A5056303769- Indoor and Outdoor Localization Technologies
- Building Energy and Comfort Optimization
- Underwater Vehicles and Communication Systems
- Surface and Thin Film Phenomena
- Semiconductor materials and interfaces
- Copper Interconnects and Reliability
- Sparse and Compressive Sensing Techniques
- Blockchain Technology Applications and Security
- Semiconductor materials and devices
- Advanced Sensor and Energy Harvesting Materials
- Smart Grid Energy Management
- Urban Heat Island Mitigation
- Network Security and Intrusion Detection
- Cryptography and Data Security
- Electric Vehicles and Infrastructure
- IoT and Edge/Fog Computing
- Context-Aware Activity Recognition Systems
- Speech and Audio Processing
- Privacy-Preserving Technologies in Data
- Energy Efficient Wireless Sensor Networks
- nanoparticles nucleation surface interactions
- Molecular Junctions and Nanostructures
- IoT-based Smart Home Systems
- Physics of Superconductivity and Magnetism
- Image and Signal Denoising Methods
Guangzhou University
2015-2025
Chongqing Medical University
2023-2024
Children's Hospital of Chongqing Medical University
2023
University of Aizu
2018-2020
Guilin University of Electronic Technology
2019-2020
Sun Yat-sen University
2019
Institute of New Materials
2015
Beijing University of Technology
2013
Shanxi Normal University
2013
IBM (United States)
1982-1988
Device-free localization (DFL), as an emerging technology that locates targets without any attached devices via wireless sensor networks, has spawned extensive applications in the Internet of Things (IoT) field. For DFL, a key problem is how to extract significant features characterize raw signals with different patterns associated locations. To address this problem, paper, DFL formulated image classification problem. Moreover, we design three-layer convolutional autoencoder (CAE) neural...
A Virtual Power Plant (VPP) is a network of distributed power generating units, flexible consumers, and storage systems. VPP balances the load on grid by allocating generated different linked units during periods peak load. Demand-side energy equipment, such as Electric Vehicles (EVs) mobile robots, can also balance supply-demand when effectively deployed. However, fluctuation various makes supply challenging goal. Moreover, communication security between aggregator end facilities critical...
Metaverse and blockchain, as the latest buzzwords, have attracted great attention from industry academia. They will inevitably promote technological innovation in field of building information modeling (BIM) future. BIM organizes various into a whole by establishing virtual three-dimensional model architectural engineering using digital technology. The metaverse seamlessly integrates real world world, conducts rich activities such creation, display, trading. Therefore, through exploration...
Due to mobile Internet technology's rapid popularization, the Industrial of Things (IIoT) can be seen everywhere in our daily lives. While IIoT brings us much convenience, a series security and scalability issues related permission operations rise surface during device communications. Hence, at present, reliable dynamic access control management system for is urgent need. Up till now, numerous architectures have been proposed IIoT. However, owing centralized models heterogeneous devices,...
Fault detection is a fundamental requirement for Industrial Internet of Things (IIoT), such as the process industry. This article first reviews recent studies focusing on applying fault techniques to IIoT networks. However, we find that numerous focus resource utilization and workload allocation. The toward facilities still in its immature stage because existing approaches are not accurate enough stringent To this end, present novel algorithm, named Gaussian Bernoulli restricted Boltzmann...
In recent years, federated learning has attracted more and attention as it could collaboratively train a global model without gathering the users' raw data. It brought many challenges. this paper, we proposed layer-based system with privacy preservation. We successfully reduced communication cost by selecting several layers of to upload for averaging enhanced protection applying local differential privacy. evaluated our in non independently identically distributed scenario on three datasets....
Medical image processing plays an important role in the interaction of real world and metaverse for healthcare. Self-supervised denoising based on sparse coding methods, without any prerequisite large-scale training samples, has been attracting extensive attention medical processing. Whereas, existing self-supervised methods suffer from poor performance low efficiency. In this paper, to achieve state-of-the-art one hand, we present a method, named weighted iterative shrinkage thresholding...
Impurities and structural defects have a strong influence on the kinetics of thin film reactions. We investigated effect Cu formation microstructure in temperature rangeof 375°–450°C. found that presence 1 weight percent Al changes both activation energy pre‐exponential factor growth law. At same time, influences growing phase smoothens reaction interface. The results are discussed terms possible diffusion mechanisms .
To build the mechanism of a SelfDN is becoming an emerging direction for future IoT. The detection device status, such as fault or normal, very fundamental module in SelfDN. In this article, we first review recent studies devoted to applying techniques IoT networks. Taking challenge processing real-valued data into account, propose novel architecture Under architecture, present algorithm, named GBRBM-based deep neural network with auto-encoder (i.e., GBRBM-DAE) transform problem...
In the era of Internet Things (IoT) and AI-driven smart environments, human behavior recognition has emerged as a pivotal technology underpinning broad spectrum intelligent applications. However, achieving high accuracy while preserving user privacy remains critical challenge. To tackle this problem, paper introduces novel privacy-preserving method for WiFi-based recognition. It employs three-dimensional convolutional neural networks(3D-CNNs) enhanced with an attention-enabled autoencoder...