Title:
RECURRENT RESIDUAL NETWORK-BASED SIGNAL MODULATION AND RECOGNITION METHOD AND DEVICE
Document Type and Number:
WIPO Patent Application WO/2021/077841
Kind Code:
A1
Abstract:
The embodiments of the present application provide a recurrent residual network-based signal modulation and recognition method and device. Said method comprises: acquiring a signal matrix of a signal to be recognized, and extracting real part information and imaginary part information of the signal matrix; generating a real part-imaginary part feature matrix of said signal according to the extracted real part information and imaginary part information; converting the real part-imaginary part feature matrix into an amplitude-phase feature matrix according to a preset matrix conversion method; and inputting the amplitude-phase feature matrix into a pre-trained recurrent residual network, so as to obtain a modulation mode corresponding to said signal. In the embodiments of the present application, the processing of a signal to be recognized is simple and easy to operate, and neither complex algorithms nor manual processing is required, so that the recognition flexibility is high, and a recognition result of a modulation mode of the signal to be recognized can be obtained accurately.
Inventors:
FENG ZHIYONG (CN)
HUANG SAI (CN)
DAI RUI (CN)
NING FAN (CN)
ZHANG QIXUN (CN)
ZHANG YIFAN (CN)
WEI ZHIQING (CN)
HUANG SAI (CN)
DAI RUI (CN)
NING FAN (CN)
ZHANG QIXUN (CN)
ZHANG YIFAN (CN)
WEI ZHIQING (CN)
Application Number:
PCT/CN2020/105749
Publication Date:
April 29, 2021
Filing Date:
July 30, 2020
Export Citation:
Assignee:
UNIV BEIJING POSTS & TELECOMM (CN)
International Classes:
H04L27/00
Foreign References:
CN110798417A | 2020-02-14 | |||
CN110309854A | 2019-10-08 | |||
CN110048980A | 2019-07-23 | |||
CN110071893A | 2019-07-30 | |||
CN108521387A | 2018-09-11 | |||
US20060188032A1 | 2006-08-24 |
Attorney, Agent or Firm:
PATENTSINO IP FIRM (CN)
Download PDF: