中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

558 records (took 0.031 seconds)

    Method and System for Graph-to-graph Prediction Based on Recurrent Neural Network
    241.
    发明公开
    Method and System for Graph-to-graph Prediction Based on Recurrent Neural Network 审中-公开

    公开(公告)号：US20240176990A1

    公开(公告)日：2024-05-30

    申请号：US18479474

    申请日：2023-10-02

    Applicant: Yangtze Delta Region Institute (Huzhou), University of Electronic Science and Technology of China

    Inventor： Guan WANG

    IPC: G06N3/0455 ,  G06N3/0985

    CPC classification number: G06N3/0455 ,  G06N3/0985

    Abstract: The present disclosure discloses a method and system for graph-to-graph prediction based on recurrent neural networks. The method includes: representing a graph structure as a sequence; based on the sequentialization representation of the graph structure, constructing a deep neural network model for graph-to-graph prediction, wherein both input and output of the deep neural network model are graph structures; the deep neural network model includes an Encoder and a Decoder, wherein, in the Encoder, employing encNodeRNN and encEdgeRNN in combination to encode an input graph, and the Decoder decodes based on an encoding vector obtained from the Encoder, thereby obtaining a corresponding predicted graph and realizing the graph-to-graph prediction.

    Flexible microwave power transistor and preparation method thereof
    244.
    发明授权
    Flexible microwave power transistor and preparation method thereof 有权

    公开(公告)号：US11973136B2

    公开(公告)日：2024-04-30

    申请号：US17264521

    申请日：2019-12-20

    Applicant: University of Electronic Science and Technology of China

    Inventor： Yuehang Xu ,  Yan Wang ,  Yunqiu Wu

    IPC: H01L29/778 ,  H01L21/02 ,  H01L23/14 ,  H01L29/20 ,  H01L29/66

    CPC classification number: H01L29/7786 ,  H01L23/145 ,  H01L29/2003 ,  H01L29/66462

    Abstract: The present disclosure provides a flexible microwave power transistor and a preparation method thereof. In view of great lattice mismatch and poor performance of a device prepared with a Si substrate in an existing preparation method, the preparation method of the present disclosure grows a gallium nitride high electron mobility transistor (GaN HEMT) layer on a rigid silicon carbide (SiC) substrate to avoid lattice mismatch between a silicon (Si) substrate and gallium nitride (GaN), improving performance of the flexible microwave power transistor. Moreover, in view of problems such as low output power, power added efficiency and power gain with the existing device preparation method, the present disclosure retains part of the rigid SiC substrate and grows a flexible substrates at room temperature to prepare a high-quality device. The present disclosure has greatly improved power output capability, efficiency and gain, and basically unchanged performance of device under 0.75% of stress.

    AUTOMATIC MODULATION CLASSIFICATION METHOD BASED ON DEEP LEARNING NETWORK FUSION
    246.
    发明公开
    AUTOMATIC MODULATION CLASSIFICATION METHOD BASED ON DEEP LEARNING NETWORK FUSION 审中-公开

    公开(公告)号：US20240112037A1

    公开(公告)日：2024-04-04

    申请号：US18076160

    申请日：2022-12-06

    Applicant: Yangtze Delta Region Institute of University of Electronic Science and Technology of China

    Inventor： Shunsheng ZHANG ,  Jie HUANG

    IPC: G06N3/091 ,  G06F17/15

    CPC classification number: G06N3/091 ,  G06F17/156

    Abstract: The present invention discloses an automatic modulation classification method based on deep learning network fusion, comprising: acquiring a WBFM sample signal within a data set RML 2016.10a, and selecting a proper threshold γ to separate a WBFM signal during a silence period; expanding a new WBFM signal to 1000 by adopting a data enhancement method, and expanding an original data set; dividing the data set expanded in the step S2 into a training set, a verification set and a test set; respectively calculating amplitude, phase and a fractional order Fourier transformation result for data in the step S3; building a multi-channel feature fusion network model composed of an LSTM network and an FPN network; performing network model training, after the end of training, inputting verification set data into a trained network model for verification, and calculating prediction accuracy; and performing parameter fine adjustment on the network model through said test set, improving prediction precision, and taking a final model as an automatic modulation classification model. The present invention enables the improvement to the average classification accuracy rate of communication signals.

    High-dimensional signal transmission method
    248.
    发明授权
    High-dimensional signal transmission method 有权

    公开(公告)号：US11923930B2

    公开(公告)日：2024-03-05

    申请号：US18033373

    申请日：2021-04-29

    Applicant: University of Electronic Science and Technology of China

    Inventor： Guangrong Yue ,  Daizhong Yu ,  Lin Yang

    IPC: H04B7/0456 ,  H04L5/00 ,  H04L25/03

    CPC classification number: H04B7/0456 ,  H04L5/0026 ,  H04L25/03343 ,  H04L25/0391 ,  H04L2025/03426

    Abstract: A high-dimensional signal transmission method is provided. The method generates M M-dimensional first signals on the basis of M original signals and generates M M-dimensional second signals on the basis of a precoding signal and of the first signals, and finally, a transmitter sums all of the second signals and then transmits by utilizing M subchannels. As such, each subchannel carries information of the M original signals; hence, when any subchannel experiences deep fading, the deep fading is shared jointly by M signals, thus preventing the deep fading from causing a particularly severe impact on any signal. Moreover, all of the original signals can be recovered by utilizing the signals on the other subchannels, thus increasing the systematic resistance against subchannel deep fading. Meanwhile, the system implements the parallel transmission of the M original signals, thus ensuring the throughput of a communication system.

    HIGH-DIMENSIONAL SIGNAL TRANSMISSION METHOD
    250.
    发明公开
    HIGH-DIMENSIONAL SIGNAL TRANSMISSION METHOD 审中-公开

    公开(公告)号：US20240022293A1

    公开(公告)日：2024-01-18

    申请号：US18033373

    申请日：2021-04-29

    Applicant: University of Electronic Science and Technology of China

    Inventor： Guangrong YUE ,  Daizhong YU ,  Lin YANG

    IPC: H04B7/0456 ,  H04L25/03 ,  H04L5/00

    CPC classification number: H04B7/0456 ,  H04L25/0391 ,  H04L5/0026 ,  H04L25/03343 ,  H04L2025/03426

    Abstract: A high-dimensional signal transmission method is provided. The method generates M M-dimensional first signals on the basis of M original signals and generates M M-dimensional second signals on the basis of a precoding signal and of the first signals, and finally, a transmitter sums all of the second signals and then transmits by utilizing M subchannels. As such, each subchannel carries information of the M original signals; hence, when any subchannel experiences deep fading, the deep fading is shared jointly by M signals, thus preventing the deep fading from causing a particularly severe impact on any signal. Moreover, all of the original signals can be recovered by utilizing the signals on the other subchannels, thus increasing the systematic resistance against subchannel deep fading. Meanwhile, the system implements the parallel transmission of the M original signals, thus ensuring the throughput of a communication system.

Patent Agency Ranking