A5005367960- Optical Network Technologies
- Advanced Photonic Communication Systems
- Advanced MIMO Systems Optimization
- Advanced Wireless Communication Techniques
- Metaheuristic Optimization Algorithms Research
- Wireless Signal Modulation Classification
- Error Correcting Code Techniques
- Advanced Algorithms and Applications
- Millimeter-Wave Propagation and Modeling
- PAPR reduction in OFDM
- Service-Oriented Architecture and Web Services
- Artificial Immune Systems Applications
- Advanced Optical Network Technologies
- Wireless Communication Networks Research
- Advanced biosensing and bioanalysis techniques
- Advanced Graph Neural Networks
- Advanced Wireless Communication Technologies
- Software System Performance and Reliability
- Underwater Vehicles and Communication Systems
- Energy Harvesting in Wireless Networks
- Vehicular Ad Hoc Networks (VANETs)
- Indoor and Outdoor Localization Technologies
- Advanced Computational Techniques and Applications
- Advanced Fiber Laser Technologies
- Power Line Communications and Noise
Southeast University
2015-2025
University of Electronic Science and Technology of China
2015-2025
Zhejiang University
2005-2025
Stomatology Hospital
2017-2025
North China University of Science and Technology
2024
Zhengzhou University of Light Industry
2018-2023
Guizhou University
2018-2023
Shanghai Electric (China)
2023
Guilin University of Technology
2010-2022
Ministry of Natural Resources
2022
Graph representation learning has emerged as a powerful technique for addressing real-world problems. Various downstream graph tasks have benefited from its recent developments, such node classification, similarity search, and classification. However, prior arts on focus domain specific problems train dedicated model each dataset, which is usually non-transferable to out-of-domain data. Inspired by the advances in pre-training natural language processing computer vision, we design...
In this paper, we propose a distributed joint computation offloading and resource allocation optimization (JCORAO) scheme in heterogeneous networks with mobile edge computing. An problem is formulated to provide the optimal strategy policy, uplink subchannel allocation, transmission power scheduling. The decomposed into two sub-problems due NP-hard property. order analyze strategy, sub-algorithm named potential game built. existence of Nash equilibrium proved. To jointly allocate subchannel,...
GNN-based anomaly detection has recently attracted considerable attention. Existing attempts have thus far focused on jointly learning the node representations and classifier for detecting anomalies. Inspired by recent advances of self-supervised (SSL) graphs, we explore another possibility decoupling representation classification detection. We conduct a preliminary study to show that decoupled training using existing graph SSL schemes represent nodes can obtain performance gains over joint...
ABSTRACT Aim To investigate the involvement of low‐density lipoprotein receptor‐related protein 5 (LRP5) in inflammation and alveolar bone loss periodontitis. Materials Methods Gingival tissues were obtained from 10 periodontitis patients healthy individuals. Wild‐type (WT) osteoblast‐specific Lrp5 conditional knock‐out C57BL/6 (LRP5fl/fl;Oc‐Cre) mice used to establish a ligature‐induced mouse model Human periodontal ligament stem cells (hPDLSCs) isolated further verify mechanism through...
Rapid rural social-economic transitions in China have given rise to increasingly serious agricultural eco-environmental (AEE) problems. The deterioration of AEE poses great challenges the sustainable agriculture and results an incoordination among economy, society ecology. Coordinating relationship between transition development has significance sustainability revitalization strategy. However, few studies focus on interactions areas. In this study, comprehensive index systems were built...
Service processes in SOA are composed dynamically by services from different service providers. At run-time, some may become faulty and cause a process to violate its end-to-end quality of (QoS) constraints. We propose an effective approach for replacing only their neighboring maintain the original QoS use iterative algorithm search reconfiguration region that has replaceable meet constraint region. Services regions be replaced using one-to-one, one-to-many, or many-to-one mappings. By...
Deep neural networks (NNs) have exhibited considerable potential for efficiently balancing the performance and complexity of multiple-input multiple-output (MIMO) detectors. We propose a receiver framework that enables efficient online training by leveraging following simple observation: although NN parameters should adapt to channels, not all them are channel-sensitive. In particular, we use deep unfolded structure represents iterative algorithms in signal detection channel decoding modules...
Background: Stathmin1 (Stmn1) is a protein highly expressed during the development of central nervous system. The phosphorylation Stmn1 involves microtubule dynamics, so plays vital part in neurite outgrowth and synaptic plasticity. Previous studies reported that genetic variants influence fear anxiety as well cognitive-affective processing. However, no study on relationship between gene polymorphism cognition Chinese. Thus, this association was investigated present study. Methods: A total...
Abstract The optimized titanium-doped silane treatment agent and its process silanized the NdFeB magnet, resulting in a dense film of on surface. Using SEM, EDS, AFM, FT-IR, NSS, microstructure characteristics films were described. findings demonstrate structure smooth surface membrane. In addition, nanoscale compound particles containing Ti F elements uniformly dispersed film. Ti-O-Si bond may was formed between doped fluorotitanate silanol, element did not form chemical with silane....
Survivability is a key concern in modern network design order to achieve fast service restorability against failures. This paper investigates the problem of survivable dynamic connection provisioning general telecom backbone networks, which are mesh structured. These networks employ optical fibers, may fail due outages such as fiber cuts, etc. Our study applies survivability wavelength-division multiplexing (WDM) well multi-protocol label switching (MPLS) networks. For study, we assume...
In this paper, we propose a new unsupervised text detection approach which is based on Histogram of Oriented Gradient and Graph Spectrum. By investigating the properties edges, proposed first extracts edges from an image localize candidate character blocks using Gradients, then Spectrum utilized to capture global relationship among cluster into groups generate bounding boxes objects in image. The method robust color size text. ICDAR 2003 locating dataset video frames were used evaluate...
Orthogonal frequency division multiplexing (OFDM), usually with sufficient cyclic prefix (CP), has been widely applied in various communication systems. The CP OFDM consumes additional resource and reduces spectrum energy efficiency. However, channel estimation signal detection are very challenging for CP-free In this paper, we propose a novel artificial intelligence (AI)-aided receiver (AI receiver) system. AI includes neural network (CE-NET) based on orthogonal approximate message passing...
Due to the characteristics of underwater acoustic (UWA) channel, process at receiver is complicated match channel. To simplify design and UWA channel better, this study proposes a deep neural network‐based orthogonal frequency division multiplexing for communication. Different from existing receivers needing network several other processing parts, proposed only uses single implement whole signal processing. Moreover, it general which suitable modulation schemes. Simulation results show that...
Orthogonal frequency division multiplexing (OFDM) has been widely applied in current communication systems. The artificial intelligence (AI)-aided OFDM receivers are currently brought to the forefront replace and improve traditional receivers. In this study, we first compare two AI-aided receivers, namely, data-driven fully connected deep neural network model-driven ComNet, through extensive simulation real-time video transmission using a 5G rapid prototyping system for an over-the-air (OTA)...
This paper proposes a novel tensor-train deep neural network (TT-DNN) based channel estimator to tackle challenges of time-varying estimation in multiple-input multiple-output (MIMO) systems. A centralized DNN can be realized by distributed TT-DNN with parallel paths. The provides compact representation decomposing each layer into TT format fewer model parameters, and is well-designed adapt the block structure, pilot density, number MIMO antennas. Moreover, performed block-by-block...