Impact of Quantization Noise on CNN-based Joint Source-Channel Coding and Modulation

This paper investigated the impact of a quantizer in analog-to-digital and digital-to-analog converters in communication devices on image quality when using deep learning-based joint source-channel coding modulation (JSCCM) for image transmission. In recent years, JSCCM, which efficiently encodes im...

Full description

Saved in:
Bibliographic Details
Published in:IEEE Consumer Communications and Networking Conference pp. 465 - 468
Main Authors: Matsumoto, Keigo, Inoue, Yoshiaki, Hara-Azumi, Yuko, Maruta, Kazuki, Nakayama, Yu, Hisano, Daisuke
Format: Conference Proceeding
Language:English
Japanese
Published: IEEE 08.01.2023
Subjects:
deep learning
Image coding
Image communication
Image quality
Joint source-channel coding
PSNR
quantization
Quantization (signal)
Symbols
Training
ISSN:2331-9860
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract This paper investigated the impact of a quantizer in analog-to-digital and digital-to-analog converters in communication devices on image quality when using deep learning-based joint source-channel coding modulation (JSCCM) for image transmission. In recent years, JSCCM, which efficiently encodes images and videos with low information entropy, has attracted great attention. JSCCM has a structure based on an autoencoder and determines the compression ratios for the image input by adjusting the number of IQ symbol output. The IQ symbol output from the encoder are allocated to symbol constellations with higher degrees of arbitrariness than those in typical square quadrature amplitude modulation and are therefore expected to be strongly affected by the quantization noise. In this paper, we employed quantization to the IQ symbol sequence and investigated its effect. Adjusting the quantizer's clipping ratio and the number of quantization bits, we examined the images' tolerance of the peak signal-to-noise ratio (PSNR). The simulation results showed that by adequately adjusting the clipping ratio, the image quality can be guaranteed to be equivalent to ideal conditions without quantization noise, and the number of required quantization bits that do not degrade the PSNR, was calculated.
AbstractList This paper investigated the impact of a quantizer in analog-to-digital and digital-to-analog converters in communication devices on image quality when using deep learning-based joint source-channel coding modulation (JSCCM) for image transmission. In recent years, JSCCM, which efficiently encodes images and videos with low information entropy, has attracted great attention. JSCCM has a structure based on an autoencoder and determines the compression ratios for the image input by adjusting the number of IQ symbol output. The IQ symbol output from the encoder are allocated to symbol constellations with higher degrees of arbitrariness than those in typical square quadrature amplitude modulation and are therefore expected to be strongly affected by the quantization noise. In this paper, we employed quantization to the IQ symbol sequence and investigated its effect. Adjusting the quantizer's clipping ratio and the number of quantization bits, we examined the images' tolerance of the peak signal-to-noise ratio (PSNR). The simulation results showed that by adequately adjusting the clipping ratio, the image quality can be guaranteed to be equivalent to ideal conditions without quantization noise, and the number of required quantization bits that do not degrade the PSNR, was calculated.
Author Hara-Azumi, Yuko
Nakayama, Yu
Inoue, Yoshiaki
Matsumoto, Keigo
Hisano, Daisuke
Maruta, Kazuki
Author_xml – sequence: 1
  givenname: Keigo
  surname: Matsumoto
  fullname: Matsumoto, Keigo
  email: k_matsumoto@pn.comm.eng.osaka-u.ac.jp
  organization: Graduate School of Engineering, Osaka University,Osaka,Japan
– sequence: 2
  givenname: Yoshiaki
  surname: Inoue
  fullname: Inoue, Yoshiaki
  email: yoshiaki@comm.eng.osaka-u.ac.jp
  organization: Graduate School of Engineering, Osaka University,Osaka,Japan
– sequence: 3
  givenname: Yuko
  surname: Hara-Azumi
  fullname: Hara-Azumi, Yuko
  email: hara@cad.ict.e.titech.ac.jp
  organization: School of Engineering, Tokyo Institute of Technology,Tokyo,Japan
– sequence: 4
  givenname: Kazuki
  surname: Maruta
  fullname: Maruta, Kazuki
  email: kazuki.maruta@ieee.org
  organization: Graduate School of Engineering, Tokyo University of Science,Tokyo,Japan
– sequence: 5
  givenname: Yu
  surname: Nakayama
  fullname: Nakayama, Yu
  email: yu.nakayama@ieee.org
  organization: Institute of Engineering, Tokyo University of Agriculture and Technology,Tokyo,Japan
– sequence: 6
  givenname: Daisuke
  surname: Hisano
  fullname: Hisano, Daisuke
  email: hisano@comm.eng.osaka-u.ac.jp
  organization: Graduate School of Engineering, Osaka University,Osaka,Japan
BookMark eNo1kMtOwzAURA0CiVLyB0j4BxKuH4ntJbJ4FJUgRFlX14kDRqldJekCvp7yWs1ojmYWc0qOYoqekAsGBWNgLq2tbckqKQsOXBQMoAKp9QHJjNKsqkpplJDikMy4ECw3uoITko3jOwAw0GVp9IysFpstNhNNHX3aYZzCJ04hRVqnMHq6N7auc4ejb-l9CnGiz2k3ND63bxij76lNbYivFGNLH1K763_aZ-S4w3702Z_OycvN9cre5cvH24W9WuaBGzPlWjqpoFWSCezQKOxUw_ehZug4aoXaM8Oxah2y5htx7TotS18aZ0A5MSfnv7vBe7_eDmGDw8f6_wjxBbsDU94
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/CCNC51644.2023.10060488
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 9781665497343
1665497343
EISSN 2331-9860
EndPage 468
ExternalDocumentID 10060488
Genre orig-research
GroupedDBID 6IE
6IF
6IH
6IK
6IL
6IN
AAJGR
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IPLJI
M43
OCL
RIE
RIL
RNS
ID FETCH-LOGICAL-i299t-84b470d7413afa97af7c284b81ab2a87a8e192a6dba1c7c2828bf845e59b907b3
IEDL.DBID RIE
ISICitedReferencesCount 1
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000982339100088&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 02:18:25 EDT 2025
IsPeerReviewed false
IsScholarly false
Language English
Japanese
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i299t-84b470d7413afa97af7c284b81ab2a87a8e192a6dba1c7c2828bf845e59b907b3
PageCount 4
ParticipantIDs ieee_primary_10060488
PublicationCentury 2000
PublicationDate 2023-01-08
PublicationDateYYYYMMDD 2023-01-08
PublicationDate_xml – month: 01
  year: 2023
  text: 2023-01-08
  day: 08
PublicationDecade 2020
PublicationTitle IEEE Consumer Communications and Networking Conference
PublicationTitleAbbrev CCNC
PublicationYear 2023
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0001085598
Score 1.854677
Snippet This paper investigated the impact of a quantizer in analog-to-digital and digital-to-analog converters in communication devices on image quality when using...
SourceID ieee
SourceType Publisher
StartPage 465
SubjectTerms deep learning
Image coding
Image communication
Image quality
Joint source-channel coding
PSNR
quantization
Quantization (signal)
Symbols
Training
Title Impact of Quantization Noise on CNN-based Joint Source-Channel Coding and Modulation
URI https://ieeexplore.ieee.org/document/10060488
WOSCitedRecordID wos000982339100088&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NSwMxEA22ePCkYsVvcvCaul9psufFooJLxQq9lXxMYEE2Yrf-fifp1uLBg7ewYZdlQnjzXuZlCLlN3MSkJjEMqbJkyDcSppVFzoNoqbjJE5er2GxC1LVcLMpZb1aPXhgAiMVnMA7DeJZvvVkHqQx3eLjqRcoBGQgx2Zi1doJKqLgqZV_DlSblXVXVFUc6EKSTLB9v3_7VRyXCyPTwnz9wREY7Qx6d_UDNMdmD9oTMH6PDkXpHX9YYoN5RSWvfrIDioKprFlDK0ifftB19jUI9C4aCFt5p5cPHqGotffa2b-M1Im_T-3n1wPomCaxBJOmYLHQhEouJQa6cKoVywiDkaJkqnSkplARM4tTEapWaMJVJ7WTBgZcaibHOT8mw9S2cEZpqoV2iOTIMW1goFcdkFlLuQAieCXVORiEky4_NPRjLbTQu_nh-SQ5C4KNgIa_IsPtcwzXZN19ds_q8iav3Dcmgmms
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NSwMxEA1aBT2pWPHbHLym7lea7HmxtNqGihV6K_mEBdmVduvvdxK3Fg8evIUsu4QJ4c17Oy-D0H3k-jrWkSZAlTkBvhERJQ1wHkBLSXUauVSGZhNMCD6f59PWrB68MNbaUHxme34Y_uWbWq-9VAYn3F_1wvku2vOts1q71lZS8TVXOW-ruOIofygKUVAgBF48SdLe5v1fnVQCkAyO_rmEY9TdWvLw9AdsTtCOrU7RbBQ8jrh2-GUNIWo9lVjU5cpiGBRCEI9TBj_VZdXg1yDVE28pqOw7Lmr_MSwrgye1aRt5ddHb4HFWDEnbJoGUgCUN4ZnKWGQgNUilkzmTjmkAHcVjqRLJmeQW0jjZN0rG2j9KuHI8o5bmCqixSs9Qp6ore45wrJhykaLAMUxmbC4ppLM2ps4yRhMmL1DXh2Tx8X0TxmITjcs_5u_QwXA2GS_GI_F8hQ79JgT5gl-jTrNc2xu0rz-bcrW8DTv5BQaunbQ
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=IEEE+Consumer+Communications+and+Networking+Conference&rft.atitle=Impact+of+Quantization+Noise+on+CNN-based+Joint+Source-Channel+Coding+and+Modulation&rft.au=Matsumoto%2C+Keigo&rft.au=Inoue%2C+Yoshiaki&rft.au=Hara-Azumi%2C+Yuko&rft.au=Maruta%2C+Kazuki&rft.date=2023-01-08&rft.pub=IEEE&rft.eissn=2331-9860&rft.spage=465&rft.epage=468&rft_id=info:doi/10.1109%2FCCNC51644.2023.10060488&rft.externalDocID=10060488