A5101908707- Face and Expression Recognition
- IoT and Edge/Fog Computing
- Rough Sets and Fuzzy Logic
- Data Mining Algorithms and Applications
- Neural Networks and Applications
- Maritime Ports and Logistics
- Advanced Wireless Communication Technologies
- Data Management and Algorithms
- Maritime Navigation and Safety
- Advanced Decision-Making Techniques
- Maritime Security and History
- UAV Applications and Optimization
- Age of Information Optimization
- Image Processing and 3D Reconstruction
- Advanced MIMO Systems Optimization
- Advanced Clustering Algorithms Research
Dalian University
2023
Center for High Pressure Science & Technology Advanced Research
2019-2021
Driven by the advancement of science and technology, issue marine ship Internet Things (IoT) users' assignment computing tasks offloading has become more challenging. When faced with complex dynamic environment, considering different quality requirements maritime applications, we have addressed this in paper. We first propose space-air-ground-edge (SAGE) communication network architecture. This novel architecture is used to offload computing-intensive applications services for IoT users...
We consider the problem of intelligent and efficient task allocation mechanism in large-scale mobile edge computing (MEC), which can reduce delay energy consumption a parallel distributed optimization. In this paper, we study joint optimization model to cooperative management among terminals (MT), macro cell base station (MBS), multiple small (SBS) for MEC applications. propose multi-block Alternating Direction Method Multipliers (ADMM) based method both requirements low system formulates...
Missing values are a common phenomenon in real-world datasets, which caused by many factors such as errors data acquisition or storage, equipment failure, human fault storage. Incomplete modeling and missing imputation have become an increasingly important task. Since the regression relationship between attributes is usually different clusters, this paper proposes method called DS-TS-ALI model to incomplete that rely on clusters. The precise established for framework of Takagi-Sugeno (TS)...
Incomplete datasets with missing values increase the difficulty of data analysis. In this paper, incomplete are modeled based on Auto-Association Neural Network(AANN) and imputation. order to strengthen correlation among attributes samples, we propose an Attribute Mutual Associative Multi-Task Learning Model Based Confidence (AM-MTL). By optimizing transmission path nodes in output layer, construct MTL architecture which main imputation task is parallel secondary classification task, use...
Digital Historical material databases are important for historical research. This article analyzes the current situation and problems of database construction proposes a method based on blockchain technology to achieve secure storage protection digital data. The also sharing building mode architecture, which supports co-building encourages exchange materials.