Overview of Deep Learning-based CSI Feedback in Massive MIMO Systems

Many performance gains achieved by massive multiple-input and multiple-output depend on the accuracy of the downlink channel state information (CSI) at the transmitter (base station), which is usually obtained by estimating at the receiver (user equipment) and feeding back to the transmitter. The ov...

Full description

Saved in:

Bibliographic Details
Published in:	IEEE transactions on communications Vol. 70; no. 12; p. 1
Main Authors:	Guo, Jiajia, Wen, Chao-Kai, Jin, Shi, Li, Geoffrey Ye
Format:	Journal Article
Language:	English
Published:	New York IEEE 01.12.2022 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:	3GPP Communication Computer architecture CSI feedback Decoding Deep learning Downlink Feedback Image coding Image reconstruction Indexes Massive MIMO Neural networks overview State-of-the-art reviews Training Wireless communication systems
ISSN:	0090-6778, 1558-0857
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Abstract	Many performance gains achieved by massive multiple-input and multiple-output depend on the accuracy of the downlink channel state information (CSI) at the transmitter (base station), which is usually obtained by estimating at the receiver (user equipment) and feeding back to the transmitter. The overhead of CSI feedback occupies substantial uplink bandwidth resources, especially when the number of transmit antennas is large. Deep learning (DL)-based CSI feedback refers to CSI compression and reconstruction by a DL-based autoencoder and can greatly reduce feedback overhead. In this paper, a comprehensive overview of state-of-the-art research on this topic is provided, beginning with basic DL concepts widely used in CSI feedback and then categorizing and describing some existing DL-based feedback works. The focus is on novel neural network architectures and utilization of communication expert knowledge to improve CSI feedback accuracy. Works on joint design of CSI feedback with other communication modules are also introduced, and some practical issues, including bitstream generation, multirate feedback, imperfect feedback, NN complexity, training dataset collection, online training, and standardization effect, are discussed. At the end of the paper, some challenges and potential research directions associated with DL-based CSI feedback in future wireless communication systems are identified.
AbstractList	Many performance gains achieved by massive multiple-input and multiple-output depend on the accuracy of the downlink channel state information (CSI) at the transmitter (base station), which is usually obtained by estimating at the receiver (user equipment) and feeding back to the transmitter. The overhead of CSI feedback occupies substantial uplink bandwidth resources, especially when the number of transmit antennas is large. Deep learning (DL)-based CSI feedback refers to CSI compression and reconstruction by a DL-based autoencoder and can greatly reduce feedback overhead. In this paper, a comprehensive overview of state-of-the-art research on this topic is provided, beginning with basic DL concepts widely used in CSI feedback and then categorizing and describing some existing DL-based feedback works. The focus is on novel neural network architectures and utilization of communication expert knowledge to improve CSI feedback accuracy. Works on joint design of CSI feedback with other communication modules are also introduced, and some practical issues, including bitstream generation, multirate feedback, imperfect feedback, NN complexity, training dataset collection, online training, and standardization effect, are discussed. At the end of the paper, some challenges and potential research directions associated with DL-based CSI feedback in future wireless communication systems are identified.
Author	Wen, Chao-Kai Guo, Jiajia Li, Geoffrey Ye Jin, Shi
Author_xml	– sequence: 1 givenname: Jiajia orcidid: 0000-0002-6220-2295 surname: Guo fullname: Guo, Jiajia organization: National Mobile Communications Research Laboratory, Southeast University, Nanjing, P. R. China – sequence: 2 givenname: Chao-Kai orcidid: 0000-0001-5952-232X surname: Wen fullname: Wen, Chao-Kai organization: Institute of Communications Engineering, National Sun Yat-sen University, Kaohsiung, Taiwan – sequence: 3 givenname: Shi orcidid: 0000-0003-0271-6021 surname: Jin fullname: Jin, Shi organization: National Mobile Communications Research Laboratory, Southeast University, Nanjing, P. R. China – sequence: 4 givenname: Geoffrey Ye orcidid: 0000-0002-7894-2415 surname: Li fullname: Li, Geoffrey Ye organization: Department of Electrical and Electronic Engineering, Imperial College London, London, U.K
BookMark	eNp9kD1PwzAQhi1UJErhD8BiiTnFH3HsjCilUKlRhpbZcpILcmmTYqdF_fe4tGJg4Ja74X3uTs81GrRdCwjdUTKmlKSPy6zI8zEjjI05ozLUBRpSIVRElJADNCQkJVEipbpC196vCCEx4XyIJsUe3N7CF-4aPAHY4jkY19r2PSqNhxpnixmeAtSlqT6wbXFuvLd7wPksL_Di4HvY-Bt02Zi1h9tzH6G36fMye43mxcsse5pHFUtFHzHgRpVlJUoBcRgNiYVJyqrhJQfKKAhaE6UqmbC6SWrDJMQJQMiCoSSERujhtHfrus8d-F6vup1rw0nNpIhTkciUhZQ6pSrXee-g0ZXtTW-7tnfGrjUl-uhM_zjTR2f67Cyg7A-6dXZj3OF_6P4E2fDrL5CmnErK-TfUyHi6
CODEN	IECMBT
CitedBy_id	crossref_primary_10_1109_COMST_2023_3315374 crossref_primary_10_1109_TWC_2024_3354238 crossref_primary_10_1109_LWC_2024_3482863 crossref_primary_10_3390_math13152371 crossref_primary_10_1109_TWC_2024_3381211 crossref_primary_10_1109_LWC_2024_3518586 crossref_primary_10_1109_LWC_2024_3511553 crossref_primary_10_3390_s24185977 crossref_primary_10_1109_JSAC_2025_3559128 crossref_primary_10_1016_j_phycom_2024_102516 crossref_primary_10_1007_s11277_025_11767_7 crossref_primary_10_1109_MWC_010_2400078 crossref_primary_10_1109_ACCESS_2024_3457805 crossref_primary_10_1007_s12243_025_01113_0 crossref_primary_10_1109_JIOT_2024_3397302 crossref_primary_10_1109_TCCN_2023_3322983 crossref_primary_10_1109_TCOMM_2024_3524028 crossref_primary_10_1016_j_eswa_2024_125985 crossref_primary_10_1109_LWC_2024_3372148 crossref_primary_10_1016_j_phycom_2025_102756 crossref_primary_10_1109_TVT_2024_3502840 crossref_primary_10_1109_TWC_2025_3531447 crossref_primary_10_1145_3705893 crossref_primary_10_1109_JPROC_2024_3437730 crossref_primary_10_1109_LWC_2024_3367968 crossref_primary_10_1109_TCOMM_2025_3534528 crossref_primary_10_1109_TWC_2023_3329945 crossref_primary_10_1109_TCOMM_2024_3416889 crossref_primary_10_1109_TWC_2024_3400616 crossref_primary_10_1360_SSI_2024_0210 crossref_primary_10_1109_COMST_2024_3361991 crossref_primary_10_12688_f1000research_169843_1 crossref_primary_10_1109_LWC_2025_3574011 crossref_primary_10_1109_TWC_2025_3561786 crossref_primary_10_1109_TSP_2024_3423660 crossref_primary_10_1109_LWC_2025_3570965 crossref_primary_10_1109_TCCN_2024_3394867 crossref_primary_10_1109_TCCN_2023_3306391 crossref_primary_10_1109_TCOMM_2024_3384256 crossref_primary_10_1109_LCOMM_2024_3424434 crossref_primary_10_1109_TVT_2024_3461830 crossref_primary_10_1109_TCOMM_2024_3377724 crossref_primary_10_1109_TWC_2024_3490600 crossref_primary_10_1109_TCOMM_2023_3317924 crossref_primary_10_1109_JIOT_2025_3585120 crossref_primary_10_1109_TWC_2024_3481054 crossref_primary_10_1007_s10791_025_09542_0 crossref_primary_10_1109_LWC_2023_3329019 crossref_primary_10_1109_TWC_2024_3426525 crossref_primary_10_1109_LWC_2022_3227315 crossref_primary_10_3390_rs16020247 crossref_primary_10_1109_LWC_2024_3469383 crossref_primary_10_1109_LCOMM_2023_3267461 crossref_primary_10_1109_TWC_2024_3418907 crossref_primary_10_1109_TWC_2025_3560231 crossref_primary_10_1109_LCOMM_2023_3258749 crossref_primary_10_1109_LNET_2024_3475269 crossref_primary_10_1109_TCCN_2024_3435915 crossref_primary_10_1109_TWC_2024_3440275 crossref_primary_10_1016_j_sigpro_2025_110192 crossref_primary_10_1016_j_phycom_2025_102682 crossref_primary_10_1109_LWC_2023_3301992 crossref_primary_10_1109_TWC_2025_3529945 crossref_primary_10_1109_TWC_2024_3476474 crossref_primary_10_1109_TWC_2024_3400826 crossref_primary_10_1109_JIOT_2024_3430894 crossref_primary_10_1109_JSAC_2025_3559165 crossref_primary_10_1109_TCOMM_2023_3241326 crossref_primary_10_1109_LWC_2024_3395843 crossref_primary_10_1109_TVT_2024_3493463 crossref_primary_10_1109_TWC_2023_3339824 crossref_primary_10_1109_TWC_2024_3376633 crossref_primary_10_1109_TMC_2024_3495719 crossref_primary_10_1109_TWC_2023_3275990 crossref_primary_10_1109_LWC_2023_3324981 crossref_primary_10_1109_TWC_2024_3422399 crossref_primary_10_1109_LWC_2024_3407813 crossref_primary_10_1109_LWC_2024_3411065 crossref_primary_10_3390_electronics13020268 crossref_primary_10_1109_TVT_2024_3452409 crossref_primary_10_1109_JSEN_2024_3377911 crossref_primary_10_1016_j_phycom_2024_102305 crossref_primary_10_1109_LWC_2024_3400279 crossref_primary_10_1109_TMC_2024_3407792 crossref_primary_10_1109_TVT_2022_3226220 crossref_primary_10_1016_j_phycom_2024_102538 crossref_primary_10_1109_MWC_010_2200304 crossref_primary_10_1109_TWC_2023_3342735 crossref_primary_10_1109_TWC_2025_3545372 crossref_primary_10_3103_S0278641925700153 crossref_primary_10_1109_TWC_2023_3327984 crossref_primary_10_3390_s23104929 crossref_primary_10_1109_TVT_2023_3345938 crossref_primary_10_1016_j_eng_2025_07_032 crossref_primary_10_1109_TWC_2024_3428863 crossref_primary_10_1016_j_compeleceng_2025_110406 crossref_primary_10_1109_TWC_2024_3452438 crossref_primary_10_1049_ell2_12854 crossref_primary_10_1109_TWC_2023_3289929 crossref_primary_10_1007_s42979_025_04290_w crossref_primary_10_1109_TCOMM_2024_3429170 crossref_primary_10_3390_app14041356
Cites_doi	10.1109/CVPR.2018.00716 10.1109/TCOMM.2021.3138097 10.1007/s41745-020-00169-2 10.23919/JCC.2021.11.008 10.1109/TVT.2020.3004842 10.1049/ell2.12080 10.1016/j.patcog.2019.01.006 10.1109/18.382009 10.23919/JCIN.2020.9200894 10.1109/LWC.2021.3083331 10.4236/jcc.2021.910009 10.1109/LCOMM.2021.3076504 10.1109/MCOMSTD.0001.2200001 10.1109/TSP.2010.2090346 10.3390/e24040441 10.1109/ACCESS.2018.2850226 10.1109/TIT.2006.883550 10.1109/TWC.2020.2968430 10.1109/LCOMM.2020.2981448 10.23919/JCC.2022.08.002 10.1109/LWC.2020.2964550 10.23915/distill.00021 10.1109/26.380120 10.1109/SPAWC.2018.8446005 10.1109/TWC.2022.3182216 10.1109/LWC.2021.3118923 10.1109/TIT.2007.909170 10.1109/MWC.011.2100016 10.1109/MCOM.001.2001187 10.1109/JSAC.2017.2692307 10.1109/JSAC.2020.3041397 10.1109/TCOMM.2019.2960361 10.1109/TWC.2021.3073309 10.1109/TWC.2021.3103120 10.1109/LWC.2022.3149416 10.1109/LWC.2019.2942908 10.1109/MWC.001.1900157 10.1109/TCOMM.2021.3086525 10.1109/LCOMM.2018.2882829 10.1109/ISCEIC53685.2021.00056 10.1109/VTC2021-Fall52928.2021.9625047 10.1109/ICNP49622.2020.9259366 10.1109/LWC.2021.3099136 10.1109/MCOMSTD.001.1800036 10.1109/LWC.2021.3064963 10.1109/TCOMM.2020.2993626 10.1109/CVPR42600.2020.01009 10.1109/JSTSP.2017.2784180 10.1109/JCN.2020.000016 10.1109/CC.2017.8233654 10.1109/CVPR52688.2022.01166 10.1109/TCCN.2021.3119945 10.1109/JSAC.2008.081002 10.1109/LWC.2021.3100493 10.1109/TWC.2022.3149946 10.1137/080716542 10.1109/JIOT.2021.3139958 10.1109/SPAWC51858.2021.9593238 10.1109/GLOBECOM46510.2021.9685912 10.1109/MWC.2019.1800601 10.1109/MWC.006.2100543 10.1162/neco.1997.9.8.1735 10.1109/TSP.2014.2324991 10.1109/TCCN.2021.3084409 10.1109/ACCESS.2019.2901066 10.1109/LWC.2021.3117032 10.1109/LWC.2018.2818160 10.1109/WCNC49053.2021.9417115 10.1109/LWC.2019.2962114 10.1109/LCOMM.2020.2989499 10.1109/LWC.2021.3085317 10.1109/CVPR.2016.90 10.1109/TPAMI.2017.2699184 10.1109/ACCESS.2020.2963896 10.1109/LWC.2017.2757490 10.1109/WOCC53213.2021.9602863 10.1109/GLOBECOM42002.2020.9322121 10.1109/TCOMM.2021.3120294 10.1109/GlobalSIP45357.2019.8969557 10.1109/JSTSP.2022.3160268 10.1109/MWC.010.2200304 10.1109/ACCESS.2019.2928049 10.1109/LCOMM.2022.3185308 10.1109/LWC.2018.2874264 10.1109/CVPR.2017.243 10.1109/TVT.2022.3183596 10.1109/LWC.2021.3112747 10.1109/MWC.2019.1800447 10.1109/MSP.2018.2789521 10.1109/LWC.2021.3092947 10.1109/LWC.2021.3140102 10.1109/ICCV.2019.00200 10.1109/WCSP52459.2021.9613524 10.1109/MWC.2021.9363048 10.1109/LCOMM.2007.061534 10.1109/TVT.2021.3110608 10.1088/1742-6596/1693/1/012173 10.1109/TWC.2020.2970707 10.1109/MWC.001.2100136 10.1109/TWC.2022.3141653 10.1109/LCOMM.2021.3116864 10.1049/iet-com.2019.1030 10.1109/MWC.001.1900473 10.1109/TWC.2021.3055202 10.1109/ICC40277.2020.9149229 10.1109/MLSP.2019.8918798 10.1109/TIT.2005.858979 10.1109/LCOMM.2020.3017188 10.1109/VTC2021-Fall52928.2021.9625585 10.1109/LWC.2022.3216352 10.1109/LCOMM.2018.2877965 10.1109/LWC.2020.3017753 10.1109/MWC.2012.6393523 10.1109/WCNC49053.2021.9417500 10.1109/TWC.2021.3087191 10.1109/LWC.2022.3153085 10.1109/LCOMM.2022.3145099 10.1109/LCOMM.2021.3138927 10.1109/LWC.2022.3157263 10.1109/LWC.2021.3122462 10.1109/TCOMM.2020.3019077 10.1109/TVT.2021.3131606 10.1109/ACCESS.2022.3208284 10.1109/ACCESS.2019.2939938 10.1109/LWC.2021.3096808 10.1109/TCOMM.2022.3180388 10.1109/MSP.2008.930649 10.1109/JSTSP.2014.2317671 10.1109/WCNC.2019.8885897 10.1109/LWC.2021.3057934 10.1109/ACCESS.2022.3194035 10.1109/LWC.2019.2898662 10.1109/23.589532 10.1109/ICASSP40776.2020.9053850 10.1016/j.crma.2008.03.014 10.1109/BigData47090.2019.9006216 10.1109/TWC.2020.3043502 10.23919/JCC.2021.01.004 10.1109/LWC.2019.2895039 10.1109/TWC.2022.3160498 10.1109/LCOMM.2021.3099841 10.1109/TVT.2020.2980905 10.1109/LCOMM.2020.3019653 10.1109/TCSVT.2003.815173 10.1109/LCOMM.2021.3123941 10.1109/TCCN.2017.2758370 10.1016/j.dcan.2021.09.014
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2022
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2022
DBID	97E RIA RIE AAYXX CITATION 7SP 8FD L7M
DOI	10.1109/TCOMM.2022.3217777
DatabaseName	IEEE All-Society Periodicals Package (ASPP) 2005-present IEEE All-Society Periodicals Package (ASPP) 1998-Present IEEE Electronic Library (IEL) CrossRef Electronics & Communications Abstracts Technology Research Database Advanced Technologies Database with Aerospace
DatabaseTitle	CrossRef Technology Research Database Advanced Technologies Database with Aerospace Electronics & Communications Abstracts
DatabaseTitleList	Technology Research Database
Database_xml	– sequence: 1 dbid: RIE name: IEEE/IET Electronic Library (IEL) (UW System Shared) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering
EISSN	1558-0857
EndPage	1
ExternalDocumentID	10_1109_TCOMM_2022_3217777 9931713
Genre	orig-research
GroupedDBID	-~X .DC 0R~ 29I 3EH 4.4 5GY 5VS 6IK 85S 97E AAJGR AASAJ AAWTH ABFSI ABQJQ ABVLG ACGFO ACGFS ACIWK ACKIV ACNCT AENEX AETIX AGQYO AGSQL AHBIQ AI. AIBXA AKJIK AKQYR ALLEH ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 E.L EBS EJD HZ~ H~9 IAAWW IBMZZ ICLAB IES IFIPE IFJZH IPLJI JAVBF LAI M43 MS~ O9- OCL P2P RIA RIE RNS TAE TN5 VH1 ZCA ZCG AAYXX CITATION 7SP 8FD AARMG ABAZT L7M
ID	FETCH-LOGICAL-c295t-2e3a8bbc5b5e43a8a045a6bcf3b3e121e51d088c762df6da27e46eee43ea103b3
IEDL.DBID	RIE
ISICitedReferencesCount	141
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000927591900022&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0090-6778
IngestDate	Mon Jun 30 10:12:09 EDT 2025 Tue Nov 18 22:24:34 EST 2025 Sat Nov 29 04:08:25 EST 2025 Tue Nov 25 14:44:28 EST 2025
IsPeerReviewed	true
IsScholarly	true
Issue	12
Language	English
License	https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html https://doi.org/10.15223/policy-029 https://doi.org/10.15223/policy-037
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c295t-2e3a8bbc5b5e43a8a045a6bcf3b3e121e51d088c762df6da27e46eee43ea103b3
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0002-6220-2295 0000-0003-0271-6021 0000-0001-5952-232X 0000-0002-7894-2415
PQID	2754956792
PQPubID	85472
PageCount	1
ParticipantIDs	proquest_journals_2754956792 crossref_citationtrail_10_1109_TCOMM_2022_3217777 crossref_primary_10_1109_TCOMM_2022_3217777 ieee_primary_9931713
PublicationCentury	2000
PublicationDate	2022-12-01
PublicationDateYYYYMMDD	2022-12-01
PublicationDate_xml	– month: 12 year: 2022 text: 2022-12-01 day: 01
PublicationDecade	2020
PublicationPlace	New York
PublicationPlace_xml	– name: New York
PublicationTitle	IEEE transactions on communications
PublicationTitleAbbrev	TCOMM
PublicationYear	2022
Publisher	IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml	– name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References	(ref182) 2022 ref57 (ref23) 2022 ref56 ref59 ref58 ref53 ref52 ref55 ref169 ref54 zhang (ref141) 2022 olah (ref33) 2015 rao (ref20) 2014; 62 ref170 ref177 (ref147) 1988 ref178 ref175 ref51 ref174 ref171 ref46 ref45 ref48 ref47 ref42 ref179 ref44 woo (ref38) 2018 ref49 xu (ref153) 2022 ref180 ref8 ref181 ref7 guo (ref109) 2022 ref9 ref4 ref3 ref6 ref5 ref100 abadi (ref82) 2016 ref101 kingma (ref36) 2013 ref34 ref31 ref148 ref30 ref146 ref32 hong (ref92) 2021; 2021 hinton (ref150) 2015 yu (ref173) 2019 ref156 staudemeyer (ref35) 2019 ref154 ref151 paszke (ref83) 2019; 32 ref152 goodfellow (ref37) 2014; 27 ref26 ref25 ref159 ref22 ref157 ref21 ref158 ref28 ref27 jiang (ref136) 2021 ref29 wang (ref94) 2019 bahdanau (ref40) 2015 ref166 ref167 ref164 shlezinger (ref86) 2020 ref165 ref162 ref163 ref160 ref161 li (ref168) 2022 ref13 ref12 ref128 ref15 ref129 ref14 ref126 ref127 ref96 ref124 ref99 ref11 ref125 ref98 ref10 ballé (ref149) 2018 ref17 ref16 ref19 ref18 campbell (ref176) 2018 ref133 ref93 ref134 ref131 ref95 ref132 ref130 ref91 ref90 ref89 ref139 ref137 iandola (ref155) 2016 ref138 ref85 (ref24) 2021 ref135 gregor (ref87) 2010 ref144 ref145 ref81 ref142 ref84 ref143 ref140 ref80 ref79 ref108 ref78 vaswani (ref39) 2017; 30 ref75 ref104 ref74 ref105 ref77 ref102 ref76 ref103 ref2 ref1 ahmed (ref172) 2019 ref71 ref111 ref70 ref112 ref73 ref72 ha (ref106) 2016 ref110 latva-aho (ref107) 2020 guo (ref97) 2020 shih (ref43) 2018 ref68 ref119 ref67 ref117 ref69 ref118 ref64 ref115 ref63 ref116 ref66 ref113 ref65 ref114 xu (ref41) 2015 hussien (ref50) 2020 ref60 ref122 ref123 ref62 ref120 ref61 ref121 liu (ref88) 2019
References_xml	– ident: ref156 doi: 10.1109/CVPR.2018.00716 – start-page: 1 year: 2015 ident: ref40 article-title: Neural machine translation by jointly learning to align and translate publication-title: Proc ICLR – ident: ref139 doi: 10.1109/TCOMM.2021.3138097 – ident: ref118 doi: 10.1007/s41745-020-00169-2 – ident: ref151 doi: 10.23919/JCC.2021.11.008 – ident: ref65 doi: 10.1109/TVT.2020.3004842 – ident: ref72 doi: 10.1049/ell2.12080 – ident: ref30 doi: 10.1016/j.patcog.2019.01.006 – ident: ref89 doi: 10.1109/18.382009 – ident: ref152 doi: 10.23919/JCIN.2020.9200894 – ident: ref53 doi: 10.1109/LWC.2021.3083331 – ident: ref67 doi: 10.4236/jcc.2021.910009 – ident: ref154 doi: 10.1109/LCOMM.2021.3076504 – ident: ref5 doi: 10.1109/MCOMSTD.0001.2200001 – ident: ref145 doi: 10.1109/TSP.2010.2090346 – ident: ref113 doi: 10.3390/e24040441 – ident: ref11 doi: 10.1109/ACCESS.2018.2850226 – ident: ref26 doi: 10.1109/TIT.2006.883550 – ident: ref44 doi: 10.1109/TWC.2020.2968430 – year: 2018 ident: ref43 publication-title: Study on Massive MIMO CSI Feedback Based on Deep Learning – ident: ref74 doi: 10.1109/LCOMM.2020.2981448 – ident: ref159 doi: 10.23919/JCC.2022.08.002 – ident: ref102 doi: 10.1109/LWC.2020.2964550 – ident: ref31 doi: 10.23915/distill.00021 – ident: ref100 doi: 10.1109/26.380120 – ident: ref133 doi: 10.1109/SPAWC.2018.8446005 – ident: ref125 doi: 10.1109/TWC.2022.3182216 – ident: ref128 doi: 10.1109/LWC.2021.3118923 – ident: ref144 doi: 10.1109/TIT.2007.909170 – year: 2013 ident: ref36 article-title: Auto-encoding variational Bayes publication-title: arXiv 1312 6114 – ident: ref179 doi: 10.1109/MWC.011.2100016 – ident: ref6 doi: 10.1109/MCOM.001.2001187 – ident: ref2 doi: 10.1109/JSAC.2017.2692307 – ident: ref112 doi: 10.1109/JSAC.2020.3041397 – ident: ref12 doi: 10.1109/TCOMM.2019.2960361 – start-page: 1 year: 2019 ident: ref172 article-title: DeepMIMO: A generic deep learning dataset for millimeter wave and massive MIMO applications publication-title: Proc Inf Theory Appl Workshop (ITA) – year: 2020 ident: ref86 article-title: Model-based deep learning publication-title: arXiv 2012 08405 – ident: ref120 doi: 10.1109/TWC.2021.3073309 – year: 2016 ident: ref155 article-title: SqueezeNet: AlexNet-level accuracy with 50X fewer parameters and <0.5 MB model size publication-title: arXiv 1602 07360 – ident: ref101 doi: 10.1109/TWC.2021.3103120 – start-page: 3 year: 2018 ident: ref38 article-title: CBAM: Convolutional block attention module publication-title: Proc Eur Conf Comput Vis (ECCV) – ident: ref59 doi: 10.1109/LWC.2022.3149416 – start-page: 698 year: 2018 ident: ref176 article-title: Bayesian coreset construction via greedy iterative geodesic ascent publication-title: Proc ICML – ident: ref121 doi: 10.1109/LWC.2019.2942908 – ident: ref180 doi: 10.1109/MWC.001.1900157 – ident: ref96 doi: 10.1109/TCOMM.2021.3086525 – ident: ref91 doi: 10.1109/LCOMM.2018.2882829 – ident: ref103 doi: 10.1109/ISCEIC53685.2021.00056 – ident: ref132 doi: 10.1109/VTC2021-Fall52928.2021.9625047 – year: 2016 ident: ref106 article-title: HyperNetworks publication-title: arXiv 1609 09106 – ident: ref166 doi: 10.1109/ICNP49622.2020.9259366 – ident: ref160 doi: 10.1109/LWC.2021.3099136 – start-page: 1 year: 2019 ident: ref173 article-title: Slimmable neural networks publication-title: Proc ICLR – ident: ref1 doi: 10.1109/MCOMSTD.001.1800036 – ident: ref66 doi: 10.1109/LWC.2021.3064963 – ident: ref105 doi: 10.1109/TCOMM.2020.2993626 – start-page: 2048 year: 2015 ident: ref41 article-title: Show, attend and tell: Neural image caption generation with visual attention publication-title: Proc ICML – ident: ref80 doi: 10.1109/CVPR42600.2020.01009 – ident: ref18 doi: 10.1109/JSTSP.2017.2784180 – start-page: 1 year: 2018 ident: ref149 article-title: Variational image compression with a scale hyperprior publication-title: Proc ICLR – ident: ref165 doi: 10.1109/JCN.2020.000016 – volume: 2021 start-page: 1 year: 2021 ident: ref92 article-title: Machine learning-based adaptive CSI feedback interval publication-title: ICT Exp – year: 2022 ident: ref168 article-title: Multi-task learning-based CSI feedback design in multiple scenarios publication-title: arXiv 2204 12698 – ident: ref8 doi: 10.1109/CC.2017.8233654 – start-page: 1198 year: 2021 ident: ref136 article-title: Federated learning-based codebook design for massive MIMO communication system publication-title: Proc Int Conf Natural Comput Fuzzy Syst Knowl Discovery – ident: ref32 doi: 10.1109/CVPR52688.2022.01166 – volume: 32 start-page: 1 year: 2019 ident: ref83 article-title: PyTorch: An imperative style, high-performance deep learning library publication-title: Proc NeurIPS – ident: ref69 doi: 10.1109/TCCN.2021.3119945 – ident: ref21 doi: 10.1109/JSAC.2008.081002 – year: 2020 ident: ref107 publication-title: Key drivers and research challenges for 6G ubiquitous wireless intelligence – ident: ref52 doi: 10.1109/LWC.2021.3100493 – ident: ref171 doi: 10.1109/TWC.2022.3149946 – ident: ref29 doi: 10.1137/080716542 – ident: ref169 doi: 10.1109/JIOT.2021.3139958 – ident: ref95 doi: 10.1109/SPAWC51858.2021.9593238 – ident: ref123 doi: 10.1109/GLOBECOM46510.2021.9685912 – ident: ref7 doi: 10.1109/MWC.2019.1800601 – volume: 27 start-page: 139 year: 2014 ident: ref37 article-title: Generative adversarial nets publication-title: Proc 28th Adv Neural Inf Process Syst (NIPS) – ident: ref174 doi: 10.1109/MWC.006.2100543 – ident: ref34 doi: 10.1162/neco.1997.9.8.1735 – volume: 62 start-page: 3261 year: 2014 ident: ref20 article-title: Distributed compressive CSIT estimation and feedback for FDD multi-user massive MIMO systems publication-title: IEEE Trans Signal Process doi: 10.1109/TSP.2014.2324991 – ident: ref135 doi: 10.1109/TCCN.2021.3084409 – ident: ref178 doi: 10.1109/ACCESS.2019.2901066 – ident: ref126 doi: 10.1109/LWC.2021.3117032 – ident: ref25 doi: 10.1109/LWC.2018.2818160 – ident: ref124 doi: 10.1109/WCNC49053.2021.9417115 – ident: ref115 doi: 10.1109/LWC.2019.2962114 – ident: ref127 doi: 10.1109/LCOMM.2020.2989499 – ident: ref70 doi: 10.1109/LWC.2021.3085317 – ident: ref76 doi: 10.1109/CVPR.2016.90 – start-page: 265 year: 2016 ident: ref82 article-title: TensorFlow: A system for large-scale machine learning publication-title: Proc 12th OSDI – start-page: 1 year: 2019 ident: ref88 article-title: ALISTA: Analytic weights are as good as learned weights in LISTA publication-title: Proc ICLR – ident: ref157 doi: 10.1109/TPAMI.2017.2699184 – ident: ref47 doi: 10.1109/ACCESS.2020.2963896 – ident: ref13 doi: 10.1109/LWC.2017.2757490 – ident: ref71 doi: 10.1109/WOCC53213.2021.9602863 – ident: ref114 doi: 10.1109/GLOBECOM42002.2020.9322121 – year: 2022 ident: ref182 publication-title: Wireless Intelligence – year: 1988 ident: ref147 publication-title: Pulse Code Modulation (PCM) of Voice Frequencies – ident: ref111 doi: 10.1109/TCOMM.2021.3120294 – ident: ref122 doi: 10.1109/GlobalSIP45357.2019.8969557 – year: 2022 ident: ref109 article-title: Deep learning for joint channel estimation and feedback in massive MIMO systems publication-title: arXiv 2011 07242 – ident: ref137 doi: 10.1109/JSTSP.2022.3160268 – ident: ref177 doi: 10.1109/MWC.010.2200304 – ident: ref162 doi: 10.1109/ACCESS.2019.2928049 – ident: ref138 doi: 10.1109/LCOMM.2022.3185308 – ident: ref90 doi: 10.1109/LWC.2018.2874264 – ident: ref81 doi: 10.1109/CVPR.2017.243 – ident: ref54 doi: 10.1109/TVT.2022.3183596 – ident: ref140 doi: 10.1109/LWC.2021.3112747 – ident: ref85 doi: 10.1109/MWC.2019.1800447 – ident: ref22 doi: 10.1109/MSP.2018.2789521 – ident: ref55 doi: 10.1109/LWC.2021.3092947 – ident: ref175 doi: 10.1109/LWC.2021.3140102 – ident: ref77 doi: 10.1109/ICCV.2019.00200 – ident: ref58 doi: 10.1109/WCSP52459.2021.9613524 – year: 2019 ident: ref35 article-title: Understanding LSTM-A tutorial into long short-term memory recurrent neural networks publication-title: arXiv 1909 09586 – ident: ref4 doi: 10.1109/MWC.2021.9363048 – ident: ref163 doi: 10.1109/LCOMM.2007.061534 – ident: ref116 doi: 10.1109/TVT.2021.3110608 – year: 2021 ident: ref24 publication-title: New SI Study on Artificial Intelligence (AI)/Machine Learning (ML) for NR Air Interface – ident: ref42 doi: 10.1088/1742-6596/1693/1/012173 – ident: ref17 doi: 10.1109/TWC.2020.2970707 – year: 2022 ident: ref153 article-title: Deep joint source-channel coding for CSI feedback: An end-to-end approach publication-title: arXiv 2203 16005 – ident: ref181 doi: 10.1109/MWC.001.2100136 – ident: ref131 doi: 10.1109/TWC.2022.3141653 – ident: ref57 doi: 10.1109/LCOMM.2021.3116864 – ident: ref78 doi: 10.1049/iet-com.2019.1030 – ident: ref130 doi: 10.1109/MWC.001.1900473 – year: 2019 ident: ref94 publication-title: Research on key technology of massive MIMO channel feedback for intelligent communications – volume: 30 start-page: 1 year: 2017 ident: ref39 article-title: Attention is all you need publication-title: Proc 31st Adv Neural Inf Process Syst (NIPS) – ident: ref117 doi: 10.1109/TWC.2021.3055202 – ident: ref46 doi: 10.1109/ICC40277.2020.9149229 – ident: ref68 doi: 10.1109/MLSP.2019.8918798 – ident: ref27 doi: 10.1109/TIT.2005.858979 – ident: ref49 doi: 10.1109/LCOMM.2020.3017188 – ident: ref64 doi: 10.1109/VTC2021-Fall52928.2021.9625585 – ident: ref63 doi: 10.1109/LWC.2022.3216352 – ident: ref14 doi: 10.1109/LCOMM.2018.2877965 – ident: ref110 doi: 10.1109/LWC.2020.3017753 – ident: ref170 doi: 10.1109/MWC.2012.6393523 – ident: ref134 doi: 10.1109/WCNC49053.2021.9417500 – ident: ref15 doi: 10.1109/TWC.2021.3087191 – ident: ref60 doi: 10.1109/LWC.2022.3153085 – ident: ref164 doi: 10.1109/LCOMM.2022.3145099 – ident: ref75 doi: 10.1109/LCOMM.2021.3138927 – year: 2022 ident: ref141 article-title: Attention mechanism based intelligent channel feedback for mmWave massive MIMO systems publication-title: arXiv 2208 06570 – ident: ref61 doi: 10.1109/LWC.2022.3157263 – ident: ref73 doi: 10.1109/LWC.2021.3122462 – year: 2022 ident: ref23 publication-title: Release 17 Description Summary of Rel-17 Work Items (v1 0 0) – ident: ref158 doi: 10.1109/TCOMM.2020.3019077 – ident: ref108 doi: 10.1109/TVT.2021.3131606 – year: 2020 ident: ref97 article-title: DL-based CSI feedback and cooperative recovery in massive MIMO publication-title: arXiv 2003 03303 – ident: ref10 doi: 10.1109/ACCESS.2022.3208284 – ident: ref3 doi: 10.1109/ACCESS.2019.2939938 – ident: ref104 doi: 10.1109/LWC.2021.3096808 – ident: ref142 doi: 10.1109/TCOMM.2022.3180388 – ident: ref143 doi: 10.1109/MSP.2008.930649 – ident: ref19 doi: 10.1109/JSTSP.2014.2317671 – ident: ref45 doi: 10.1109/WCNC.2019.8885897 – ident: ref79 doi: 10.1109/LWC.2021.3057934 – ident: ref62 doi: 10.1109/ACCESS.2022.3194035 – ident: ref93 doi: 10.1109/LWC.2019.2898662 – ident: ref84 doi: 10.1109/23.589532 – ident: ref129 doi: 10.1109/ICASSP40776.2020.9053850 – year: 2020 ident: ref50 article-title: PRVNet: A novel partially-regularized variational autoencoders for massive MIMO CSI feedback publication-title: arXiv 2011 04178 [cs] – ident: ref28 doi: 10.1016/j.crma.2008.03.014 – ident: ref161 doi: 10.1109/BigData47090.2019.9006216 – start-page: 1 year: 2015 ident: ref150 article-title: Distilling the knowledge in a neural network publication-title: Proc Adv Neural Inf Process Syst Workshops (NIPSW) – ident: ref98 doi: 10.1109/TWC.2020.3043502 – year: 2015 ident: ref33 publication-title: Understanding LSTM Networks – ident: ref51 doi: 10.23919/JCC.2021.01.004 – ident: ref146 doi: 10.1109/LWC.2019.2895039 – ident: ref99 doi: 10.1109/TWC.2022.3160498 – ident: ref56 doi: 10.1109/LCOMM.2021.3099841 – ident: ref119 doi: 10.1109/TVT.2020.2980905 – ident: ref48 doi: 10.1109/LCOMM.2020.3019653 – start-page: 399 year: 2010 ident: ref87 article-title: Learning fast approximations of sparse coding publication-title: Proc 27th Int Conf Mach Learn – ident: ref148 doi: 10.1109/TCSVT.2003.815173 – ident: ref167 doi: 10.1109/LCOMM.2021.3123941 – ident: ref16 doi: 10.1109/TCCN.2017.2758370 – ident: ref9 doi: 10.1016/j.dcan.2021.09.014
SSID	ssj0004033
Score	2.714227
Snippet	Many performance gains achieved by massive multiple-input and multiple-output depend on the accuracy of the downlink channel state information (CSI) at the...
SourceID	proquest crossref ieee
SourceType	Aggregation Database Enrichment Source Index Database Publisher
StartPage	1
SubjectTerms	3GPP Communication Computer architecture CSI feedback Decoding Deep learning Downlink Feedback Image coding Image reconstruction Indexes Massive MIMO Neural networks overview State-of-the-art reviews Training Wireless communication systems
Title	Overview of Deep Learning-based CSI Feedback in Massive MIMO Systems
URI	https://ieeexplore.ieee.org/document/9931713 https://www.proquest.com/docview/2754956792
Volume	70
WOSCitedRecordID	wos000927591900022&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
journalDatabaseRights	– providerCode: PRVIEE databaseName: IEEE/IET Electronic Library (IEL) (UW System Shared) customDbUrl: eissn: 1558-0857 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0004033 issn: 0090-6778 databaseCode: RIE dateStart: 19720101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwELZKxQADr4IoFOSBDdImjp3HiFoqOqRFokjdItu5IARKq774-5xTt4BASGTycI6Sz_E9nLvvCLnCeFqY3icO2sLI4Zp7TiTNSRw69wr9fcPxXTabCPv9aDSKHyrkZlMLAwBl8hk0zbD8l5-N9cIclbXQlnqhaVG7FYbBqlbrswbS9S3jpElnD6N1gYwbt4btQZJgKMhY00cPHK9vRqjsqvJDFZf2pbv_vyc7IHvWj6S3q4U_JBUojsjuF3bBGukMlkYRwDsd57QDMKGWTPXZMbYro-3HHu2i9VJSv9KXgiboSKPyo0kvGVBLZX5Mnrp3w_a9Y5smOJrFYu4w8GWklBZKAMehRJ9NBkrnvvLBYx4IL0PNolEJZnmQSRYCD_B9uA_Sc1HohFSLcQGnhIKbMS5UkJtNK3mkuIZci4xrFAwgqBNvjWKqLaO4aWzxlpaRhRunJfKpQT61yNfJ9WbOZMWn8ad0zWC9kbQw10ljvVip3XKzlIXCBHthzM5-n3VOdsy9V7koDVKdTxdwQbb1cv4ym16WX9MHTlzD-g
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LS8NAEB6KCurBVxXrcw_eNJpsdvM4Smux2LSCFbyF7GYiRWlF2_r3nU23VVEEc9rDLEm-zc5jM_MNwAnF09L0PnHIFkaO0MJzosycxJFzr8jfNxzfZbOJsNOJHh7i2wqczWthELFMPsNzMyz_5edDPTZHZRdkS73QtKhdlEJwd1qt9VkF6fqWc9IktIfRrETGjS969W6SUDDI-blPPjhd38xQ2VflhzIuLUxz_X_PtgFr1pNkl9Ol34QKDrZg9Qu_YBUa3YlRBfjOhgVrIL4wS6f66BjrlbP6XYs1yX6pTD-x_oAl5EqT-mNJK-kyS2a-DffNq1792rFtExzNYzlyOPpZpJSWSqKgYUZeWxYoXfjKR497KL2cdIsmNZgXQZ7xEEVA7yN8zDyXhHZgYTAc4C4wdHMupAoKs20zESmhsdAyF5oEAwxq4M1QTLXlFDetLZ7TMrZw47REPjXIpxb5GpzO57xMGTX-lK4arOeSFuYaHMwWK7Wb7i3loTThXhjzvd9nHcPydS9pp-1W52YfVsx9ppkpB7Aweh3jISzpyaj_9npUflkfq_THQQ
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%3Ajournal&rft.genre=article&rft.atitle=Overview+of+Deep+Learning-Based+CSI+Feedback+in+Massive+MIMO+Systems&rft.jtitle=IEEE+transactions+on+communications&rft.au=Guo%2C+Jiajia&rft.au=Chao-Kai%2C+Wen&rft.au=Shi%2C+Jin&rft.au=Geoffrey+Ye+Li&rft.date=2022-12-01&rft.pub=The+Institute+of+Electrical+and+Electronics+Engineers%2C+Inc.+%28IEEE%29&rft.issn=0090-6778&rft.eissn=1558-0857&rft.volume=70&rft.issue=12&rft.spage=8017&rft_id=info:doi/10.1109%2FTCOMM.2022.3217777&rft.externalDBID=NO_FULL_TEXT
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0090-6778&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0090-6778&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0090-6778&client=summon