IIoT sensors collect large volumes of vibration time-series data at high sampling rates. These data are nonstationary and multi-scale and are essential for condition monitoring across different devices and operating conditions. Deep compression sensing reduces data volume while preserving critical information, enabling efficient and low-cost data processing in IIoT systems. However, existing methods for industrial time-series data processing struggle to preserve features effectively under stringent bandwidth and storage constraints. Moreover, most deep compression sensing models overlook computational limitations and robustness demands in noisy industrial settings. Therefore, we propose an Attention-Based Deep Compressed Sensing with Multi-Loss Optimization for IIoT Vibration Data (DCS-AMTD). The model achieves efficient compression and high-quality reconstruction while improving both resource efficiency and noise robustness. Specifically, we design a dual-path convolutional module that incorporates dilated convolutions to capture multi-scale local and global features. We design a one-dimensional convolutional block attention module (CBAM1D) for industrial vibration signals to dynamically reweight multi-scale features and enhance discriminative representations. Furthermore, we design a joint time-frequency loss with multi-domain constraints to improve reconstruction quality under strict bandwidth and computational constraints. Experiments on the Case Western Reserve University (CWRU) and Paderborn (PB) datasets demonstrate that our method achieves superior reconstruction performance across various compression ratios, outperforming existing approaches.

In smart grids, smart meters periodically collect users' fine-grained multi-dimensional energy data, which poses great concerns on users' privacy and security. Existing privacy preserving multi-dimensional aggregation schemes suffer from heavy computational burdens, especially for smart meters with limited computational resources. To address these limitations, in this paper we propose an efficient and flexible multi-dimensional data aggregation scheme called MDA-SMuSha, by which smart meters employ the Shamir's multi-secret sharing to generate a set of shared secrets, with the first one kept locally, while the remained ones are packaged and then uploaded to a control center via an aggregator. By the MDA-SMuSha scheme, aggregation results of smart meters' multi-dimensional energy data during multiple periods can be obtained, with only one time of Paillier encryption conducted on the smart meters. In addition, it allows the control center to send query requests flexibly, i.e., at a pre-specified frequency or whenever it wants to obtain statistical data of interests. Rigorous security analyses show that the MDA-SMuSha scheme satisfies security requirements of privacy-preservation, authenticity and data integrity as well as fault-tolerance. Both theoretical analyses and experiment results show that the MDA-SMuSha scheme outperforms state-of-the art methods in terms of computation costs, with comparable communication costs.

In pursuit of safe and efficient distributed cooperative control of multi-robot systems (MRSs), a logic-adaptive discrete neural dynamics (LADND)-based minimum infinity norm (MIN) strategy is introduced in this paper. The MIN strategy is employed to address critical safety concerns caused by excessively high velocity in the individual joint. To further improve the adaptivity of the neural dynamics solver, a fuzzy system is integrated to enable adaptive parameter adjustment based on the real-time behavior of MRSs. Specifically, the cooperative control problem is formulated as a linear program incorporating the MIN along with constraints associated with distributed network topology and orientation maintenance, thereby enhancing the safety and effectiveness of MRSs. To efficiently solve the proposed linear program, an LADND solver is developed, which adaptively adjusts its parameter in real time according to the trajectory tracking error and its derivative. Furthermore, theoretical analyses confirm the convergence and robustness of the proposed LADND solver. Simulative and experimental results validate the effectiveness of the proposed LADND-based MIN strategy in cooperative trajectory tracking tasks.

Driven by the new generation of information technology, modern manufacturing is shifting from traditional rigid production to flexible, customized models. This transformation imposes higher demands on systems' dynamic adaptability and real-time responsiveness. The emergence of Edge Computing (EC) and the Age of Information (AoI) offers promising solutions to these challenges. Significant progress has been made in areas such as edge resource allocation and information update strategies, contributing to enhanced system responsiveness. However, most existing studies assume task independence, which limits their applicability in complex scenarios (such as high-end manufacturing), where task dependency is dynamic and ubiquitous. In addition, ensuring end-to-end integration of sensing, transmission, computation, and control (STCC) remains a major challenge. To address this gap, this study proposes a collaborative framework focusing on sensing, transmission, computation, and control (STCC), centering around task chains. For the first time, it introduces and defines the Dependency-Aware Age of Information (DAoI) metric to quantify inter-task dependencies and the effects of delay propagation, thereby enabling a more accurate assessment of data timeliness. Additionally, an optimal controller using the Linear Quadratic Regulator (LQR) is designed. Our model optimizes control performance and energy consumption through a Markov Decision Process (MDP). The proposed Dependency-Aware Heterogeneous Task and Resource Co-scheduling (MAPPO-DHTCO) algorithm efficiently manages task dependencies and resource allocation. Experimental results show that this method has significantly improved performance compared to the benchmark method.

Meeting the deterministic demands of industrial tasks can be quite challenging due to the diversity of devices and the unclear relationship between tasks and platforms in industrial edge computing scenarios. To tackle this issue, this study introduces an entropy-weighted scheduling method grounded in resource quantification. First, we scrutinized the affinity challenge when tasks operate across different platforms and broadened the scope of scheduling evaluation criteria within existing real-time systems. This expansion was accomplished by examining the alignment between various task attributes and platform characteristics through resource quantification. Subsequently, we employed the entropy weight method to handle the information entropy of all scheduling evaluation criteria and calculated the weighted sums to allocate the optimal scheduling device for each task. Ultimately, the entropy-weighted scheduling algorithm, which relies on resource quantification, was formulated to assess the algorithm’s scheduling performance under various parameter configurations. The experimental analysis indicated that the scheduling method based on resource quantification could effectively optimize the resource demand relationship between tasks and platforms, and the scheduling success rate of the proposed algorithm was 5.1%, 7.7%, and 34.5% higher than those of multitarget tracking sensor scheduling algorithm, D-Quantify, and RRA algorithms, respectively.

Blockchain-based vehicular edge computing (VEC) is regarded as a promising computing paradigm that can enhance the computing capabilities of mobile vehicles while ensuring security during task offloading. However, the blockchain consensus for secure task offloading inevitably increases the communication and computation resource consumption. More importantly, the frequent handover among roadside units during the fast movement of vehicles also raises the communication cost for blockchain consensus. To address these issues, this article proposes intelligent secure task offloading and caching (ISTOC) scheme for VEC networks. Specifically, we first establish a digital twin-assisted VEC network that migrates the blockchain consensus process from the physical space to the cyber space, supporting the dynamic handover of vehicles. Correspondingly, we propose a lightweight blockchain scheme named diffused delegated Byzantine fault tolerance (d2BFT). Then, aiming at simultaneously reducing the task processing latency and improving the blockchain transaction throughput, we formulate the joint blockchain, communication, computation, and caching (B3C) optimization problem subject to task division, communication bandwidth, computing frequency, cache storage, task deadline, and blockchain stability. Due to the nonconvexity of B3C, we transform it into a Markov decision process, and propose a multiagent double actor-critic (MADAC) algorithm in light of the distributed characteristic of blockchain. Through offline training and online execution, we jointly optimize the task division, communication bandwidth, computing frequency and cache storage allocation, block size, and block generation interval for ISTOC. Experimental results show that the proposed MADAC-based ISTOC scheme can stably converge with a much higher reward than the benchmark schemes based on MADDPG, soft actor-critic, deep deterministic policy gradient, and TD3. The improvement of MADAC-ISTOC over SAC-ISTOC is more than 25.93%.

The coexistence of ultra-reliable low-latency communication (URLLC) and enhanced mobile broadband (eMBB) services in 5G-based industrial wireless networks (IWNs) poses significant resource slicing challenges due to their inherent performance requirement conflicts. To address this challenge, this paper proposes a puncturing method that uses a model-aided deep reinforcement learning (DRL) algorithm for URLLC over eMBB services in uplink 5G networks. First, a puncturing-based optimization problem is formulated to maximize the eMBB accumulated rate under strict URLLC latency and reliability constraints. Next, we design a random repetition coding-based contention (RRCC) scheme for sporadic URLLC traffic and derive its analytical reliability model. To jointly optimize the scheduling parameters of URLLC and eMBB, a DRL solution based on the reliability model is developed, which is capable of dynamically adapting to changing environments. The accelerated convergence of the model-aided DRL algorithm is demonstrated using simulations, and the superiority in resource efficiency of the proposed method over existing approaches is validated.

Blockchain-empowered end-edge collaborative computing is a promising technology for enhancing the timeliness and trustworthiness of Industrial Internet of Things (IIoT). However, integrating task offloading with blockchain consensus inevitably escalates resource consumption across communication, computation, and energy domains. Thus, the joint optimization of task offloading, resource allocation and blockchain consensus is very important for IIoT. This paper studies a general end-edge collaborative computing scenario with multiple end devices and multiple edge servers. We first propose a novel dynamic blockchain (DBC) scheme by developing a dynamic leader election mechanism and designing a dynamic consensus waiting time window. Then, by fully considering the constraints of multi-task size and deadline, communication bandwidth, computing frequency, battery capacity, Byzantine fault tolerant and trustworthiness, we formulate the trustworthy processing efficiency (TPE) maximization problem with respect to end-edge task division, communication and computation resource allocation, leader election and consensus waiting window. To address this problem, we transform it into a Markov decision process and design a compound reward by fully considering the penalty for computing timeout and consensus failure. After that, we propose a rotating multi-agent deep reinforcement learning (R-MADRL) algorithm tailored to the proposed DBC scheme, where an entropy-based dual-critic DRL algorithm is proposed for rotating multi-agent training and decentralized execution. Extensive experiments validate the effectiveness and superiority of the proposed DBC with R-MADRL, where three benchmark DRL algorithms and three blockchain consensus schemes are compared. The results demonstrate that R-MADRL achieves stable convergence with more than 60.32% TPE reward than other algorithms while the task timeout ratio of DBC is reduced by more than 66.49% compared with other schemes.

Deterministic response is the core foundation for the safe operation of industrial production systems. However, with the increasing demand for intelligence, flexibility, and agility, ensuring the deterministic response of computing and control tasks while meeting new demands has become the primary issue that manufacturers urgently need to address. In response to this issue, this article focuses on AI-driven cyber–physical production systems (AI-CPPSs) and conducts research on the adaptive interaction mechanism of network, computing, and manufacturing resources with guaranteed performance. The efficient adaptive configuration of network, computing, and manufacturing resources is used to meet the response requirements of dynamic tasks. To achieve on-demand configuration of multidimensional resources for tasks, we first propose an AI-CPPS-oriented modeling method named the hourglass method, which redefines task models and multidimensional resources with resources as the core. Furthermore, through the proposed method of computing power quantification and a heterogeneous frame structure, we achieve the unified arrangement of network, computing, and manufacturing resources in the time dimension. Finally, to ensure the security and reliability of resource interaction, a multidimensional resource interaction mechanism is proposed for network computing control, namely, the quicksand mechanism. The experimental results indicate that the proposed quicksand mechanism can optimize resource utilization based on ensuring a deterministic task response.

Industrial Internet, motivated by the deep integration of new-generation information and communication technology (ICT) and advanced manufacturing technology, will open up the production chain, value chain, and industry chain by establishing complete interconnections between humans, machines, and things. This will also help establish novel manufacturing and service modes, where personalized and customized production for differentiated services is a typical paradigm of future intelligent manufacturing. Thus, there is an urgent requirement to break through the existing chimney-like service mode provided by the hierarchical heterogeneous network architecture and establish a transparent channel for manufacturing and services using a flat network architecture. Starting from the basic concepts of process manufacturing and discrete manufacturing, we first analyze the basic requirements of typical manufacturing tasks. Then, with an overview on the developing process of industrial Internet, we systematically compare the current networking technologies and further analyze the problems of the present industrial Internet. On this basis, we propose to establish a novel “thin waist” that integrates sensing, communication, computing, and control for the future industrial Internet. Furthermore, we perform a deep analysis and engage in a discussion on the key challenges and future research issues regarding the multi-dimensional collaborative sensing of task–resource, the end-to-end deterministic communication of heterogeneous networks, and virtual computing and operation control of industrial Internet.

Task offloading and resource allocation are fundamental and crucial for the edge computing-enhanced industrial Internet of things, where the security and credibility among massive heterogeneous devices are being challenged. This paper first proposes a novel blockchain consensus scheme named replicated and Byzantine fault tolerant, which can enhance the trust among nodes with the low communication cost. Then, with the objective of minimizing the task completion time, which includes credible verification, task offloading and transaction record, a joint task offloading and resource allocation problem with respect to blockchain verification ratio, offloading decision, communication and computing resources is formulated. Due to its non-convexity and the decentralized characteristic of blockchain, a multi-agent deep reinforcement learning algorithm with deterministic policy gradient is proposed to appropriate the optimal solution. Experiment results confirm the effectiveness of the proposed scheme in guaranteeing the timeliness and security.

The cyber–physical production system (CPPS) has practical requirements, such as distributed, reconfigurable, and high-performance, which bring a new challenge to performance guarantee of remote manufacturing with time delay under shared resources. Time delay has a great influence on system performance, and the mainstream methods mainly reduce its influence by optimizing control. However, due to the uncertainty of time delays, the existing methods have great limitations, and the configuration complexity is high. To address this issue, we take teleoperation as the control model, then we introduce 5G slicing and edge computing technologies to turn this control problem into a control–communication–computing co-design problem. An industrial teleoperation testbed is implemented to help clarify this problem and explore the key points in solving it. Then, a novel co-design teleoperation platform (CdTP) is designed that can quantitatively describe the relationship between the time delay and system configuration. Based on CdTP, system performance assurance does not have to be achieved by blindly increasing the total amount of resources, but can be achieved by improving the utilization of shared control–communication–computing resources. In addition, CdTP can dynamic configure resources based on changing requirements, which make it combines flexibility, real time, and reliability. Finally, we propose a resource allocation method to minimize the maximum job delay. The evaluation and experimental results indicate that our platform can achieve an all-in-one configuration, and validation of the proposed method is conducted to provide deterministic delay guarantees.

With the interdisciplinary advances of mobile communication and edge computing, massive heterogeneous tasks are accessing wireless networks and competing for the edge-end computing and communication resources. Digital twin (DT), which establishes the digital models of physical objects for simulation, analysis and optimization, provides a promising method for network scheduling and management. This paper proposes a DT-driven edge-end collaborative scheduling algorithm for heterogeneous tasks and heterogeneous computing/communication resources. Specifically, multiple end devices (EDs) cooperate with each other to accomplish a complex job, where each ED can offload individual task to multiple edge servers (ESs) for parallel computing. By fully considering deadline requirements of heterogeneous tasks, maximum computing capabilities of ESs and EDs, computing resource estimation deviations of DT, maximum transmit powers of EDs and tolerable peak interference powers to coexisting EDs, we formulate a job completion time minimization problem to jointly optimize the edge-end task division, transmit power control, computing resource type matching and allocation. To solve this non-convex problem, we first reformulate it by multi-agent Markov decision process, where a compound reward leveraging latency reward and deadline reward according to the task criticality is designed. Then, we propose a multi-agent deep reinforcement learning-based scheduling algorithm, where Actor-Critic framework with estimation and target networks is designed for policy and value iterations. Meanwhile, a step-by-step ϵ-greedy algorithm is proposed to balance exploration and exploitation, avoiding local optimal trap. Through offline centralized training by DT and online distributed execution by EDs, we realize edge-end collaborative computing for heterogeneous tasks. Experimental results demonstrate that, comparing with typical benchmark algorithms, the proposed algorithm converges with the highest reward and achieves the smallest job completion time, where the deadlines of heterogeneous tasks can be well satisfied respectively.

Compared to industrial wired networks, 5G can improve device mobility and reduce the cost of networking. However, the real-time performance and reliability of 5G new radio (NR) still need to be improved to satisfy industrial applications’ requirements. In factories, the main factor that affects the performance of 5G NR is the unstable signal quality caused by high temperatures and metal. Although assigning dedicated resources to all transmissions and retransmissions is an effective method to improve the performance of 5G NR, the unstable signal quality causes the resources required for retransmissions to be uncertain. To address the problem, we introduce the mixed-criticality task model to 5G NR. When high-criticality packets cannot be transmitted, they are allowed to preempt the resources shared with low-criticality packets. The mixed-criticality scheduling problem of 5G NR is NP-hard. We formulate it as an optimization modulo theories (OMT) specification and propose a scheduling algorithm based on bin packing methods to make 5G NR satisfy industrial applications’ requirements. Finally, we conduct extensive evaluations based on an industrial 5G testbed and random test cases. The evaluation results indicate that our algorithm makes communication reliability greater than 99.9% on unlicensed spectrum, and for most test cases, our algorithm is close to optimal solutions.

Edge artificial intelligence will empower the ever simple industrial wireless networks (IWNs) supporting complex and dynamic tasks by collaboratively exploiting the computation and communication resources of both machine-type devices (MTDs) and edge servers. In this paper, we propose a multi-agent deep reinforcement learning based resource allocation (MADRL-RA) algorithm for end-edge orchestrated IWNs to support computation-intensive and delay-sensitive applications. First, we present the system model of IWNs, wherein each MTD is regarded as a self-learning agent. Then, we apply the Markov decision process to formulate a minimum system overhead problem with joint optimization of delay and energy consumption. Next, we employ MADRL to defeat the explosive state space and learn an effective resource allocation policy with respect to computing decision, computation capacity, and transmission power. To break the time correlation of training data while accelerating the learning process of MADRL-RA, we design a weighted experience replay to store and sample experiences categorically. Furthermore, we propose a step-by-step ε -greedy method to balance exploitation and exploration. Finally, we verify the effectiveness of MADRL-RA by comparing it with some benchmark algorithms in many experiments, showing that MADRL-RA converges quickly and learns an effective resource allocation policy achieving the minimum system overhead.

In Industry 4.0, massive heterogeneous industrial devices generate a great deal of data with different quality of service requirements, and communicate via industrial wireless networks (IWNs). However, the limited time-frequency resources of IWNs cannot well support the high concurrent access of massive industrial devices with strict real-time and reliable communication requirements. To address this problem, a deep reinforcement learning-based dynamic priority multichannel access (DRL-DPMCA) algorithm is proposed in this article. Firstly, according to the time-sensitivity of industrial data, industrial devices are assigned with different priorities, based on which their channel access probabilities are dynamically adjusted. Then, the Markov decision process is utilized to model the dynamic priority multichannel access problem. To cope with the explosion of state space caused by the multichannel access of massive industrial devices with dynamic priorities, DRL is used to establish the mapping from states to actions. Next, the long-term cumulative reward is maximized to obtain an effective policy. Especially, with joint consideration of the access reward and priority reward, a compound reward for multichannel access and dynamic priority is designed. For breaking the time correlation of training data while accelerating the convergence of DRL-DPMCA, an experience replay with experience-weight is proposed to store and sample experiences categorically. Besides, the gated recurrent unit, dueling architecture and step-by-step$\varepsilon$-greedy method are employed to make states more comprehensive and reduce model oscillation. Extensive experiments show that, compared with slotted-Aloha and deep Q network algorithms, DRL-DPMCA converges quickly, and guarantees the highest channel access probability and the minimum queuing delay for high-priority industrial devices in the context of minimum access conflict and nearly 100% channel utilization.

This article studies a green-energy-powered cognitive radio network (GCRN) in an underlay paradigm, wherein multiple battery-free secondary users (SUs) capture both the spectrum and the energy of primary users (PUs) to communicate with an access point (AP). By time division multiple access, each SU transmits data to AP in the allocated time and harvests energy from the RF signals of PUs otherwise, all in the same licensed spectrum concurrently with PUs. Thus, the transmit power of each SU is jointly constrained by the peak interference power at PU and the harvested energy of SU. With the formulated green coexistence paradigm, we investigate the sum-throughput maximization problem with respect to time and power allocation, which is non-convex. To obtain the optimal resource allocation, we propose a joint optimal time and power allocation (JOTPA) algorithm that first transforms the original problem into a convex optimization problem with respect to time and energy allocation, and then solve it by iterative Lagrange dual decomposition. To comprehensively evaluate the performance of the GCRN with JOTPA, we deploy the GCRN in three typical scenarios and compare JOTPA with the equal time and optimal power allocation (ETOPA) algorithm. Extensive simulations show that the deployment of the GCRN significantly influences the throughput performance and JOTPA outperforms ETOPA under all considered scenarios.

This paper studies a green paradigm for the underlay coexistence of primary users (PUs) and secondary users (SUs) in energy harvesting cognitive radio networks (EH-CRNs), wherein battery-free SUs capture both the spectrum and the energy of PUs to enhance spectrum efficiency and green energy utilization. To lower the transmit powers of SUs, we employ multi-hop transmission with time division multiple access, by which SUs first harvest energy from the RF signals of PUs, and then, transmit data in the allocated time concurrently with PUs, all in the licensed spectrum. In this way, the available transmit energy of each SU mainly depends on the harvested energy before the turn to transmit, namely energy causality. Meanwhile, the transmit powers of SUs must be strictly controlled to protect PUs from harmful interference. Thus, subject to the energy causality constraint and the interference power constraint, we study the end-to-end throughput maximization problem for optimal time and power allocation. To solve this nonconvex problem, we first equivalently transform it into a convex optimization problem and then propose the joint optimal time and power allocation (JOTPA) algorithm that iteratively solves a series of feasibility problems until convergence. Extensive simulations evaluate the performance of EH-CRNs with JOTPA in three typical deployment scenarios and validate the superiority of JOTPA by making comparisons with two other resource allocation algorithms.