A5100347765- Mobile Ad Hoc Networks
- Energy Efficient Wireless Sensor Networks
- Opportunistic and Delay-Tolerant Networks
- Domain Adaptation and Few-Shot Learning
- Energy Harvesting in Wireless Networks
- Advanced Neural Network Applications
- Cooperative Communication and Network Coding
- Human Mobility and Location-Based Analysis
- Topic Modeling
- Vehicular Ad Hoc Networks (VANETs)
- Software-Defined Networks and 5G
- Wireless Networks and Protocols
- Caching and Content Delivery
- Multimodal Machine Learning Applications
- Natural Language Processing Techniques
- Anomaly Detection Techniques and Applications
- Radiomics and Machine Learning in Medical Imaging
- Indoor and Outdoor Localization Technologies
- Machine Learning and Data Classification
- IoT and Edge/Fog Computing
- Wireless Signal Modulation Classification
- Digital Media Forensic Detection
- Neural Networks and Applications
- Advanced MIMO Systems Optimization
- Video Surveillance and Tracking Methods
University of Electronic Science and Technology of China
2016-2025
Beijing Jiaotong University
2006-2024
China Electronics Technology Group Corporation
2021-2024
Guangzhou Automobile Group (China)
2024
Quzhou University
2022-2024
Deakin University
2020-2024
Harbin Institute of Technology
2012-2024
Fourth Hospital of Hebei Medical University
2023-2024
Hebei Medical University
2023-2024
Wenzhou Medical University
2022-2023
Although end-to-end neural text-to-speech (TTS) methods (such as Tacotron2) are proposed and achieve state-of-theart performance, they still suffer from two problems: 1) low efficiency during training inference; 2) hard to model long dependency using current recurrent networks (RNNs). Inspired by the success of Transformer network in machine translation (NMT), this paper, we introduce adapt multi-head attention mechanism replace RNN structures also original Tacotron2. With help...
Interactive AI-powered services require low-latency evaluation of deep neural network (DNN) models-aka ""real-time AI"". The growing demand for computationally expensive, state-of-the-art DNNs, coupled with diminishing performance gains general-purpose architectures, has fueled an explosion specialized Neural Processing Units (NPUs). NPUs interactive should satisfy two requirements: (1) execution DNN models low latency, high throughput, and efficiency, (2) flexibility to accommodate evolving...
Accurate semantic segmentation of remote sensing data plays a crucial role in the success geoscience research and applications. Recently, multimodal fusion-based models have attracted much attention due to their outstanding performance as compared conventional single-modal techniques. However, most these perform fusion operation using convolutional neural networks (CNN) or vision transformer (Vit), resulting insufficient local-global contextual modeling representative capabilities. In this...
Vehicular edge computing has been a promising paradigm to offer low-latency and high reliability vehicular services for users. Nevertheless, compute-intensive vehicle applications, most previous researches cannot perform them efficiently due both the inadequate of infrastructure construction resource bottleneck server. Motivated by fact that there is large number outside parked vehicles with rich underutilized resources in urban area, we propose idea parking computing, which makes use assist...
Dangling bond formation for COF materials in a rational manner is an enormous challenge, especially through post-treatment which facile strategy while has not been reported yet. In this work, "chemical scissor" proposed the first time to rationally design dangling bonds materials. It found that Zn2+ coordination post-metallization of TDCOF can act as "inducer" elongates target and facilitates its fracture hydrolyzation reactions create bonds. The number well-modulated by controlling time....
Diagnosing liver lesions is crucial for treatment choices and patient outcomes. This study develops an automatic diagnosis system using multiphase enhanced computed tomography (CT). A total of 4039 patients from six data centers are enrolled to develop Liver Lesion Network (LiLNet). LiLNet identifies focal lesions, including hepatocellular carcinoma (HCC), intrahepatic cholangiocarcinoma (ICC), metastatic tumors (MET), nodular hyperplasia (FNH), hemangioma (HEM), cysts (CYST). Validated in...
In Vehicular Ad Hoc Networks (VANETs), content distribution directly relies on the fleeting and dynamic contacts between moving vehicles, which often leads to prolonged downloading delay terrible user experience. Deploying Wifi-based Access Points (APs) could relieve this problem, but it requires a large amount of investment, especially at city scale. paper, we propose idea ParkCast, doesn't need leverages roadside parking distribute contents in urban VANETs. With wireless device rechargable...
Federated Learning aims to learn machine learning models from multiple decentralized edge devices (e.g. mobiles) or servers without sacrificing local data privacy. Recent Natural Language Processing techniques rely on deep and large pre-trained language models. However, both big neural are trained with huge amounts of which often lies the server side. Since text is widely originated end users, in this work, we look into recent NLP use federated as framework. Our survey discusses major...
Radio frequency fingerprint (RFF) identification technique has drawn great attention to wireless terminal authentication. Long-Term Evolution (LTE) been widely deployed all over the world. RFF-based LTE identifications can prevent potential impersonation or denial of service (DoS) attacks in physical layer. This paper proposes a novel multi-channel convolutional neural network (MCCNN) for identification. Differential constellation trace figure (DCTF) is extracted from random access preamble...
This retrospective study aimed to develop and validate deep-learning-based models for grading clear cell renal carcinoma (ccRCC) patients. A cohort enrolling 706 patients (n = 706) with pathologically verified ccRCC was used in this study. temporal split applied verify our models: the first 83.9% of cases (years 2010-2017) development last 16.1% (year 2018-2019) validation (development cohort: n 592; 114). Here, we demonstrated a deep learning(DL) framework initialized by self-supervised...
Information interaction is a crucial part of modern transportation activities. In this paper, we propose the idea Vehicle-to-Passenger communication (V2P), which allows direct, instant, and flexible between moving vehicles roadside passengers. With pocket wireless devices, passengers can easily join VANETs as nodes, express their travel demands, e.g., taking free ride or calling taxi via radio queries over VANETs. Once matched vehicle found through disseminated queries, driver decide whether...
Objective: To investigate whether the tag single nucleotide polymorphisms (tSNPs) in VSX1, COL4A3, COL4A4, IL1A and IL1B genes were associated with keratoconus (KTCN) Han Chinese population.Methods: Ninety-seven KTCN patients 101 healthy controls enrolled this study. All cases diagnosed on basis of clinical examination. Twenty-one tSNPs selected for association study five genes. SNP genotyping was performed by Sequenom MassARRAY RS1000. Typer 4.0 Software, PLINK, Haploview SHEsis software...
Since pervasive smartphones own advanced computing capability and are equipped with various sensors, they have been used for dangerous driving behaviors detection, such as drunk driving. However, sensory data gathered by noisy, which results in inaccurate estimations. Some existing works try to filter noise from sensor readings, but usually only the outlier filtered. The noises caused hardware of smartphone cannot be removed reading. In this paper, we propose DrivingSense, a reliable...
The energy problem is one of the most important challenges in application sensor networks. With development wireless charging technology and intelligent mobile charger (MC), can be solved by strategy. In practical rechargeable networks (WRSNs), consumption rate nodes dynamically changed due to many uncertainties, such as death different transmission tasks nodes. However, existing works focus on on-demand schemes, which not fully consider real-time global scheduling. this article, a novel...
In this paper, we propose a novel data augmentation strategy named Cut-Thumbnail, that aims to improve the shape bias of network. We reduce an image certain size and replace random region original with reduced image. The generated not only retains most information but also has global in call as thumbnail. Furthermore, find idea thumbnail can be perfectly integrated Mixed Sample Data Augmentation, so put one image's on another while ground truth labels are mixed, making great achievements...
This review presents a survey of representative mononuclear and multinuclear actinide SMMs to look back at how far we have come, trying seek out feasible strategy for top-performing An-SMMs.
A novel synthetic route to the triple-decker benzene tetra-anion complexes [(η5-C5iPr5)M(μ:η6:η6-C6H6)M(η5-C5iPr5)] is reported for a range of early and late rare-earth elements, i.e., M = Y, La, Sm, Gd, Dy (1M). The lanthanum complex 1La first largest element. Aromaticity in 10π-electron ligands confirmed through crystallographic studies all compounds nucleus-independent chemical shift calculations on 1Y 1La. Analysis bonding using density functional theory revealed strong covalency...