Skip to main navigation Skip to search Skip to main content

HED-CNN based Ionospheric Clutter Extraction for HF Range-Doppler Spectrum

  • School of Electronics and Information Engineering, Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

High Frequency Surface Wave Radar (HFSWR) suffers seriously with the ionospheric clutter formed from ionosphere echoes. The ionospheric clutter could be extensive and exists all day long, which restricts the detection performance of HFSWR. It is necessary to eliminate the interference of ionospheric clutter which overwhelms target echoes always. However, there is not a prior knowledge about clutter each work cycle, and anti-ionospheric interference technology adapting to all kinds of situations. With the purpose of extracting the ionospheric clutter separately for clutter cancellation, image processing method is adopted to study and analyze the application of deep learning in edge extraction of ionospheric clutter existing in Range-Doppler (RD) spectrum. In this paper, holistically-nested edge detection (HED) based algorithm is adopted and Canny algorithm is used for comparison. It shows that HED algorithm is effective and efficient in edge extraction of ionospheric clutter in RD spectrum.

Original languageEnglish
Title of host publicationProceedings - 2022 8th Annual International Conference on Network and Information Systems for Computers, ICNISC 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages155-159
Number of pages5
ISBN (Electronic)9781665453516
DOIs
StatePublished - 2022
Externally publishedYes
Event8th Annual International Conference on Network and Information Systems for Computers, ICNISC 2022 - Virtual, Online, China
Duration: 16 Sep 202219 Sep 2022

Publication series

NameProceedings - 2022 8th Annual International Conference on Network and Information Systems for Computers, ICNISC 2022

Conference

Conference8th Annual International Conference on Network and Information Systems for Computers, ICNISC 2022
Country/TerritoryChina
CityVirtual, Online
Period16/09/2219/09/22

Keywords

  • HED
  • deep learning
  • edge extraction
  • ionospheric clutter

Fingerprint

Dive into the research topics of 'HED-CNN based Ionospheric Clutter Extraction for HF Range-Doppler Spectrum'. Together they form a unique fingerprint.

Cite this