A5101453506- IoT and Edge/Fog Computing
- Industrial Vision Systems and Defect Detection
- Advanced Optimization Algorithms Research
- Metaheuristic Optimization Algorithms Research
- Advanced Graph Neural Networks
- Glass properties and applications
- Integrated Circuits and Semiconductor Failure Analysis
- Energy Efficient Wireless Sensor Networks
- VLSI and Analog Circuit Testing
- Network Security and Intrusion Detection
- Opportunistic and Delay-Tolerant Networks
- Photonic and Optical Devices
- Advanced Neural Network Applications
- Traffic Prediction and Management Techniques
- Polynomial and algebraic computation
- Advanced MIMO Systems Optimization
- Advanced Image and Video Retrieval Techniques
- Optical Network Technologies
- Energy Load and Power Forecasting
- Recommender Systems and Techniques
- Wireless Networks and Protocols
- Digital and Cyber Forensics
- Structural Engineering and Vibration Analysis
- Cloud Data Security Solutions
- Advanced Computing and Algorithms
Shenzhen Institutes of Advanced Technology
2022-2025
University of Chinese Academy of Sciences
2022-2025
Jinan University
2014-2024
State Grid Corporation of China (China)
2024
Data Assurance and Communication Security
2024
Chinese Academy of Sciences
2022
Harbin Institute of Technology
2021
Guangdong University Of Finances and Economics
2021
Beijing Jiaotong University
2019-2020
University of Jinan
2009
Graph Neural Networks (GNNs) have significant advantages in handling non-Euclidean data and been widely applied across various areas, thus receiving increasing attention recent years. The framework of GNN models mainly includes the information propagation phase aggregation phase, treating nodes edges as entities channels, respectively. However, most existing face challenge disconnection between node edge feature information, these typically treat learning features independent tasks. To...
This letter investigates the proactive eavesdropping for multiple suspicious links either through interfering or assisting links. Considering power constraint at eavesdropper, our objective is to maximize weighted sum rate of via jointly optimizing their intervention strategies (jamming relaying) and corresponding transmit eavesdropper. The formulated problem shown be a mixed-integer nonlinear programming (MINLP) problem, which NP-hard in general. By identifying separable structure we...
With the rapid development of wireless communication, traditional cloud computing cannot fully support low-latency services, especially in sensor networks. Mobile edge (MEC) can improve quality experience end users and save energy consumption mobile devices by providing resources storage space. However, it may cause discontinuity services if these roam around different MEC servers' areas. To solve aforementioned problem, this article, we propose an effective edge-intelligent service...
The continuously increasing number of connected smart devices has led to the emergence a crucial fault detection challenge Internet Things (IoT). In this study, we aim identify method for effective faults in IoT devices. An network model is first established, and data edge verification mechanism based on blockchain proposed; used ensure that cannot be tampered with, their accuracy verified using edge. Finally, set weighted random forest particle swarm optimization proposed. simulation...
Automated tuning can significantly improve productivity and save the costs of manual operation in microwave filter manufacturing industry. This article proposes a mathematical model scattering data optimization to find accurate coupling matrix for multiple-version filters, core step automated tuning. For large-scale problem coefficient combination, we propose decision set decomposition strategy that evenly divides entire frequency interval into several subintervals according correlation...
The data in industrial informatics may be high-dimensional and mislabeled. Irrelevant or noisy features pose a significant challenge to the detection of mislabeling. traditional method usually adopts two-step solution, first finding relevant subspace then using it for mislabeling detection. This struggles provide optimal performance, since separates procedures feature selection label error To solve this problem, article, we integrate two steps propose sequential ensemble noise filter (SENF)....
With the development of 5G technology and Internet Things, all kinds real life data are collected recorded by a large number sensors. It is great significance to mine analyze hidden information in for applications like future prediction. However, due interferences or instability collection equipment, sensory often incomplete, this incompleteness hinders in-depth analysis cloud. Therefore, processing around missing values significant. Relying on cloud machine learning methods not enough deal...
The social relationships among users can be effectively represented using graph structures, which has led to increasing interests in utilizing neural networks (GNNs) for recommendation.However, there are still some inevitable issues the existing methods: 1) problem of sparse supervision signals GNNbased recommendation models not been well addressed.2) methods often neglect guiding effect auxiliary behaviors on target behaviors, where only single behavior data used model training.3) In...
As an emerging multi-submersible system, Human Occupied Vehicle (HOV) under a convoy of set Autonomous Underwater Vehicles (AUVs) is regarded as the future framework for underwater exploration. In this work, to improve interoperability and communication efficiency formations, we treat system paradigm Internet (IoV) show how utilize Software-Defined Networking (SDN) technique optimize architecture. With assistance SDN, consider ocean current factors propose artificial flow potential field...