A5101531860- Energy Harvesting in Wireless Networks
- IoT and Edge/Fog Computing
- UAV Applications and Optimization
- Photonic and Optical Devices
- Advanced Battery Materials and Technologies
- Advanced Battery Technologies Research
- Advancements in Battery Materials
- Image and Signal Denoising Methods
- Wireless Power Transfer Systems
- Remote-Sensing Image Classification
- Privacy-Preserving Technologies in Data
- Advanced Neural Network Applications
- Formal Methods in Verification
- Distributed and Parallel Computing Systems
- Robotic Path Planning Algorithms
- Age of Information Optimization
- Mobile Crowdsensing and Crowdsourcing
- Advanced Image Fusion Techniques
- IoT Networks and Protocols
- Sparse and Compressive Sensing Techniques
- Advanced Fiber Optic Sensors
- Distributed Control Multi-Agent Systems
- AI-based Problem Solving and Planning
- Adversarial Robustness in Machine Learning
- Plasmonic and Surface Plasmon Research
Northeastern University
2025
China Mobile (China)
2025
Shanghai Jiao Tong University
2022-2024
Sun Yat-sen University
2023-2024
Zhangzhou Normal University
2021-2024
Goodyear (United Kingdom)
2023
Harbin Institute of Technology
2023
Shenzhen University
2023
Tianjin University of Technology and Education
2023
University of Electronic Science and Technology of China
2023
A fast and accurate capacity estimation method for lithium-ion batteries is developed. This applies our developed semi-empirical model to a discharge curve of battery the determination its maximum stored charge after each cycle. provides an state-of-health (SoH) with difference less than 2.22% when compared electrochemistry-based electrical (ECBE) SoH calculation. The parameters derived from can also be applied other cells in same pack 2.5% complex ECBE model, showing extendibility model....
<title>Abstract</title> To address the challenges of age and gender recognition in uncontrolled scenarios with facial absence or severe occlusion, this paper proposes a Spatial Correlation Guided Cross Scale Feature Fusion Network (SCGNet). The method integrates multi-granularity semantic features through Cross-Scale Combination (CSC) module, enhances local detail representation using Local (LFGF) designs Composition Analysis (SCCA) module based on Getis-Ord Gi* statistics for feature...
Unmanned aerial vehicles (UAVs) can be used to relay sensing information and computational workloads from ground users (GUs) a remote base station (RBS) for further processing. In this paper, we employ multiple UAVs assist with the collection of in terrestrial wireless sensor network. All collected by forwarded RBS. We aim improve energy efficiency sensing-data transmission optimizing UAV trajectory, scheduling, access-control strategies. Considering time-slotted frame structure, flight,...
Inaccurate state-of-health (SoH) estimation of battery can lead to over-discharge as the actual depth discharge will be deeper, or a more-than-necessary number charges calculated SoC underestimated, depending on whether inaccuracy in maximum stored charge is over under estimated. Both increased degradation battery. SoH also continuous use below 80% that could catastrophic failures. Therefore, an accurate and rapid on-line method for lithium ion batteries, different operating conditions such...
Ensuring safe, real-time motion planning in arbitrary environments requires a robotic manipulator to avoid collisions, obey joint limits, and account for uncertainties the mass inertia of objects robot itself. This paper proposes Autonomous Robust Manipulation via Optimization with Uncertainty-aware Reachability (ARMOUR), provably-safe, receding-horizon trajectory planner tracking controller framework manipulators address these challenges. ARMOUR first constructs robust that tracks desired...
Mobile edge computing (MEC) is viewed as a promising technology to address the challenges of intensive demands in hotspots (HSs). In this paper, we consider unmanned aerial vehicle (UAV)-assisted backscattering MEC system. The UAVs can fly from parking aprons HSs, providing energy HSs via RF beamforming and collecting data wireless users through backscattering. We aim maximize long-term utility all subject stability HSs' queues. This problem joint optimization offloading decision contract...
Synthetic aperture radar (SAR) images are inherently affected by speckle noise. Deep learning-based methods have shown good potential in image denoising task. Most deep learning for focus on additive Gaussian noise removal. However, SAR usually contaminated non-Gaussian multiplicative In this paper, we propose a novel unrolling network named SAR-DURNet to deal with the despeckling problem. We establish optimization problem of removal using priori distribution, which can be sovled...
In this paper, we propose a novel dictionary learning model based on the idea of deep unrolling to deal with synthetic aperture radar (SAR) image change detection problem. Deep aims at iterative algorithm into trainable neural network. our proposed method, is applied Iterative Shrinkage Threshold Algorithm (ISTA), which one classic algorithms for learning. Then, Unrolling (UR-ISTA), utilized obtain sparse codes difference results. Finally, map computed by k-means clustering algorithm. The...
In this paper, we consider a mobile edge computing (MEC) system with multiple unmanned aerial vehicles (UAVs) and stochastic energy harvesting. The UAVs' mobility can help data offloading over larger geographical area containing multi- hotspots (HSs). If HSs have requests, the dispatch agent (DA) recruit different types of UAVs to fly close computation. We aim maximize long-term utility all HSs, subject stability queue. proposed problem is joint optimization strategy contract design in...
随着大数据的普及和算力的提升，深度学习已成为一个热门研究领域，但其强大的性能过分依赖网络结构和参数设置。因此，如何在提高模型性能的同时降低模型的复杂度，关键在于模型优化。为了更加精简地描述优化问题，本文以有监督深度学习作为切入点，对其提升拟合能力和泛化能力的优化方法进行归纳分析。给出优化的基本公式并阐述其核心；其次，从拟合能力的角度将优化问题分解为3个优化方向，即收敛性、收敛速度和全局质量问题，并总结分析这3个优化方向中的具体方法与研究成果；从提升模型泛化能力的角度出发，分为数据预处理和模型参数限制两类对正则化方法的研究现状进行梳理；结合上述理论基础，以生成对抗网络（generative...
Mobile crowdsensing is considered as a promising technology to exploit the computing and sensing capabilities of decentralized wireless sensor nodes. Typically, quality information obtained from largely affected by various factors, such diverse requirements tasks, varying across different crowd workers, dynamic changes channels conditions environment. In this paper, considering dynamics’ we focus on spatial-temporal model aim maximize value at point interest, optimizing recruiting range time...