heavy fluctuation of network traffic. Though there are several methodologies to the UAV movement, it is still indistinct which one is better, especially while employing different routing protocols. A detailed comparison between two typical schemes based on the mean value and the collision domain is made for UAV-assisted wireless mesh networks with multimedia burst. Specifically, we designed two schemes accordingly to select the UAV position and the mesh router as the gateway to connect the UAV. A numerical experiments with different routing protocols are evaluated and compared in a realistic simulation environment of NS-3 and show that they both have advantages and disadvantages with respect to QoS when employing different routing protocols in the considered scenario.

The design of a wireless network is often critically affected by issues such as determining the optimal density of Access Points (APs) and the optimal channel assignment by exploiting partially overlapped channels (POCs) for significantly improving the network performance in terms of maximizing the overall network capacity. Contemporary research works have traditionally dealt with these two problems in an isolated manner though they should be considered within the same problem formulation. Furthermore, even though deployment of additional APs can improve the network capacity in case there are a few APs in a given area, the APs cannot be indefinitely added to the wireless network. This means that there is an upper bound to the network capacity maximization with respect to the number of APs. In fact, the network capacity starts to dramatically decrease when the number of deployed APs becomes excessive. This performance decrease can be accredited to the substantial interference among the high number of deployed APs. In order to address this challenge, in this paper, we propose an approach to jointly optimize the number of APs and POCs assignment. Our proposal derives the existence of the optimal density of APs with POCs, and models the POC assignment to the deployed APs from a novel perspective. Computer-based simulations are conducted to demonstrate the effectiveness of our proposal.