South China University of Technology

Intelligent reflecting surface (IRS)-assisted wireless communication has been recognized as an important way to enhance the security of unmanned aerial vehicle (UAV) networks. However, a single IRS may be unable to meet the transmission requirements in complex communication scenarios. In particular, due to the inherent instability of UAV platforms, the inevitable jitter caused by airflow and body vibration can have a great impact on transmission quality. In this paper, we study the multi-aerial IRS (AIRS) assisted secure simultaneous wireless information and power transfer (SWIPT) system with UAV jitter taken into account. For the purpose of exposition, two IRSs are deployed on two UAVs to reflect signals transmitted from the base station to an information user and an energy user; meanwhile an eavesdropper intends to eavesdrop on their messages. Angle estimation errors due to UAV jitter is transformed into the bounded channel state information (CSI) errors by applying linear approximations, and a joint optimization problem of the beamforming vector, AIRS phase shift matrices, and UAV trajectories is formulated to maximize the average secrecy rate (ASR). Since the problem is non-convex and the variables are strongly coupled, we propose an alternating optimization (AO) algorithm to deal with it. We decompose it into three sub-problems and adopt the Schur Complement, General S-Procedure, penalty dual decomposition (PDD), and successive convex approximation (SCA) methods to solve these non-convex sub-problems successfully. Numerical results show that UAV jitter could lead to system performance loss and demonstrate the performance gains of our proposed robust algorithm over other benchmark schemes.

Amorphous carbons are promising candidates as the anode materials for potassium-ion hybrid capacitors (PIHCs). The insufficient storage sites and inferior diffusion kinetics limit their potassium-ion storage capability. Edge nitrogen and morphology engineering are effective pathways to construct accessible active sites and enhanced diffusion kinetics. However, the organic integration of both pathways in amorphous carbon is still challenging. Herein, a “twice-cooking” strategy, including two-step carbonization processes at 700 °C, is designed to synthesize edge-nitrogen-rich lignin-derived carbon nanosheet framework (EN-LCNF). In the first-step carbonization process, the staged gas releases of CO and CO 2 from CaC 2 O 4 decomposition exfoliate the carbon matrix into a carbon nanosheet framework. In the second-step carbonization process, the generated CaO reacts with the cyanamide units of graphitic carbon nitride (g-C 3 N 4 ) to form an edge-nitrogen-rich framework, which is then integrated into the meso-/macropores of carbon nanosheet framework through sp ³ -hybridized C–N bonds. EN-LCNF with a high edge-nitrogen level of 7.0 at.% delivers an excellent capacity of 310.3 mAh g ⁻¹ at 50 mA g ⁻¹ , a robust rate capability of 126.4 mAh g ⁻¹ at 5000 mA g ⁻¹ , and long cycle life. The as-assembled PIHCs based on EN-LCNF anode and commercial activated carbon cathode show a high energy density of 110.8 Wh kg ⁻¹ at a power density of 100 W kg ⁻¹ and excellent capacitance retention of 98.7% after 6000 cycles. This work provides a general strategy for the synthesis of edge-nitrogen-rich lignin-derived carbon materials for advanced potassium-ion storage. Graphical Abstract

Evolutionary multitasking(EMT), which conducts evolutionary research on multiple tasks simultaneously, is an emerging research topic in the computation intelligence community. It aims to enhance the convergence characteristics by simultaneously conducting evolutionary research on multiple tasks, thereby facilitating knowledge transfer among tasks and achieving exceptional performance in solution quality. However, most of the existing EMT algorithms still suffer from the high computational burden especially when the number of tasks is large. To address this issue, this paper proposes a GPU-based multitasking evolutionary framework, which is able to handle thousands of tasks that arrive asynchronous in a short time. Besides, a concurrent multi-island mechanism is proposed to enable the parallel EMT algorithm to efficiently solve high-dimensional problems. Experimental results on eight problems with differing characteristics have demonstrated that the proposed framework is effective in solving high-dimensional problems and can significantly reduce the search time.

In this article, a proactive control strategy is developed for robots interacting with humans by integrating the estimation of human partner’s motion intention. A human control model is used and a least square-based observer is employed to estimate the human control input without a force sensor. Using the estimate of the human intention, a neural network (NN)-enhanced robot controller is designed to make the robot actively follow the human trajectory. NNs are integrated into robot controller to approximate and compensate the system uncertainties, so that a tracking performance can be guaranteed. Rigorous analysis based on Lyapunov theory proves that all the error signals are uniformly ultimately bounded. Implementations show that the proposed control method has adaptive properties.

Green Mobile Edge Networks (GMENs) are emerging networks that harvest green energy for powering mobile edge nodes, thereby reducing carbon dioxide emissions and energy costs. In GMENs, network service providers can flexibly place multiple virtual network functions (VNFs) that form a service function chain (SFC) in a specific order on geographically distributed edge nodes based on the level of harvested green energy, providing customized and sustainable network services for users. To meet the diversified availability requirements of users, backup SFCs need to be provided in addition to the primary SFC. These backup SFCs can be activated for providing uninterrupted services when the primary SFC is unavailable. However, due to the dynamic nature of wireless communication links, the uncertainty and unpredictability of green energy, and the limited resources available at edge nodes, optimizing the VNF placement and route traffic in real-time is challenging to minimize energy costs of all nodes and form expected SFCs with higher availability than user demand value. In this paper, the above problem is first formulated as an integer nonlinear programming and proven to be NP-hard. Then, it is discretized into a sequence of one-slot optimization problems to handle real-time changes in green energy and link availability. Finally, an online approximation strategy with a constant approximation ratio is proposed to solve the one-slot problems in polynomial time. This is the first study into online link availability-aware VNF placement and traffic routing problems in GMENs, motivated by sustainability concerns. The evaluation results indicate that the proposed scheme can ensure service availability while reducing the energy costs of all edge nodes and has achieved better performance when compared with other state-of-the-art methods.

To reduce switching losses of very-high-frequency (VHF) DC-DC converters, the quasi-resonance method is popular for achieving soft-switching. Yet, the widely adopted Class E and Class $\Phi _{2}$ circuits may result in high voltage stress on switching devices, making devices selection difficult. Focusing on this issue, the zero-voltage-transition (ZVT) technique is introduced into VHF DC-DC converters and a novel ZVT clamped VHF DC-DC topology is proposed in this paper. It consists of two boost units. And the drive signals of both MOSFETs are complementary and have a duty cycle of 0.5. And by appropriate design considerations, the parasitic parameters of both MOSFETs and both diodes are absorbed in the circuit. Moreover, the resonance operation of this circuit only occurs before and after the switching action. And thanks to the clamping effect of the diodes, the voltage stress on all switching devices is significantly limited. Finally, an experimental prototype with operating frequency of 10 MHz, 9 V input and 21 V/20 W output, is built to verify the performances of the proposed converter, which has a higher measurement efficiency of 88%.

Drawing inspiration from the mechanism of human skill acquisition, imitation learning has demonstrated remarkable performance. Over recent years, modelbased imitation learning combined with machine learning and control theory has been continuously developed and adapted to unstructured environments. However, most results for dual-arm tasks focus on relatively safe and stable environments, which still lack robustness to generalize skills. In this article, we propose a novel robust imitation learning framework for dual-arm object-moving tasks. During demonstration, we present a shared teleoperation strategy that actively assists the operator in remotely executing dual-arm tasks, aiming to reduce the operational difficulty and stress. During modeling and generalization, we propose a coupled linear parameter-varying dynamical system (CLPV-DS), which possesses the ability to protect and restore states against possible disturbances in the environment while maintaining good tracking accuracy and stability. To address the risk of box slipping caused by disturbances, we further introduce amutual following strategy, enabling the arms to compliantly follow each other while maintaining appropriate contact force. Considering potential obstacles in a complex generalization environment, we introduce a reactive obstacle avoidance strategy in real time that ensures global asymptotic stability. Finally, we verified the effectiveness of the proposed framework through comprehensive testing in both 2-D simulations and real-robot experiments.

Combinatorial optimization problems (COPs) are prevalent in various domains and present formidable challenges for modern computers. Searching for the ground state of the Ising model emerges as a promising approach to solve these problems. Recent studies have proposed some annealing processing architectures based on the Ising model, aimed at accelerating the solution of COPs. However, most of them suffer from low solution accuracy and inefficient parallel processing. This article presents a novel parallel tempering processing architecture (PTPA) based on the fully-connected Ising model to address these issues. The proposed modified parallel tempering algorithm supports multi-spin concurrent updates per replica and employs an efficient multi-replica swap scheme, with fast speed and high accuracy. Furthermore, an independent pipelined spin update architecture is designed for each replica, which supports replica scalability while enabling efficient parallel processing. The PTPA prototype is implemented on FPGA with 8 replicas, each with 1,024 fully-connected spins. It supports up to 64 spins for concurrent updates per replica and operates at 200 MHz. Different concurrency strategies are considered to further improve the efficiency of solving COPs. In the test of various G-set problems, PTPA achieves 3.2× faster solution speed along with 0.27% better average cut accuracy compared to a state-of-the-art FPGA-based Ising machine.

Fabric metaverse employs intelligence fibers embedded with flexible sensors to unknowingly gather and transmit massive hypermodal data around humans to a deep neural network-based metaverse inference service (DMS) for continual and real-time analysis. Each DMS has one primary branch and multiple side branches that allow early termination of service with differential accuracy and energy consumption. However, the continual provisioning of compute-intensive DMS with varying requirements for service model, accuracy, delay, and reliability poses a challenge for edge servers characterized by restricted computing resources and intermittent green energy. In this paper, we focus on a continual individualized DMS provisioning problem in the fabric metaverse consisting of a side branch insertion subproblem and a server activation and service deployment subproblem, and formulate them as Integer linear Programming and Markov Decision Process, respectively. Then, we propose a green continual inference (GCI) system, where a pruner with provable approximation ratios trims superfluous branches of every model to the given number K to minimize total overflow accuracy between accuracy demands and reserved branches assigned to users. Based on this exit result, each DMS is further divided into several blocks with dependencies to exploit constrained resources of computing and energy in a fine-grained manner. Finally, a learning-based scheduler is merged into GCI to maximize request throughput while minimizing the activation number of edge servers on different demand scenarios, by adaptively activating suitable servers and deploying required blocks and their corresponding backups on selected servers. Theoretical analyses, simulations, and experiments demonstrate that the GCI is promising compared with baseline algorithms.

This article investigates the fixed-time attitude control problem of quadrotor unmanned aerial vehicles (QUAVs) subject to stochastic disturbances. A fixed-time stability criterion for stochastic systems, which can compress the upper bound of the convergence time compared with the fixed-time stable Lyapunov theorem commonly used in the existing works, is proposed and theoretically proofed. The impact of stochastic disturbances on a QUAV is considered and a stochastic model of the attitude system of a QUAV is established. Based on the proposed criterion, a novel fixed-time attitude tracking controller for QUAV is designed, which can theoretically ensure the stability of attitude tracking within a fixed-time frame. To demonstrate the effectiveness of the controller, a simulation example is presented. A simulation comparison of different methods is also conducted. By artificially adding stochastic disturbances to the attitude angular velocity signals from sensors, an experiment is carried out on the three-DoF QUAV experimental platform. The experimental results show the feasibility of the designed control method in practical application.

Federated class-incremental learning (FCIL) allows multiple clients in a distributed environment to learn models collaboratively from evolving data streams, where new classes arrive continually at each client. Some existing works in FCIL combine traditional federated learning methods with class-incremental methods. However, the global model affected by data heterogeneity can aggravate local forgetting through the direct combination of traditional methods. To tackle this issue, we propose FCIDF, a novel Federated Class-Incremental learning approach based on Dynamic feature extractor Fusion . FCIDF learns personalized and incremental models for each client by introducing personalized fusion rates to integrate global knowledge into local features. Leveraging meta-learning during each incremental round, FCIDF ensures involvement of both old and new task knowledge in personalized training. Besides, we further propose a new Storing strategy based on Accumulated Global Feature Means (AGFMS), which helps the model review unbiased old knowledge and compensates for local forgetting. Experiment results show that FCIDF outperforms the baseline methods in both accuracy and forgetting on most settings, and AGFMS improves the performance of FCIDF on most evaluated scales.

Imbalanced data poses a substantial challenge to conventional classification methods, which often disproportionately favor samples from the majority class. To mitigate this issue, various oversampling techniques have been deployed, but opportunities for optimizing data distributions remain underexplored. By exploiting the ability of metric learning to refine the sample distribution, we propose a novel approach, Imbalance Large Margin Nearest Neighbor (ILMNN). Initially, ILMNN is applied to establish a latent feature space, pulling intra-class samples closer and distancing inter-class samples, thereby amplifying the efficacy of oversampling techniques. Subsequently, we allocate varying weights to samples contingent upon their local distribution and relative class frequency, thereby equalizing contributions from minority and majority class samples. Lastly, we employ Kullback-Leibler (KL) divergence as a safeguard to maintain distributional similarity to the original dataset, mitigating severe intra-class imbalances. Comparative experiments on various class-imbalanced datasets verify that our ILMNN approach yields superior results.

Edge-enabled Vehicular Metaverse (EVM) is a new paradise supported by various compute-intensive Virtual Vehicle Services (VVSs), where users can immerse and enjoy their spiritual world. User immersion is critical during VVS provisioning in the EVM, yet it can be weakened or curtailed by a sense of disengagement caused by unknown failures. Providing redundant backups VVSs (BVVSs) and keeping the Age of Backup Information (AoBI) could effectively resist and avoid this disengagement when failures occur. However, the trajectories of mobile vehicles are unknown and dynamic, which makes it challenging to optimally migrate VVSs and BVVSs or adjust the update frequency of backup information in real-time, so as to ensure service reliability and AoBI while minimizing the cost of accepting VVS-based metaverse services. In this paper, the above long-term issue is first decomposed into discrete single-slot sub-problems that are modeled as integer linear programming problems. Then, a comprehensive resource explorer named spotlighter is designed, where the first and second parts are a metaverse service home prediction algorithm based on deep learning and a VVS migration algorithm based on randomized rounding, respectively. By tracking the dynamical locations of service homes based on current and historical information, the former can help the latter to adaptively minimize migration costs on VVS re-instantiation and traffic transmission among services and moving vehicles. Finally, a cost-adaptive AoBI guarantee algorithm is merged in spotlighter to ensure the freshness of backup status, by trading-off synchronization cost on BVVS migration, backup update, and backup synchronization. Theoretical analyses and experiments based on real databases show that our algorithms are promising compared with baseline algorithms.

This paper investigates the implementation of Federated Learning (FL) in an over-the-air computation system with volatile clients, where each client operates under a limited energy budget and may unexpectedly drop out during local training sessions. The dropout of clients not only wastes energy but also diminishes their participation frequency, necessitating careful client selection by the server in each communication round. However, the diversity of training tasks and the random nature of client dropout present challenges such as the absence of an explicit objective function and the unavailability of client performance metrics. To address these challenges, we first analyze the convergence of the over-the-air federated learning system with volatile clients to identify the key factor influencing the model's convergence speed. Building upon this analysis, we propose an approximation of the objective function as the optimization goal for client selection. To mitigate energy waste, we introduce a dynamic client selection strategy termed DCSE, based on Exp3 with multiple plays and energy constraints, aiming to reconcile the dilemma of unknown local training states and limited resource constraints. Theoretical analysis demonstrates that our proposed solution maintains a constant bound on the difference from the optimal solution, affirming its theoretical feasibility. Furthermore, experimental results validate the effectiveness of the proposed strategy in enhancing FL by accelerating convergence speed, improving test accuracy, and reducing wasted energy.

Due to the advantages of fast training speed and competitive performance, Broad Learning System (BLS) has been widely used for classification tasks across various domains. However, the random weight generation mechanism in BLS makes the model unstable, and the performance of BLS may be limited when dealing with some complex datasets. On the other hand, the instability of BLS brings diversity to ensemble learning, and ensemble methods can also reduce the variance and bias of the single BLS. Therefore, we propose an ensemble learning algorithm based on BLS, which includes three modules. To improve the stability and generalization ability of BLS, we utilize BLS as the base classifier in an AdaBoost framework first. Taking advantage of the incremental learning mechanism of BLS, we then propose a selective ensemble method to raise the accuracy and diversity of the BLS ensemble method. In addition, based on the former selective Adaboost framework, we suggest a hierarchical ensemble algorithm, which combines sample and feature dimensions to further improve the fitting ability of the ensemble BLS. Extensive experiments have demonstrated that the proposed method performs better than the original BLS and other state-of-the-art models, proving the effectiveness and versatility of our proposed approaches.

The degradation of AC filter (ACF) arresters in converter stations can be accelerated by the frequent occurrence of unipolar polarity impulse currents during energization of ACFs. However, the present approaches for suppressing frequent unipolar polarity impulse currents are costly. In this paper, accelerated degradation tests with positive and bipolar polarity impulse currents are carried out on zinc oxide (ZnO) varistors to investigate the discrepancy in the degradation characteristics between these impulse currents, and the mechanism is briefly discussed. ZnO varistors have a slower degradation rate and better stability under bipolar polarity impulse currents. Moreover, the polarity and waveform features of the impulse current on ACF arresters during energization of ACFs are determined through PSCAD/EMTDC simulations. Combined with the monitored statistics, the test results and simulation results, a theoretically economical approach for improving the operating lifespan of ACF arresters in a converter station already in operation is proposed.

Federated learning (FL) is a distributed machine learning paradigm that can be organized in two layers. In the outer layer of users, there is a model interaction process between the task publisher and users, through which all parties obtain their respective utilities. However, these parties' utilities are coupled, both depending on the training sample size and local iterations. In the inner layer of users, a user' multiple devices (e.g., computers and smart phones) can be used to jointly train local models efficiently. Yet, due to device heterogeneity, it is challenging for users to determine which devices to participate in local training and allocate how many computing and communication resources to minimize training costs. In this paper, we tackle this novel two-layer optimization problem by designing utility game and resource control strategies. In the outer layer, we model the relationship between the task publisher and users as a Stackelberg game and obtain the optimal solution for both parties by solving a unique Stackelberg equilibrium point; while in the inner layer, we formulate the optimization problem as a mixed integer nonlinear programming problem, which is decomposed into sub-problems and solved by devising resource control algorithm based on successive convex approximation. Finally, extensive experiments show that the proposed algorithms outperform baseline algorithms.

Recently, rule-based classification on multivariate time series (MTS) data has gained lots of attention, which could improve the interpretability of classification. However, state-of-the-art approaches suffer from three major issues. 1) few existing studies consider temporal relations among features in a rule, which could not adequately express the essential characteristics of MTS data. 2) due to the concept drift and time warping of MTS data, traditional methods could not mine essential characteristics of MTS data. 3) existing online learning algorithms could not effectively update shapelet-based temporal association rules of MTS data due to its temporal relationships among features of different variables. To handle these issues, we propose an online learning method for temporal association rule on dynamically collected MTS data (OTARL). First, a new type of rule named temporal association rule is defined and mined to represent temporal relationships among features in a rule. Second, an online learning mechanism with a probability correlation-based evaluation criterion is proposed to realize the online learning of temporal association rules on dynamically collected MTS data. Finally, an ensemble classification approach based on maximum-likelihood estimation is advanced to further enhance the classification performance. We conduct experiments on ten real-world datasets to verify the effectiveness and efficiency of our approach

The pipeline safety warning system (PSEW) is an important guarantee for the safe transportation of energy pipelines. Given the constraints of deploying detection models at resource-limited pipeline stations, there is a compelling need to develop efficient, lightweight models suitable for edge device applications. This brief introduces an adaptive heterogeneous model knowledge distillation network (AHKDnet) for edge deployment of pipeline network detection models. The global information and long-distance dependency relationships from the ViT-based teacher network are transferred to the CNN-based shallow student network. We introduce the learnable modulation parameters to optimize target information enhancement, reducing the impact of irrelevant information. By embedding the model selection at each stage of knowledge distillation, the performance collapse of student models caused by misleading cross-architecture knowledge is avoided, and model convergence is accelerated. Experiments on three actual scene datasets of pipeline networks show that AHKDnet outperforms the state-of-the-art KD methods and has strong generalization ability. Notably, AHKDnet enhances the recognition performance of shallow student networks by an average of 10%, highlighting its efficacy and potential for practical applications. Our method can provide a new reference for edge deployment of PSEW.