Highly flexible Automated Guided Vehicles (AGVs) are interconnected via Industrial Wireless Control Networks (IWCNs) in Multi-access Edge Computing (MEC)-assisted smart factories. The MEC alleviates the lack of computational resources in AGV systems through task offloading. However, IWCNs with limited communication resources struggle to support the highly concurrent offloading of AGVs. In the distributed AGV systems with multi-MEC servers, AGV mobility leads to uneven distribution across MEC server areas, potentially resulting in severe competition for communication resources. Therefore, in this paper, we design a Transferable joint Task Offloading and Multi-Channel Access (T2OMCA) algorithm based on multi-agent deep reinforcement learning. Specifically, AGV observations are modelled as graphs, in which edge relationships are learned through Transformer. This enables AGVs to utilize domain information to collaborate and alleviate concurrent offloading. Moreover, the T2OMCA algorithm converts network input into fixed embeddings to accommodate varying numbers of AGVs. Finally, to encourage exploration in the high-dimensional action space, the T2OMCA algorithm introduces a noisy network and a prioritized experience replay mechanism. Extensive simulations show that the T2OMCA algorithm outperforms existing algorithms in terms of average completion rate, processing delay, and access conflict rate under time-varying AGV topologies.

Time-Critical Wireless Network (TCWN) is a promising communication technology that can satisfy the low latency, high reliability, and deterministic requirements of mission-critical applications. Multiple TCWNs required by various applications inevitably coexist with each other. Most existing works aim to achieve acceptable latency or consider the simplest topology (i.e., line topology). As latency requirements become more stringent, exploring the minimum data collection latency becomes an interesting problem. In this paper, the coexisting system consists of multiple tree-topology-based TCWNs. We first establish a conversion framework to convert an arbitrary tree topology into multiple analogous line topologies to reduce the analysis complexity. We then propose a Time-Critical wireless network Scheduling (TCS) algorithm to minimize the data collection latency of coexisting TCWNs. The TCS algorithm consists of two phases. In the internetwork scheduling phase, we strictly derive a general expression to characterize the practical network requirements. In the intranetwork scheduling phase, we design two levels of priority assignment algorithms to accurately characterize the critical states and resource requirements of different nodes. We conduct extensive simulations to verify the effectiveness of the TCS algorithm. The evaluation results show that the TCS algorithm can achieve minimum data collection latency in more than 99.956% cases, and the maximum difference compared to the optimal value is one time slot.

Unmanned Aerial Vehicles (UAVs) exhibit great advantages in data collection from ground sensors in vast tracts of fields. Due to their limited power supply, most works assume that the UAV simply traverses each sensor’s fixed transmission range to collect data, thereby shortening the flight path. However, they neglect the quality of collected data, which may deteriorate dramatically as the transmission distance increases. In this paper, by leveraging the physical-layer protocol – LoRa, we propose a Packet Reception Ratio (PRR)-based probabilistic coverage model to evaluate the quality of data transmission, which directly determines the data acquisition efficiency. On this basis, to minimize the energy consumption of UAV and sensors while ensuring high-quality data acquisition, we formulate the UAV trajectory planning as a joint Energy Consumption and data Acquisition Efficiency (ECAE) optimization problem. To tackle the ECAE problem, we propose a Deep Reinforcement Learning (DRL)-based two-stage scheme. First, an attention-based encoder-decoder model is trained to generate an initial trajectory. Then an intuitive optimization algorithm is devised to further explore the optimal trajectory. Evaluation results show that our scheme can reduce the total energy cost of UAV and sensors by 27.1% as compared to the best baseline’s policy while maintaining a promising PRR.

The Industrial Internet of Things (IIoTs) plays an important role in various industrial applications, which require multiple time-critical networks to be deployed in the same region. The limited communication resources inevitably incur network coexistence problems. For scenarios where coexisting networks cannot coordinate effectively, the centralized or partial-information-based decentralized resource allocation methods cannot be implemented. To address this concern, we propose a Cooperation-Free Reinforcement Learning (CF-RL) algorithm for the fully distributed resource allocation problem in coexisting IIoT systems. Each network adopts the proposed algorithm to minimize collisions through a trial-and-error approach without any information interaction. To resist the influence of environmental dynamics, each coexisting network learns the state transition probability of the resource block instead of the resource block's position. Moreover, to potentially ensure the overall system performance, each network additionally considers the period offset in the initialization phase and action selection phase, so that the coexisting networks have different preferences for different state transitions. We conduct extensive simulations to verify the convergence performance. Evaluation results show that the CF-RL algorithm almost achieves (more than 99.88%) the effect of centralized resource allocation and has obvious superiorities over other cooperation-free algorithms in terms of the convergence rate, the number of collisions, and the resource utilization ratio.

The popularization of unmanned aerial vehicles (UAVs) has boosted various civil applications such as traffic monitoring, in which the effective coordination of the UAV swarm plays a significant role in expanding the monitoring range and enhancing the execution efficiency. However, due to the isolated local environments as well as the heterogeneous execution capabilities, it is challenging to achieve highly consistent actions. In this article, we incorporate the federated learning framework with the imitation learning technique to coordinate the UAVs' maneuvers by interactively imitating the leader UAV's operations. During the interagent global model download phase, we utilize the generative adversarial imitation learning (GAIL) model to accurately follow the leader UAV's operations by removing the biased estimates of imitation parameters. While in the intraagent local model training phase, we utilize the self-imitation learning (SIL) model to correct delicate imitation errors by virtue of the follower UAVs' own historical valuable experiences. In order to achieve more efficient distributed parameter interactions, we regularize the federated gradient updates and eventually yield coordinated swarm policies. We evaluate the proposed algorithm in the UAV-based traffic monitoring scenario. Evaluation results demonstrate the superiorities on training and execution efficiencies.

The ongoing coronavirus disease 2019 (COVID-19) is a pandemic causing millions of deaths, devastating social and economic disruptions. Testing individuals for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the pathogen of COVID-19, is critical for mitigating and containing COVID-19. Many countries are implementing group testing strategies against COVID-19 to improve testing capacity and efficiency while saving required workloads and consumables. A group of individuals’ nasopharyngeal/oropharyngeal (NP/OP) swab samples is mixed to conduct one test. However, existing group testing methods neglect the fact that mixing samples usually leads to substantial dilution of viral ribonucleic acid (RNA) of SARS-CoV-2, which seriously impacts the sensitivity of tests. In this paper, we aim to screen individuals infected with COVID-19 with as few tests as possible, under the premise that the sensitivity of tests is high enough. To achieve this goal, we propose an Adaptive Group Testing (AdaGT) method. By collecting information on the number of positive and negative samples that have been identified during the screening process, the AdaGT method can estimate the ratio of positive samples in real-time. Based on this ratio, the AdaGT algorithm adjusts its testing strategy adaptively between an individual testing strategy and a group testing strategy. The group size of the group testing strategy is carefully selected to guarantee that the sensitivity of each test is higher than a predetermined threshold and that this group contains at most one positive sample on average. Theoretical performance analysis on the AdaGT algorithm is provided and then validated in experiments. Experimental results also show that the AdaGT algorithm outperforms existing methods in terms of efficiency and sensitivity.Note to Practitioners—Real-time reverse transcription-polymerase chain reaction (rRT-PCR) tests provide scope for automation and are one of the most widely used laboratory methods for detecting the SARS-CoV-2 virus. This paper is motivated by the following challenges: (1) Many countries are experiencing an acute shortage of professionals and consumables for conducting rRT-PCR tests; (2) Group sizes of existing group testing methods against COVID-19 may not be optimal, which adversely impacts the efficiency of the screening of the SARS-CoV-2 virus; (3) Existing group testing methods do not consider the fact that the sensitivity of rRT-PCR tests usually decreases with the group size. The objective of this paper is to improve the efficiency and sensitivity of large-scale screening against COVID-19. For achieving this goal, we propose an Adaptive Group Testing (AdaGT) algorithm, which has the following advantages: (1) It can improve the efficiency for screening the SARS-CoV-2 virus, mainly by adaptively adjusting its testing strategy between an individual testing strategy and a group testing strategy based upon an estimated ratio of positive samples during the screening process; (2) It can guarantee a high sensitivity of the rRT-PCR tests by determining the group sizes of the group testing strategy based upon some constraints; (3) We derive an appropriate threshold for the estimated ratio of positive samples such that the AdaGT algorithm can achieve a minimum average number of rRT-PCR tests and can be directly employed in practical applications.

The pathogen of the ongoing coronavirus disease 2019 (COVID-19) pandemic is a newly discovered virus called severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Testing individuals for SARS-CoV-2 plays a critical role in containing COVID-19. For saving medical personnel and consumables, many countries are implementing group testing against SARS-CoV-2. However, existing group testing methods have the following limitations: (1) The group size is determined without theoretical analysis, and hence is usually not optimal. This adversely impacts the screening efficiency. (2) These methods neglect the fact that mixing samples together usually leads to substantial dilution of the SARS-CoV-2 virus, which seriously impacts the sensitivity of tests. In this paper, we aim to screen individuals infected with COVID-19 with as few tests as possible, under the premise that the sensitivity of tests is high enough. We propose an eXpectation Maximization based Adaptive Group Testing (XMAGT) method. The basic idea is to adaptively adjust its testing strategy between a group testing strategy and an individual testing strategy such that the expected number of samples identified by a single test is larger. During the screening process, the XMAGT method can estimate the ratio of positive samples. With this ratio, the XMAGT method can determine a group size under which the group testing strategy can achieve a maximal expected number of negative samples and the sensitivity of tests is higher than a user-specified threshold. Experimental results show that the XMAGT method outperforms existing methods in terms of both efficiency and sensitivity.

Concordant operations among automatic industrial devices (i.e., agents) play a significant role in achieving efficient smart manufacturing. Existing multiagent cooperation methods focus on centralized training with decentralized execution, which is unsuitable for the edge-based distributed industrial scenarios. To harmonize distributed devices' actions and enhance the production efficiency, in this article, we propose an Interactive-Imitation-based Distributed Coordination (IIDC) algorithm. Specifically, we leverage the generative adversarial imitation learning (GAIL) model to direct one agent's actions by following external professional demonstrations. We also adopt the self-imitation learning (SIL) model to direct one agent's potential actions by following its own previous good experiences. As the imperfect demonstrations in the GAIL-based interagent imitation process may degrade the imitation accuracy, we present a confidence-based matching method to reduce the gap between the professional and imitative behaviors. Furthermore, during the SIL-based intra-agent imitation process, the rewardless explorations also lead to nonoptimal imitation policy. We then present a Stein variational policy gradient based self-imitation method to learn an expected optimal policy. We validate the IIDC algorithm's effectiveness via the sequential assembly task. Evaluation results demonstrate that the IIDC algorithm can enhance the production efficiency evidently.

Blockchain is multi-centralized, immutable and traceable, thus is very suitable for distributed storage, privacy and security management in IoTs. However, most existing researches focus on the integration of public blockchain and IoTs. In fact, problems such as slow consensus, low transmission throughput, and completely open storage on the public blockchain are intolerable in IoT scenarios. Although consortium blockchain represented by Hyperledger Fabric has improved the transmission rate, its data security completely relies on the PKI-based certificate mechanism, resulting in transmission inefficiency and privacy leakage. In this paper, a key-derived Controllable Lightweight Secure Certificateless Signature (CLS2) algorithm is proposed to significantly improve the transmission efficiency and keep similar computation overhead of consortium blockchain. Compared with the existing certificateless signatures, CLS2achieves more secure transactions, whose controllable anonymity and key-derived mechanism not only prevents public key replacement attacks and forged signature attacks, but also supports hierarchical privacy protection. Armed with CLS2, we design a consortium blockchain security architecture based on Hyper-ledger Fabric and edge computing. To the best of our knowledge, this is the first implementation of certificateless signature in consortium blockchain. We formally prove the security of our schemes in the random oracle model. Specifically, the security of the proposed scheme is reduced to the Elliptic curve discrete logarithm problem (ECDLP). Security analysis and experiments in IoT scenarios verify the feasibility and effectiveness of CLS2.

The limited spectrum resources inevitably incur the spectrum sharing among coexisting industrial wireless networks (IWNs), and multiple coexistence IWNs form a heterogeneous environment. An effective resource allocation thus plays a crucial role in coordinating the efficient operations of multiple IWNs. Existing works only study the constrained coexistence problem among specified types of networks with a limited number of nodes over one single channel. In this paper, we investigate a general coexistence problem over multiple channels among arbitrary types of networks with line topologies, and the number of nodes in each network is also arbitrary. We rigorously analyze theoretical scheduling latency of this general coexistence problem, then we propose an algorithm to attain the optimal result. The presented Coexisting Line topology Networks Resource Allocation (CLNRA) algorithm consists of two phases. In the inter-network resource allocation phase, non-overlapped channels are allocated to each network according to the corresponding transmission priority. While in the intra-network resource allocation phase, we filter out the nodes that may generate continuous empty buffers so as to enhance the resource utilization ratio. We also verify the effectiveness of the CLNRA algorithm through extensive simulations. Evaluation results show that the CLNRA algorithm can attain the theoretical optimal result in 99:3% cases, and it has obvious superiorities on resource utilization ratio and scheduling latency.

As it is difficult for power-restricted wireless sensor nodes to achieve rapid neighbor discovery under the scenarios of asynchronous clocks, misaligned time slots, and asymmetric duty-cycle (i.e., wake-up/sleep) scheduling periods, we propose a low-power neighbor discovery algorithm termed Pharos by alternately utilizing the fully and the partially awake time slots. The partially awake time slots of one node are certain to detect the counterpart's awake slots while reducing the power consumption as compared to the fully awake time slots. We analyze the theoretical neighbor discovery latency and derive the optimal parameters for both symmetric and asymmetric duty-cycle schedules. We also verify the effectiveness of the Pharos algorithm through extensive simulations. Evaluation results display that Pharos costs much less discovery latency and power than the state-of-the-art neighbor discovery algorithms.

In distributed Internet of Things (IoTs), channel hopping (CH) is an effective scheme for neighbor nodes to achieve blind rendezvous over common available channels and to establish communication links. When nodes are unaware of each other's local clocks and the global channels and have no pre-assigned CH strategies or identifiers (IDs), it is particularly challenging to guarantee blind rendezvous within a finite period of time, which has not been solved yet by using only one radio. In this paper, we propose a novel quaternary-encoding-based CH (QECH) algorithm to tackle the above issue. The QECH algorithm encodes a randomly selected channel into a quaternary string according to the 6B/8B encoding. We also append a common prefix string as well as the randomly selected channel before the quaternary string to guarantee overlaps in the asynchronous scenario. For all kinds of quaternary digits, we construct four mutually co-prime numbers to enumerate all possible combinations of the common available channels. We theoretically analyze the deterministic rendezvous principle and the upper bounded rendezvous latency of the QECH algorithm. We also verify the effectiveness of the QECH algorithm through extensive simulations. Evaluation results show the superiority of the QECH algorithm in terms of rendezvous latency.

Effective collaborations among autonomous unmanned aerial vehicles (UAVs) rely on timely information sharing. However, the time-varying flight environment and the intermittent link connectivity pose great challenges to message delivery. In this paper, we leverage the deep reinforcement learning (DRL) technique to address the UAVs' optimal links discovery and selection problem in uncertain environments. As the multi-agent learning efficiency is constrained by the high-dimensional and continuous action spaces, we slice the whole action spaces into a number of tractable fractions to achieve efficient convergences of optimal policies in continuous domains. Moreover, for the nonstationarity issue that particularly challenges the multi-agent DRL with local perceptions, we present a multi-agent mutual sampling method that jointly interacts the intra-agent and inter-agent state-action information to stabilize and expedite the training procedure. We evaluate the proposed algorithm on the UAVs' continuous network connection task. Results show that the associated UAVs can quickly select the optimal connected links, which facilitate the UAVs' teamwork significantly.

Neighbor discovery is a significant communication primitive for adjacent unmanned aerial vehicles (UAVs) to construct a flying ad hoc network (FANET). The multi-channel nature of FANETs makes channel hopping (CH) a feasible rendezvous method for UAVs to hop to the same available channel simultaneously and initiate a connection. However, due to the intrinsic uncoordinated constraints of dispersed UAVs (e.g., lack of clock synchronization, heterogeneous local channels, symmetric roles, and oblivious identifiers), it is challenging to design a performant CH algorithm that can achieve fast neighbor discovery in dynamic FANETs. In this paper, we present a fully uncoordinated matrix-based CH algorithm termed ABIO, which consists of one fixed Anchor column and several variable Binary (i.e., I/O-bit) extended columns in each CH period. The deterministic overlaps as well as the co-primality property of channel numbers among different kinds of columns provide the rendezvous guarantee. Furthermore, for the case with frequently varying channel status, we present a probability-based dynamic discovery (PDD) algorithm. By virtue of the cumulative probability estimation and selection of the qualified channels, the PDD algorithm can achieve timely rendezvous in the unstable environment with high probability. We rigorously analyze the theoretical neighbor discovery latency. We also validate the feasibility and efficiency of the proposed algorithms through extensive simulations. Evaluation results demonstrate the superiority of our algorithms in both stable and unstable communication environments.

Channel rendezvous is a prerequisite for secondary users (SUs) to set up communications in cognitive radio networks (CRNs). It is expected that the rendezvous can be achieved within a short finite time for delay-sensitive applications and over all available channels to increase the robustness to unstable channels. Some existing works suffer from a small number of rendezvous channels and can only guarantee rendezvous under the undesired requirements such as synchronous clock, homogeneous available channels, predetermined roles and explicit SUs' identifiers (IDs). In this paper, to address these limitations, we employ the notion of Disjoint Set Cover (DSC) and propose a DSC-based Rendezvous (DSCR) algorithm. We first present an approximation algorithm to construct one DSC. The variant permutations of elements in the ingeniously constructed DSC are then utilized to regulate the order of accessing channels, enabling SUs to rendezvous on all available channels within a short duration. We derive the theoretical maximum and expected rendezvous latency and prove the full rendezvous degree of the DSCR algorithm. Extensive simulations show that the DSCR algorithm can significantly reduce the rendezvous latency compared to existing algorithms.

Rendezvous is a vital step for secondary users who want to initiate a communication in cognitive radio networks. In this paper, we propose a novel Padded-Dyck-Path-based (PDP) rendezvous algorithm that generates channel hopping sequences utilizing global channels. PDP is designed according to the roundabout Dyck path so as to increase rendezvous opportunities. As the global channels may not be shared in distributed environments, we also propose a local PDP (L-PDP) heterogeneous rendezvous algorithm that generates channel hopping sequences utilizing only local available channels. L-PDP can significantly reduce rendezvous latency and allow for distributed implementations. We prove that both PDP and L-PDP can provide guaranteed rendezvous and derive their upper bounds of rendezvous latency. Analytical and simulation results show that PDP and L-PDP outperform existing algorithms in terms of time-to-rendezvous in global and local scenarios, respectively.