- Topic Modeling
- Caching and Content Delivery
- Cloud Computing and Resource Management
- Natural Language Processing Techniques
- Advanced Computational Techniques and Applications
- Advanced Data Storage Technologies
- Distributed and Parallel Computing Systems
- Speech and Audio Processing
- Opportunistic and Delay-Tolerant Networks
- Advanced Text Analysis Techniques
- Energy Efficient Wireless Sensor Networks
- Data Management and Algorithms
- Indoor and Outdoor Localization Technologies
- Distributed systems and fault tolerance
- Service-Oriented Architecture and Web Services
- Wireless Sensor Networks and IoT
- Advanced Database Systems and Queries
- Semantic Web and Ontologies
- Advanced Clustering Algorithms Research
- Advanced Algorithms and Applications
- Metaheuristic Optimization Algorithms Research
- IoT and Edge/Fog Computing
- Peer-to-Peer Network Technologies
- Mobile Ad Hoc Networks
- IoT Networks and Protocols
Guangzhou Institute of Applied Software Technology, Chinese Academy of Sciences
2024
Chengdu University of Technology
2023
Cornell University
2023
Macau University of Science and Technology
2023
Northeastern University
2016-2022
Tsinghua University
2011-2021
Chongqing University of Posts and Telecommunications
2020-2021
Fuzhou University
2021
Institute of Microelectronics
2021
Huazhong University of Science and Technology
2012-2020
We present nbodykit, an open-source, massively parallel Python toolkit for analyzing large-scale structure (LSS) data. Using bindings of the Message Passing Interface (MPI), we provide implementations many commonly used algorithms in LSS. nbodykit is both interactive and scalable piece scientific software, performing well a supercomputing environment while still taking advantage tools provided by ecosystem. Existing functionality includes estimators power spectrum, 2 3-point correlation...
User experience of smart mobile devices can be improved in numerous scenarios with the assist in-air gesture recognition. Most existing methods proposed by industry and academia are based on special sensors. On contrary, a sensor-independent recognition method named Dolphin is this paper which applied to off-the-shelf directly. The only sensors needs loudspeaker microphone embedded device. emits continuous 21 KHz tone receive gesture-reflecting ultrasonic wave microphone. performed encoded...
Transformers have emerged as a powerful tool for broad range of natural language processing tasks. A key component that drives the impressive performance is self-attention mechanism encodes influence or dependence other tokens on each specific token. While beneficial, quadratic complexity input sequence length has limited its application to longer sequences -- topic being actively studied in community. To address this limitation, we propose Nystr\"{o}mformer model exhibits favorable...
Wheat end-use quality mainly derives from two interrelated characteristics: the compositions of gluten proteins and grain hardness. The composition determines dough rheological properties thus confers unique viscoelastic property on dough. One group proteins, high molecular weight glutenin subunits (HMW-GS), plays an important role in functional properties. On other hand, hardness, which influences milling process flour, is controlled by Puroindoline a (Pina) b (Pinb) genes. However, little...
One of the key performance measures in wireless cellular systems is handoff dropping probability. The paper studies for an integrated voice/data system. Specifically, we propose a new control policy, called dual trunk reservation with queuing (DTR-Q). It uses two thresholds, one to reserve bandwidth voice handoff, and other managing data traffic. By taking advantage adaptiveness traffic, it single queue buffer request further case requested not available. unique features proposed DTR-Q...
The combination of new-generation information technology and manufacturing has had in a significant profound impact on the future development paradigm manufacturing. Machine tools are basis virtually everything that is manufactured industry, exploring machine tool domain considerable significance to identify opportunity develop industry promote sustainable current tightening constraints resource environment. Although much attention been paid studies specific recent years, it challenging...
The long propagation delay, limited bandwidth and high bit error rate pose great challenges in Media Access Control (MAC) protocol design for underwater acoustic networks. In this paper, we proposed an MAC based on Carrier Sensing Multiple with Collision Avoidance (CSMA/CA), which is called UW-CSMA/CA (CSMA/CA underwater). Three areas of improvement are investigated, namely, the control rules, deferral backoff slot time. simulations ns2 demonstrate that throughput performance clearly higher...
Web service composition can make use of distributed services with various functions to accomplish a specific task. Service binding and become keywords computing. In the context workflow management, plays an important part in forming concrete after abstract one is submitted. Existing research devoted selecting set servers run corresponding application achieving better QoS. Due data intensity flow, communication between two server nodes may bring significant delay that not be neglected when...
The effectiveness of automatic storage management depends on the accuracy performance models that are used for making resource allocation decisions. Several approaches have been proposed modeling. Black-box most promising in real-world systems because they require minimal device specific information, and self-evolving with respect to changes system. However, blackbox techniques traditionally considered inaccurate non-converging systems. This paper evaluates a popular off-the-shelf black-box...
This paper proposes a novel 3D discrete cosine transform (DCT) based image compression method for medical endoscopic applications. Due to the high correlation among color components of wireless capsule endoscopy (WCE) images, original 2D Bayer data pattern is reconstructed into new pattern, and DCT adopted compress ratio quality. For low computational complexity 3D-DCT, an optimized 4-point butterfly structure without multiplication operation proposed. unique characteristics quantization...
We present a system which extracts word-based bigram and n-gram collocation information from 60MB corpus then locates pairs using strength spread as defined in the Xtract system. In order for to work effectively with Chinese, we have readjusted parameters. To obtain higher recall rate, modified algorithm identify collocations low-frequency of occurrence, method works particularly well case bigrams one word is high-frequency other low-frequency. preliminary experiments, our precision 61%, an...
The relationship between the color of sphalerite and trace elements has aroused interest researchers since a long time, but most study is still in qualitative stage. In this contribution, we made attempt to reveal reflection its element, order discuss genetic significance various colors through quantitative sphalerite. Based on LA-ICP-MS analysis microspectrophotometer samples different types deposits, with significant correlation index were found analyzed by factor analysis. Different...
Opportunistic networks have emerged as prospective network architecture for smart mobile devices related applications. The main obstacle designing routing protocol in opportunistic is their inherent dynamic nature. Unknown future link patterns lead to blind and inefficient packet forwarding behavior. Lots of efforts focused on the prediction problem, but contact among node pairs are too different be predicted by one single method. To this end, we propose a combination method named Combo-Pre....
We present an automatic Chinese collocation extraction system using lexical statistics and syntactical knowledge. This extracts collocations from manually segmented tagged news corpus in three stages. First, the bidirectional bigram statistical measures, including strength spread, /spl chi//sup 2/ test value, are employed to extract candidate two-word pairs. These word pairs then used high frequency multiword their context. In third stage, precision is further improved by knowledge of...
Storage management is becoming the largest component in overall cost of storage ownership. Most organizations are trying to either consolidate their operations or outsource them a service provider (SSP) order contain costs. Currently, there do not exist any planning tools that help clients and SSPs figuring out best outsourcing option. In this paper we present tool, Brahma, specifically addresses above mentioned problem, as Brahma capable providing solutions where tasks split between client...
This document explains and demonstrates how the CoE protocol works in application layer of EtherCAT, its function. Using PIC24H series MCU as controller, FB1111-0142 ESC TwinCAT master, experiment results show work mechanism function exactly.