Edge Intelligence: Empowering Intelligence to the Edge of Network

Edge intelligence refers to a set of connected systems and devices for data collection, caching, processing, and analysis proximity to where data are captured based on artificial intelligence. Edge intelligence aims at enhancing data processing and protects the privacy and security of the data and u...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings of the IEEE Jg. 109; H. 11; S. 1778 - 1837
Hauptverfasser: Xu, Dianlei, Li, Tong, Li, Yong, Su, Xiang, Tarkoma, Sasu, Jiang, Tao, Crowcroft, Jon, Hui, Pan
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York IEEE 01.11.2021
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Artificial intelligence
Artificial intelligence (AI)
Caching
Data collection
Data privacy
Data processing
edge caching
Edge computing
Electronic devices
inference
Inference algorithms
model training
offloading
Systematics
Taxonomy
Training data
ISSN:0018-9219, 1558-2256
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Edge intelligence refers to a set of connected systems and devices for data collection, caching, processing, and analysis proximity to where data are captured based on artificial intelligence. Edge intelligence aims at enhancing data processing and protects the privacy and security of the data and users. Although recently emerged, spanning the period from 2011 to now, this field of research has shown explosive growth over the past five years. In this article, we present a thorough and comprehensive survey of the literature surrounding edge intelligence. We first identify four fundamental components of edge intelligence, i.e., edge caching, edge training, edge inference, and edge offloading based on theoretical and practical results pertaining to proposed and deployed systems. We then aim for a systematic classification of the state of the solutions by examining research results and observations for each of the four components and present a taxonomy that includes practical problems, adopted techniques, and application goals. For each category, we elaborate, compare, and analyze the literature from the perspectives of adopted techniques, objectives, performance, advantages and drawbacks, and so on. This article provides a comprehensive survey of edge intelligence and its application areas. In addition, we summarize the development of the emerging research fields and the current state of the art and discuss the important open issues and possible theoretical and technical directions.
AbstractList Edge intelligence refers to a set of connected systems and devices for data collection, caching, processing, and analysis proximity to where data are captured based on artificial intelligence. Edge intelligence aims at enhancing data processing and protects the privacy and security of the data and users. Although recently emerged, spanning the period from 2011 to now, this field of research has shown explosive growth over the past five years. In this article, we present a thorough and comprehensive survey of the literature surrounding edge intelligence. We first identify four fundamental components of edge intelligence, i.e., edge caching, edge training, edge inference, and edge offloading based on theoretical and practical results pertaining to proposed and deployed systems. We then aim for a systematic classification of the state of the solutions by examining research results and observations for each of the four components and present a taxonomy that includes practical problems, adopted techniques, and application goals. For each category, we elaborate, compare, and analyze the literature from the perspectives of adopted techniques, objectives, performance, advantages and drawbacks, and so on. This article provides a comprehensive survey of edge intelligence and its application areas. In addition, we summarize the development of the emerging research fields and the current state of the art and discuss the important open issues and possible theoretical and technical directions.
Author Li, Tong
Su, Xiang
Xu, Dianlei
Crowcroft, Jon
Li, Yong
Jiang, Tao
Hui, Pan
Tarkoma, Sasu
Author_xml – sequence: 1
  givenname: Dianlei
  orcidid: 0000-0002-9091-5129
  surname: Xu
  fullname: Xu, Dianlei
  email: dianlei.xu@helsinki.fi
  organization: Department of Computer Science, University of Helsinki, Helsinki, Finland
– sequence: 2
  givenname: Tong
  orcidid: 0000-0002-4343-703X
  surname: Li
  fullname: Li, Tong
  email: t.li@connect.ust.hk
  organization: Department of Computer Science, University of Helsinki, Helsinki, Finland
– sequence: 3
  givenname: Yong
  orcidid: 0000-0001-5617-1659
  surname: Li
  fullname: Li, Yong
  email: liyong07@tsinghua.edu.cn
  organization: Beijing National Research Center for Information Science and Technology (BNRist), Department of Electronic Engineering, Tsinghua University, Beijing 100084, China
– sequence: 4
  givenname: Xiang
  orcidid: 0000-0003-1342-6759
  surname: Su
  fullname: Su, Xiang
  email: xiang.su@ntnu.no
  organization: Department of Computer Science, Norwegian University of Science and Technology, Trondheim, Norway
– sequence: 5
  givenname: Sasu
  orcidid: 0000-0003-4220-3650
  surname: Tarkoma
  fullname: Tarkoma, Sasu
  email: sasu.tarkoma@helsinki.fi
  organization: Department of Computer Science, University of Helsinki, Helsinki, Finland
– sequence: 6
  givenname: Tao
  surname: Jiang
  fullname: Jiang, Tao
  email: taojiang@ieee.org
  organization: School of Electronics Information and Communications, Huazhong University of Science and Technology, Wuhan, China
– sequence: 7
  givenname: Jon
  orcidid: 0000-0002-7013-0121
  surname: Crowcroft
  fullname: Crowcroft, Jon
  email: jon.crowcroft@cl.cam.ac.uk
  organization: Computer Laboratory, University of Cambridge, Cambridge, U.K
– sequence: 8
  givenname: Pan
  orcidid: 0000-0002-0848-2599
  surname: Hui
  fullname: Hui, Pan
  email: panhui@cse.ust.hk
  organization: Department of Computer Science, University of Helsinki, Helsinki, Finland
BookMark eNp9kE1Lw0AURQepYFv9A7oJuE59M5OXZNyVUrVSrIiuhyR9U1PTTJ1MKf570w8EXbh6i3vPfXB6rFPbmhi75DDgHNTN4_PLbDQQIPhAcq4UwgnrcsQ0FALjDusC8DRUgqsz1muaJQBIjGWXDcfzBQWT2lNVlQuqC7oNxqu13ZIr68WvIPA28O8U7AlrgifyW-s-ztmpyaqGLo63z97uxq-jh3A6u5-MhtOwEAp9mCPOIxQIxiDnfG6kiRKIRZRIynJAxLjNI8SEK2Mgz6MdB5EwRZZCkcs-uz7srp393FDj9dJuXN2-1AKVaJeUTNpWemgVzjaNI6OL0me-tLV3WVlpDnonTO-F6Z0wfRTWouIPunblKnNf_0NXB6gkoh9AoYpjDvIb_id3Bw
CODEN IEEPAD
CitedBy_id crossref_primary_10_1016_j_hcc_2023_100150
crossref_primary_10_1016_j_suscom_2025_101132
crossref_primary_10_1109_ACCESS_2024_3521579
crossref_primary_10_1109_MC_2022_3203613
crossref_primary_10_1080_15623599_2025_2537727
crossref_primary_10_1016_j_scib_2023_11_042
crossref_primary_10_1145_3663482
crossref_primary_10_1109_JSYST_2024_3442958
crossref_primary_10_3390_fi16110421
crossref_primary_10_1109_TSC_2024_3451170
crossref_primary_10_1007_s11356_023_31779_9
crossref_primary_10_3390_s22239404
crossref_primary_10_1109_TPDS_2024_3413751
crossref_primary_10_1145_3639825
crossref_primary_10_1109_TMC_2025_3567459
crossref_primary_10_1109_TCE_2024_3378509
crossref_primary_10_3390_electronics13183763
crossref_primary_10_1109_TMM_2024_3521770
crossref_primary_10_3390_fi17040175
crossref_primary_10_1109_COMST_2024_3482978
crossref_primary_10_4018_IJeC_388554
crossref_primary_10_1007_s11432_024_4463_0
crossref_primary_10_1016_j_ins_2024_120259
crossref_primary_10_3390_mi14050897
crossref_primary_10_3390_electronics14122401
crossref_primary_10_1016_j_adhoc_2025_103916
crossref_primary_10_1016_j_comnet_2024_110756
crossref_primary_10_1109_TCC_2023_3258982
crossref_primary_10_1109_TNSE_2025_3551148
crossref_primary_10_1109_JIOT_2025_3567545
crossref_primary_10_3390_math12152383
crossref_primary_10_1109_MIS_2022_3181504
crossref_primary_10_1109_TNSM_2024_3416947
crossref_primary_10_1186_s13677_023_00493_9
crossref_primary_10_1051_itmconf_20246503015
crossref_primary_10_3390_fi17040140
crossref_primary_10_1109_JIOT_2023_3317785
crossref_primary_10_1109_TMC_2023_3349315
crossref_primary_10_1109_JIOT_2025_3529185
crossref_primary_10_1109_JIOT_2025_3572900
crossref_primary_10_1109_TNSM_2025_3561739
crossref_primary_10_1109_COMST_2023_3244674
crossref_primary_10_1109_TVT_2023_3265479
crossref_primary_10_1016_j_jksuci_2022_12_013
crossref_primary_10_1109_JPROC_2022_3226481
crossref_primary_10_1109_ACCESS_2024_3384233
crossref_primary_10_1109_TVT_2024_3518836
crossref_primary_10_1109_COMST_2022_3211462
crossref_primary_10_1145_3657283
crossref_primary_10_1016_j_ins_2023_119169
crossref_primary_10_1016_j_future_2024_107588
crossref_primary_10_1145_3700792
crossref_primary_10_1109_ACCESS_2024_3421652
crossref_primary_10_1109_JPROC_2024_3437365
crossref_primary_10_1109_TSC_2022_3187118
crossref_primary_10_1109_ACCESS_2022_3183634
crossref_primary_10_1109_TCC_2025_3591482
crossref_primary_10_1109_TMC_2023_3329444
crossref_primary_10_1109_TMC_2025_3556243
crossref_primary_10_1016_j_future_2022_10_006
crossref_primary_10_1109_MNET_2024_3469988
crossref_primary_10_1049_ipr2_12802
crossref_primary_10_32604_cmes_2023_023684
crossref_primary_10_1007_s11227_025_07295_7
crossref_primary_10_1109_IOTM_001_2300169
crossref_primary_10_1109_TMC_2024_3361876
crossref_primary_10_1002_cpe_70269
crossref_primary_10_3390_info14070414
crossref_primary_10_3390_electronics12010001
crossref_primary_10_3390_fi16100373
crossref_primary_10_1049_cth2_12761
crossref_primary_10_1109_JIOT_2025_3546577
crossref_primary_10_1109_LWC_2024_3412674
crossref_primary_10_1016_j_compeleceng_2025_110146
crossref_primary_10_1109_TCOMM_2024_3418901
crossref_primary_10_1007_s00779_024_01821_9
crossref_primary_10_3390_bdcc7010044
crossref_primary_10_1109_ACCESS_2024_3425350
crossref_primary_10_3390_w17152269
crossref_primary_10_3390_diagnostics14232708
crossref_primary_10_1016_j_adhoc_2024_103751
crossref_primary_10_1145_3724420
crossref_primary_10_3390_s25072089
crossref_primary_10_1109_MNET_2024_3380647
crossref_primary_10_1109_JIOT_2023_3340298
crossref_primary_10_1109_TCSII_2022_3161717
crossref_primary_10_1007_s11831_023_10032_z
crossref_primary_10_1109_ACCESS_2024_3459925
crossref_primary_10_1109_JIOT_2023_3280069
crossref_primary_10_1109_MWC_006_2100699
crossref_primary_10_1109_TITS_2023_3241479
crossref_primary_10_1109_TCCN_2024_3439589
crossref_primary_10_1109_TSUSC_2024_3521401
crossref_primary_10_1016_j_comnet_2024_110624
crossref_primary_10_1016_j_patcog_2025_111831
crossref_primary_10_1109_TMC_2025_3554816
crossref_primary_10_3390_jsan11030047
crossref_primary_10_1109_TII_2022_3154473
crossref_primary_10_1016_j_future_2024_107691
crossref_primary_10_1109_ACCESS_2024_3469956
crossref_primary_10_1007_s13198_025_02940_z
crossref_primary_10_1007_s11276_025_03903_9
crossref_primary_10_1109_COMST_2022_3163176
crossref_primary_10_1063_5_0194194
crossref_primary_10_1007_s10844_024_00909_1
crossref_primary_10_1016_j_procs_2025_04_204
crossref_primary_10_1109_JIOT_2023_3281603
crossref_primary_10_1109_JIOT_2022_3160691
Cites_doi 10.1109/MSP.2007.4286571
10.1109/TVT.2020.2973651
10.1007/978-3-319-69462-7_27
10.1109/ACCESS.2017.2706741
10.1109/TWC.2020.2971981
10.1109/TCSI.2020.2966243
10.1145/3194658.3194666
10.1109/TITS.2014.2311123
10.1007/978-3-319-46448-0_2
10.1007/978-3-030-01249-6_18
10.1145/3298981
10.1109/JIOT.2021.3056185
10.1007/s11222-007-9033-z
10.1109/MCOM.2014.6736753
10.1109/WACV.2014.6836013
10.1609/aimag.v35i4.2513
10.1109/LCOMM.2018.2875978
10.3390/jlpea7020010
10.1007/s11277-018-5360-y
10.1145/1150402.1150464
10.1109/TSC.2020.3032724
10.1109/TNSE.2020.3033938
10.1109/TIM.2020.3018831
10.1109/ICDCS.2017.94
10.1109/ICDCS.2018.00154
10.1109/JIOT.2021.3049173
10.1109/PERCOMW.2019.8730706
10.1109/ICNSC.2019.8743164
10.3390/fi9040066
10.1109/TITS.2018.2877059
10.1145/3363347.3363366
10.1109/JSTSP.2020.2977090
10.1007/s11265-006-4190-4
10.1109/ICNP.2017.8117585
10.1145/3089801.3089804
10.1145/1536414.1536440
10.1109/TPAMI.2018.2884462
10.1145/2968219.2971461
10.1109/TII.2019.2900665
10.1109/ICPR.2018.8546129
10.1109/TII.2018.2842821
10.1145/2742647.2742663
10.1006/ijhc.2001.0499
10.1109/SiPS.2016.31
10.1109/ACCESS.2018.2846609
10.1109/TII.2018.2877217
10.1145/2938559.2949718
10.1109/INFCOM.2012.6195845
10.1109/HPCA.2018.00018
10.1109/JSSC.2016.2616357
10.1016/j.adhoc.2018.05.014
10.1109/VNC.2017.8275621
10.1109/PerCom45495.2020.9127360
10.1109/ACCESS.2016.2607719
10.1109/MNET.2018.1700202
10.1109/VTS-APWCS.2019.8851649
10.1109/WoWMoM.2019.8792973
10.1137/0718026
10.1109/CVPR.2017.195
10.1145/3123266.3123453
10.1098/rsta.2019.0155
10.1109/VNC.2016.7835943
10.1145/3081333.3081336
10.1145/2030112.2030164
10.1145/3081333.3081358
10.1007/s00521-010-0362-z
10.1145/2906388.2906395
10.1186/s40537-014-0008-6
10.1109/MVT.2018.2883777
10.1109/NCA.2017.8171350
10.1109/EMC2.2018.00012
10.1002/nem.2146
10.1109/TPAMI.2013.50
10.1109/PDP2018.2018.00023
10.1109/JIOT.2020.3037194
10.1038/nbt1206-1565
10.23919/JCC.2020.09.008
10.1145/3081333.3081347
10.1109/MSP.2006.1657817
10.1016/j.future.2019.01.059
10.1109/MCC.2018.1081070
10.1145/2594368.2594388
10.1109/WACV45572.2020.9093331
10.1109/MedHocNet.2012.6257116
10.1145/2964284.2967280
10.1109/MCOM.2017.1600863
10.1109/CAC.2017.8243585
10.1109/COMST.2018.2844341
10.1109/TVT.2013.2291811
10.1109/BigData.2018.8622396
10.1007/s11042-016-3316-3
10.1109/ICCV.2019.00140
10.1109/IC2E48712.2020.00009
10.1109/SocialSens.2018.00017
10.1016/j.agsy.2017.01.023
10.1016/j.acha.2019.06.004
10.1109/TITS.2019.2906038
10.1109/MC.2017.3641638
10.1109/INFOCOM.2015.7218644
10.1109/TWC.2019.2961673
10.1145/3243734.3243781
10.1109/TCSVT.2019.2903421
10.1109/ICC40277.2020.9148862
10.1109/TCOMM.2019.2956472
10.1145/3338501.3357371
10.1109/ACCESS.2018.2810264
10.1145/3132211.3134457
10.1109/ICCE.2016.7430525
10.1109/CVPR.2018.00745
10.1145/3378679.3394533
10.1016/j.adhoc.2016.02.002
10.1109/TITS.2020.3040557
10.1109/TII.2019.2952645
10.1109/AVSS.2018.8639121
10.1145/3196494.3196532
10.1145/3131895
10.1145/3173162.3173185
10.1109/CVPR.2016.521
10.1109/INFOCOM.2016.7524381
10.1109/SEC.2016.38
10.1145/3359993.3366644
10.1145/3213344.3213345
10.1145/2906388.2906396
10.1145/3152127
10.1109/JETCAS.2019.2910232
10.1145/2964284.2973801
10.1109/CVPR.2017.634
10.1109/ICIP.2012.6467224
10.1109/ICCSE.2016.7581655
10.1109/TII.2017.2679740
10.1145/3267305.3267505
10.1109/MCC.2016.124
10.1109/TCOMM.2020.2979149
10.1016/j.procs.2015.05.043
10.1007/s00779-016-0932-x
10.1109/JIOT.2016.2579198
10.1109/ICASSP.2013.6638947
10.1145/2976749.2978318
10.1145/3020078.3021736
10.1145/3140582.3081040
10.1109/JIOT.2020.2984887
10.1109/TSC.2017.2662008
10.1109/CVPR.2016.91
10.1109/ICASSP40776.2020.9054634
10.1109/INFOCOM.2018.8485905
10.1145/3397461
10.1109/LCN.2016.039
10.1109/COMST.2020.3007787
10.1109/MNET.001.1900287
10.1016/j.aci.2018.05.002
10.1145/3351260
10.1109/ACCESS.2020.3038287
10.1016/j.phycom.2021.101381
10.1109/MPRV.2017.2940968
10.1145/2619239.2626296
10.1016/j.patcog.2020.107461
10.1145/3089801.3089805
10.1002/nav.3800010110
10.1109/IUCC/DSCI/SmartCNS.2019.00126
10.4108/eai.30-11-2016.2267463
10.1109/TSG.2016.2553647
10.1109/ICECC.2011.6066743
10.1109/TIFS.2015.2400395
10.1109/ACCESS.2018.2843341
10.1109/GLOBECOM38437.2019.9013587
10.1145/3241539.3241565
10.1109/TSE.2017.2685387
10.1109/ICSCC.2019.8843640
10.1145/3126555
10.1016/j.future.2017.03.034
10.1109/TWC.2020.3037554
10.1145/2030613.2030625
10.1109/COMST.2018.2820021
10.1007/978-3-319-46493-0_38
10.1109/ICDCS.2018.00041
10.1109/JIOT.2019.2897005
10.1145/3229556.3229562
10.1109/MVT.2019.2921208
10.1109/IPDPS47924.2020.00033
10.1145/3299710.3211336
10.1145/3126686.3126739
10.1007/s11633-017-1054-2
10.1109/JIOT.2020.2988697
10.1109/CVPR.2016.308
10.2200/S00737ED1V01Y201610AIM033
10.3390/s18051532
10.1109/ICPADS.2017.00069
10.1049/cje.2016.11.016
10.1016/j.compag.2020.105233
10.1038/s41928-018-0059-3
10.3390/s18072203
10.1145/3197231.3197256
10.1109/TMC.2020.3019988
10.1109/JIOT.2020.2979691
10.1007/978-981-10-8339-6_6
10.1109/JIOT.2017.2750180
10.1126/science.aaa8415
10.1109/VNC51378.2020.9318400
10.1109/ICCCBDA51879.2021.9442600
10.1145/3132211.3134459
10.1145/3131672.3131675
10.1145/3109761.3109804
10.3390/s20237000
10.1109/JIOT.2019.2902141
10.1145/2973750.2973777
10.1145/3083187.3084016
10.1145/3210240.3210337
10.1109/MCOM.001.1900103
10.1109/AINA.2015.254
10.1007/BF00992698
10.1109/WACV45572.2020.9093377
10.1145/3093337.3037698
10.5244/C.28.88
10.1109/MNET.011.2000180
10.1109/COMST.2020.2970550
10.1109/ICWS49710.2020.00073
10.1109/TMC.2016.2567378
10.1145/3081333.3089331
10.1007/978-3-319-57959-7
10.1145/3241539.3241557
10.1109/MIS.2020.2988604
10.1109/JIOT.2020.2964162
10.1109/SEC50012.2020.00041
10.1016/j.comcom.2019.12.054
10.1109/ICRA.2019.8794182
10.1109/CVPR.2005.177
10.1109/ICC.2019.8761315
10.1056/NEJMsa1204142
10.1145/3089801.3089803
10.1145/2639189.2670273
10.1109/IPSN.2018.00048
10.1109/CSTIC.2019.8755642
10.1109/COMST.2018.2808242
10.1109/JIOT.2020.2978286
10.1145/3213344.3213351
10.1109/ACCESS.2019.2951587
10.1145/3278721.3278778
10.1145/3097895.3097903
10.1016/j.compeleceng.2017.12.009
10.1016/j.ins.2019.01.014
10.1109/SmartCloud.2016.18
10.1109/MNET.001.1900506
10.1145/3089801.3089806
10.1145/2831347.2831354
10.1145/3220192.3220460
10.1007/978-1-4302-4000-6_9
10.1109/CIT/IUCC/DASC/PICOM.2015.170
10.3390/electronics5040088
10.1109/TWC.2020.2974748
10.1145/1999995.2000000
10.1145/3191752
10.24963/ijcai.2018/429
10.1016/j.jss.2018.03.032
10.1007/978-3-319-46493-0_32
10.1109/ICDCS.2018.00139
10.1109/HotWeb.2015.22
10.1109/TITS.2021.3081560
10.1109/JIOT.2020.3036157
10.1145/2753509.2753518
10.1145/3194554.3194565
10.21437/Interspeech.2016-128
10.1109/ITAIC49862.2020.9338964
10.1109/SMARTCOMP.2018.00087
10.3233/JIFS-169699
10.1109/GLOBECOM38437.2019.9013821
10.1109/NGCAS.2017.16
10.1109/TWC.2020.3003744
10.1145/2594368.2594383
10.1109/IJCNN48605.2020.9207469
10.1109/JSAC.2019.2904348
10.1016/0169-7439(87)80084-9
10.1109/ACCESS.2014.2325029
10.1145/2342509.2342513
10.3390/app7080826
10.1109/EMC2-NIPS53020.2019.00013
10.1016/j.ecolind.2020.107124
10.1016/S0043-1648(00)00427-0
10.1109/TMC.2020.2984364
10.1109/TMC.2018.2815694
10.1109/JPROC.2019.2922285
10.1145/3154503
10.1109/TITS.2020.3002712
10.5244/C.29.41
10.1109/PCCC.2017.8280500
10.1109/LCA.2020.2979965
10.1109/CVPR.2018.00474
10.1145/2906388.2906405
10.1109/PERCOMW.2016.7457169
10.1109/TMC.2016.2607716
10.1145/3152130.3152135
10.1016/j.jpdc.2018.11.006
10.1145/1390156.1390177
10.1007/s11042-020-09740-6
10.1145/2935643.2935650
10.1109/CVPR.2018.00716
10.1109/COMSNETS.2018.8328270
10.1145/2809695.2809711
10.1145/3240508.3240697
10.1145/3373087.3375312
10.1109/T-C.1971.223174
10.1109/TPDS.2020.3016344
10.3390/ijgi8090412
10.3115/v1/D14-1179
10.1109/CFEC.2018.8358733
10.23919/INM.2017.7987297
10.1109/TITS.2020.3023446
10.1109/ICASSP.2015.7178146
10.1109/TCOMM.2020.3019527
10.3390/electronics8080896
10.1145/3212725.3212728
10.1002/cem.873
10.1109/JPROC.2019.2918951
10.1002/wcm.2591
10.1145/3161174
10.1145/3240508.3240561
10.1016/j.pmcj.2017.07.014
10.1109/INFOCOM42981.2021.9488804
10.1109/CVPR.2017.643
10.1162/neco.1997.9.8.1735
10.2196/24207
10.1109/TNSE.2020.3014455
10.1109/IPSN.2016.7460664
10.1109/TVLSI.2018.2825145
10.1609/aaai.v33i01.33014780
10.1109/TVT.2010.2043968
10.1145/2757384.2757397
10.1109/COMST.2019.2904897
10.1109/CVPR.2016.90
10.1109/ICACSIS.2017.8355051
10.1109/PERCOMW.2015.7134002
10.1109/ICCCN.2018.8487352
10.1109/PerCom45495.2020.9127387
10.1109/MNET.2019.1700470
10.1109/IC2E.2018.00042
10.1109/CHASE.2017.115
10.1145/3193025.3193041
10.1145/3161181
10.1109/5.726791
10.1109/ICITBS.2019.00104
10.1145/2633661.2633671
10.1109/GCWkshps45667.2019.9024587
10.1145/2994551.2994564
10.1007/978-3-642-31537-4_13
10.1109/MVT.2018.2848498
10.1145/3131885.3131906
10.1109/MCOM.2014.6736756
10.1080/03071847.2019.1693801
10.1109/Oceans-Spain.2011.6003654
10.1109/JSAC.2016.2545413
10.1007/3-540-45748-8_24
10.1109/LES.2018.2815954
10.1007/978-1-4419-5906-5_752
10.1109/TIFS.2020.2988575
10.1109/ICCV.2017.298
10.1016/j.compenvurbsys.2017.12.005
10.1007/s00779-020-01440-0
10.1109/MIS.2020.2988525
10.1145/2906388.2906418
10.1109/MNET.2019.1800286
10.1109/MC.2016.145
10.3390/s18113726
10.1109/JSEN.2006.886995
10.1109/ACCESS.2019.2918213
10.1109/ICCNC.2019.8685481
10.1109/VTCSpring.2018.8417759
10.1007/978-3-319-16292-8_28
10.1109/CVPR.2018.00907
10.1007/978-3-030-30490-4_56
10.1109/MC.2018.2381112
10.1145/2820975.2820980
10.1109/JPROC.2019.2920341
10.1109/WF-IoT.2015.7389148
10.1109/JSAC.2018.2844939
10.1145/3204949.3204975
10.1145/3212725.3212729
10.1145/3241539.3241563
10.1145/3081333.3081359
10.1145/3038912.3052577
10.1109/ICSENS.2010.5690033
10.1109/JIOT.2020.2967734
10.3115/1658616.1658635
10.1109/ICDCS.2017.226
10.1109/MC.2018.2381129
10.1145/2750858.2804262
10.1145/2809695.2809718
10.1145/3371154
10.1002/ett.3942
10.1109/TNET.2019.2936939
10.1109/TNSE.2021.3053588
10.1145/3154448.3154452
10.1109/MCI.2006.329691
10.1145/2413176.2413189
10.1145/3090076
10.1145/3154815
10.1109/TCSVT.2020.2996231
10.1109/CVPR.2019.00293
10.1109/CySWater.2016.7469060
10.1145/3133956.3133982
10.1109/JIOT.2020.2967772
10.1109/ICDCS.2019.00106
10.1109/CVPR.2015.7298594
10.1109/CVPR.2018.00171
10.1007/978-3-642-38652-7_2
10.1109/ISCIT.2017.8261188
10.1016/j.procs.2015.05.038
10.1007/978-3-319-97909-0_46
10.1023/A:1019956318069
ContentType Journal Article
Copyright Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2021
Copyright_xml – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2021
DBID 97E
RIA
RIE
AAYXX
CITATION
7SP
8FD
L7M
DOI 10.1109/JPROC.2021.3119950
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 Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 1558-2256
EndPage 1837
ExternalDocumentID 10_1109_JPROC_2021_3119950
9596610
Genre orig-research
GrantInformation_xml – fundername: Beijing National Research Center for Information Science and Technology
  grantid: 20031887521
  funderid: 10.13039/501100017582
– fundername: Academy of Finland
  grantid: 319669; 319670; 325570; 326305; 325774; 335934
  funderid: 10.13039/501100002341
– fundername: National Natural Science Foundation of China
  grantid: U1936217; 61971267; 61972223; 61941117; 61861136003
  funderid: 10.13039/501100001809
– fundername: National Key Research and Development Program of China
  grantid: 2020YFA0711400; 2020AAA0106000
  funderid: 10.13039/501100012166
– fundername: Beijing Natural Science Foundation
  grantid: L182038
  funderid: 10.13039/501100004826
GroupedDBID -DZ
-~X
.DC
0R~
123
1OL
29P
3EH
4.4
6IK
85S
97E
9M8
AAJGR
AAWTH
ABAZT
ABFSI
ABJNI
ABQJQ
ABVLG
ACBEA
ACGFS
AENEX
AETEA
AETIX
AFOGA
AGNAY
AGQYO
AGSQL
AHBIQ
AIBXA
ALLEH
ALMA_UNASSIGNED_HOLDINGS
AZLTO
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
DU5
EBS
EJD
FA8
HZ~
H~9
IAAWW
IBMZZ
ICLAB
IDIHD
IFIPE
IFJZH
IPLJI
JAVBF
LAI
M43
MVM
O9-
OCL
RIA
RIE
RIU
RNS
TAE
TN5
TWZ
UDY
UHB
UKR
UQL
VOH
WHG
XJT
XOL
YNT
ZCA
ZXP
ZY4
~02
AAYXX
CITATION
7SP
8FD
L7M
ID FETCH-LOGICAL-c295t-b55d45250ff5111df3f47062473eab05556d45455719ff0bb4c295042fca80cb3
IEDL.DBID RIE
ISICitedReferencesCount 144
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000714203700008&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0018-9219
IngestDate Thu Sep 25 04:52:14 EDT 2025
Sat Nov 29 06:01:44 EST 2025
Tue Nov 18 21:14:42 EST 2025
Wed Aug 27 02:26:04 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 11
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c295t-b55d45250ff5111df3f47062473eab05556d45455719ff0bb4c295042fca80cb3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ORCID 0000-0003-4220-3650
0000-0002-0848-2599
0000-0002-9091-5129
0000-0003-1342-6759
0000-0001-5617-1659
0000-0002-7013-0121
0000-0002-4343-703X
PQID 2592624937
PQPubID 85453
PageCount 60
ParticipantIDs crossref_citationtrail_10_1109_JPROC_2021_3119950
ieee_primary_9596610
crossref_primary_10_1109_JPROC_2021_3119950
proquest_journals_2592624937
PublicationCentury 2000
PublicationDate 2021-11-01
PublicationDateYYYYMMDD 2021-11-01
PublicationDate_xml – month: 11
  year: 2021
  text: 2021-11-01
  day: 01
PublicationDecade 2020
PublicationPlace New York
PublicationPlace_xml – name: New York
PublicationTitle Proceedings of the IEEE
PublicationTitleAbbrev JPROC
PublicationYear 2021
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 ref207
ref57
ref449
ref328
hennessy (ref426) 2011
ref56
ref329
ref59
ref326
ref58
ref448
ref327
ref53
ref445
ref324
ref52
ref446
ref201
ref55
ref322
ref54
ref444
ref323
lin (ref298) 2015
peltonen (ref32) 2020
ref452
ref331
ref453
ref332
ref51
ref450
zhang (ref232) 2017
ref50
ref451
chen (ref31) 2019
ref339
ref219
drolia (ref132) 2017
ref45
kone?n? (ref155) 2016
lin (ref167) 2017
ref458
ref337
ref217
ref47
ref459
ref338
ref42
oskouei (ref316) 2015
ref456
ref215
ref41
ref457
ref336
ref44
ref454
ref333
lopes (ref266) 2017
ref43
ref455
ref334
ref49
ref221
ref463
ref342
ref101
ref222
ref343
ref40
ref461
ref340
ref220
ref462
ref341
ref460
ref35
ref548
ref427
ref34
ref549
ref307
ref428
ref37
ref546
ref304
ref425
ref36
ref547
ref544
ref302
ref545
ref303
ref424
ref33
ref542
ref300
ref543
ref301
mcmahan (ref193) 2017
marjanovi? (ref66) 2017
ref38
ref308
ref429
ref309
zhuo (ref194) 2019
han (ref277) 2015
gebhart (ref508) 2017; 18
ref551
ref430
ref552
ref310
ref431
van leeuwen (ref515) 1976
ref550
you (ref553) 2020; 64
ref24
ref317
ref438
ref23
ref318
ref439
ref26
ref436
ref25
ref437
bagdasaryan (ref204) 2020
ref555
ref20
ref313
ref434
ref314
han (ref293) 2015
ref22
ref432
ref554
ref21
ref312
ref433
hasselt (ref414) 2010; 23
ref28
ref27
ref319
ref29
mo (ref171) 2020
bonawitz (ref208) 2019
ref320
ref441
ref321
ref442
ref440
bo (ref248) 2013
you (ref281) 2019
ref128
ref249
ref129
ref126
ref247
ref368
ref97
ref127
ref369
ref96
ref124
ref245
ref487
ref99
ref125
ref246
ref488
ref367
ref98
tseng (ref100) 2017
das (ref541) 2014
ref254
ref496
ref375
ref93
ref134
ref497
ref376
ref252
ref95
ref495
ref250
ref492
ref371
ref130
ref251
ref372
ref490
ref91
ref90
ref89
ref139
ref258
ref379
ref86
ref138
ref85
ref135
ref256
ref498
ref377
ref88
ref257
ref499
ref378
ref87
sau (ref262) 2016
xu (ref370) 2017
iandola (ref270) 2016
ref144
ref386
ref82
ref145
ref387
li (ref447) 2019
ref81
ref142
ref384
ref84
ref143
ref385
bengio (ref3) 2003; 3
ref83
bhagoji (ref205) 2018
ref140
ref382
ref141
ref383
ref380
ref80
ref381
ref229
ref108
lv (ref8) 2015; 16
ref79
ref78
ref227
ref106
crowley (ref263) 2018
ref348
ref469
ref107
ref349
ref104
ref346
ref467
ref75
ref226
ref105
ref347
ref468
ref74
ref102
ref344
ref77
goodfellow (ref423) 2014
ref224
ref103
ref466
ref76
mccarthy (ref399) 1998
chen (ref210) 2019
talagala (ref374) 2018
(ref465) 2019
ref111
ref353
ref474
ref71
ref233
ref112
ref354
ref475
ref70
ref351
ref73
ref110
ref473
ref72
elsken (ref493) 2018
ref470
ref350
ref119
ref68
hinton (ref259) 2015
ref238
ref117
ref359
ref239
ref118
ref69
ref236
ref115
ref357
ref64
ref237
ref116
ref358
ref63
ref234
ref113
ref355
ref476
ref235
ref114
ref356
ref477
ref65
(ref464) 2019
xu (ref311) 2017
courbariaux (ref296) 2016
chabanne (ref534) 2017; 2017
ref243
ref122
ref364
ref485
ref60
ref244
ref365
ref486
ref241
ref120
ref362
ref483
ref62
smith (ref152) 2017
ref242
ref363
ref484
wang (ref160) 2017
ref360
ref481
ref240
ref361
ref482
(ref12) 2016
ref480
molchanov (ref280) 2016
gilad-bachrach (ref533) 2016
ref168
ref289
kendall (ref5) 2017
nguyen (ref223) 2018; abs 1804 7474
ref290
ref170
lee (ref537) 2017
ref177
ref178
ref175
ref176
romero (ref260) 2014
ref173
ref294
ref174
ref172
kone?ný (ref166) 2017
ref179
kim (ref255) 2015
ref180
ref181
appleyard (ref522) 2016
ref188
kone?ný (ref154) 2015
ref186
ref187
ref184
ref185
ref182
ref183
montgomery (ref407) 2021
ref269
ref148
ref149
ref267
ref146
ref268
ref147
ref389
ref390
cui (ref531) 2017
ref276
ref397
ref398
ref274
ref153
ref395
ref275
ref396
ref151
ref393
ref273
ref394
ref391
ref150
ref392
ref159
ref278
seraj (ref67) 2015
ref279
ref158
russell (ref400) 1995
chen (ref292) 2015
ref287
ref288
mazhar (ref61) 2019
ref285
ref164
ref286
ref283
ref162
ref284
ref163
ref282
ref161
tian (ref366) 2019
qayyum (ref218) 2021
xu (ref46) 2017
kong (ref491) 2017
hard (ref209) 2018
silver (ref417) 2014
schwabacher (ref388) 2007
potes (ref11) 2016
zaremba (ref422) 2014
vanhoucke (ref299) 2011
fang (ref137) 2020
ref1
yang (ref211) 2018
ref191
ref192
ref190
sun (ref2) 2015
ref199
rastegari (ref297) 2016
denil (ref510) 2013
geyer (ref189) 2017
ogden (ref136) 2018
mcmahan (ref39) 2017; 3
yin (ref202) 2018; 80
gong (ref291) 2014
lv (ref131) 2007
blanchard (ref200) 2017
szegedy (ref271) 2017
hadidi (ref373) 2018
malki (ref516) 2003; 3
esser (ref306) 2015
ref7
ref9
ref4
simonyan (ref472) 2014
psaras (ref92) 2018
osband (ref416) 2016
ma (ref345) 2017
morris (ref518) 1971; c 20
steinhardt (ref197) 2017
alhaija (ref6) 2017; 1
sheller (ref213) 2018
hu (ref18) 2015; 11
denton (ref253) 2014
venugopal (ref133) 2018
chen (ref165) 2016
chen (ref30) 2018
chen (ref335) 2018; 2
liu (ref230) 2018
hassibi (ref514) 1993
motamedi (ref325) 2016
gao (ref216) 2019
cheng (ref195) 2019
zhou (ref265) 2018
wistuba (ref494) 2019
trivedi (ref443) 2020
komodakis (ref261) 2017
o’connor (ref121) 2016
liu (ref94) 2020
li (ref264) 2018
chen (ref203) 2017
courbariaux (ref295) 2015
jaggi (ref489) 2014
he (ref479) 2016
rallapalli (ref330) 2016
konda (ref418) 2000
duong (ref225) 2018
buckler (ref123) 2018
mavromoustakis (ref109) 2015; 141
mcmahan (ref156) 2016
roy (ref214) 2019
sprague (ref415) 2003
zhao (ref157) 2018
caldas (ref169) 2018
polese (ref435) 2020
hamerly (ref408) 2004; 16
liu (ref352) 2017
liu (ref520) 2016
shafiee (ref272) 2017
gustafson (ref517) 2017; 4
loukadakis (ref315) 2018
howard (ref231) 2017
ref526
ref405
ref13
ref527
hesamifard (ref535) 2017
ref406
ref524
ref403
ref15
ref525
ref404
ref14
ref401
ref523
ref402
ref521
ref10
ref409
ref17
ref16
ref528
ref19
ref529
ref530
nielsen (ref419) 2015; 25
hyvärinen (ref411) 1999
ref538
ref536
ref412
ref413
biggio (ref196) 2012
ref410
ref532
gal (ref206) 2016
o’shea (ref421) 2015
ref539
(ref48) 2019
lecun (ref513) 1990
ref540
ref420
krizhevsky (ref471) 2012
ref504
ref505
ref502
ref503
fung (ref198) 2018
ref500
ref501
ref509
ref506
ref507
soudry (ref305) 2014
ioffe (ref478) 2015
ref511
ref512
ramaswamy (ref212) 2019
ref519
cortes (ref228) 2017
References_xml – ident: ref512
  doi: 10.1109/MSP.2007.4286571
– ident: ref221
  doi: 10.1109/TVT.2020.2973651
– ident: ref377
  doi: 10.1007/978-3-319-69462-7_27
– year: 2018
  ident: ref205
  article-title: Analyzing federated learning through an adversarial lens
  publication-title: arXiv 1811 12470
– ident: ref84
  doi: 10.1109/ACCESS.2017.2706741
– ident: ref158
  doi: 10.1109/TWC.2020.2971981
– volume: 1
  start-page: 2
  year: 2017
  ident: ref6
  article-title: Augmented reality meets deep learning for car instance segmentation in urban scenes
  publication-title: Proc Brit Mach Vis Conf
– start-page: 119
  year: 2017
  ident: ref200
  article-title: Machine learning with adversaries: Byzantine tolerant gradient descent
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref469
  doi: 10.1109/TCSI.2020.2966243
– ident: ref151
  doi: 10.1145/3194658.3194666
– ident: ref7
  doi: 10.1109/TITS.2014.2311123
– start-page: 21
  year: 2016
  ident: ref520
  article-title: SSD: Single shot multibox detector
  publication-title: Computer Vision-ECCV 2016
  doi: 10.1007/978-3-319-46448-0_2
– ident: ref344
  doi: 10.1007/978-3-030-01249-6_18
– ident: ref40
  doi: 10.1145/3298981
– ident: ref217
  doi: 10.1109/JIOT.2021.3056185
– ident: ref409
  doi: 10.1007/s11222-007-9033-z
– year: 2015
  ident: ref154
  article-title: Federated optimization: Distributed optimization beyond the datacenter
  publication-title: arXiv 1511 03575
– start-page: 1
  year: 2017
  ident: ref311
  article-title: Fast and low-power behavior analysis on vehicles using smartphones
  publication-title: Proc 6th Int Symp Next Gener Electron (ISNE)
– ident: ref438
  doi: 10.1109/MCOM.2014.6736753
– volume: 2
  start-page: 3l
  year: 2018
  ident: ref335
  article-title: Understanding the limitations of existing energy-efficient design approaches for deep neural networks
  publication-title: Energy
– ident: ref473
  doi: 10.1109/WACV.2014.6836013
– year: 2016
  ident: ref325
  article-title: Fast and energy-efficient CNN inference on IoT devices
  publication-title: arXiv 1611 07151
– year: 2019
  ident: ref465
  publication-title: Microsoft HoloLens
– start-page: 387
  year: 2014
  ident: ref417
  article-title: Deterministic policy gradient algorithms
  publication-title: Proc Int Conf Mach Learn
– volume: 16
  start-page: 865
  year: 2015
  ident: ref8
  article-title: Traffic flow prediction with big data: A deep learning approach
  publication-title: IEEE Trans Intell Transp Syst
– ident: ref482
  doi: 10.1609/aimag.v35i4.2513
– year: 2017
  ident: ref352
  article-title: DEEProtect: Enabling inference-based access control on mobile sensing applications
  publication-title: arXiv 1702 06159
– year: 2018
  ident: ref30
  article-title: Federated meta-learning with fast convergence and efficient communication
  publication-title: arXiv 1802 07876
– volume: 80
  start-page: 5650
  year: 2018
  ident: ref202
  article-title: Byzantine-robust distributed learning: Towards optimal statistical rates
  publication-title: Proc 35th Int Conf Mach Learn
– ident: ref28
  doi: 10.1109/LCOMM.2018.2875978
– ident: ref322
  doi: 10.3390/jlpea7020010
– ident: ref455
  doi: 10.1007/s11277-018-5360-y
– ident: ref258
  doi: 10.1145/1150402.1150464
– start-page: 1
  year: 2018
  ident: ref264
  article-title: DeepRebirth: Accelerating deep neural network execution on mobile devices
  publication-title: Proc 32nd AAAI Conf Artif Intell
– ident: ref441
  doi: 10.1109/TSC.2020.3032724
– ident: ref86
  doi: 10.1109/TNSE.2020.3033938
– year: 2019
  ident: ref216
  article-title: HHHFL: Hierarchical heterogeneous horizontal federated learning for electroencephalography
  publication-title: arXiv 1909 05784
– ident: ref312
  doi: 10.1109/TIM.2020.3018831
– ident: ref130
  doi: 10.1109/ICDCS.2017.94
– year: 2016
  ident: ref262
  article-title: Deep model compression: Distilling knowledge from noisy teachers
  publication-title: arXiv 1610 09650
– ident: ref368
  doi: 10.1109/ICDCS.2018.00154
– ident: ref56
  doi: 10.1109/JIOT.2021.3049173
– start-page: 1117
  year: 2015
  ident: ref306
  article-title: Backpropagation for energy-efficient neuromorphic computing
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref9
  doi: 10.1109/PERCOMW.2019.8730706
– ident: ref538
  doi: 10.1109/ICNSC.2019.8743164
– start-page: 950
  year: 2007
  ident: ref131
  article-title: Multi-probe LSH: Efficient indexing for high-dimensional similarity search
  publication-title: Proc Int Conf On Very Large Data Bases
– ident: ref320
  doi: 10.3390/fi9040066
– ident: ref60
  doi: 10.1109/TITS.2018.2877059
– year: 2020
  ident: ref171
  article-title: Energy-efficient federated edge learning with joint communication and computation design
  publication-title: arXiv 2003 00199
– ident: ref380
  doi: 10.1145/3363347.3363366
– ident: ref290
  doi: 10.1109/JSTSP.2020.2977090
– year: 2016
  ident: ref270
  article-title: SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and < 0.5 MB model size
  publication-title: arXiv 1602 07360
– ident: ref122
  doi: 10.1007/s11265-006-4190-4
– ident: ref348
  doi: 10.1109/ICNP.2017.8117585
– ident: ref323
  doi: 10.1145/3089801.3089804
– ident: ref188
  doi: 10.1145/1536414.1536440
– ident: ref503
  doi: 10.1109/TPAMI.2018.2884462
– ident: ref241
  doi: 10.1145/2968219.2971461
– ident: ref185
  doi: 10.1109/TII.2019.2900665
– ident: ref279
  doi: 10.1109/ICPR.2018.8546129
– year: 2018
  ident: ref493
  article-title: Neural architecture search: A survey
  publication-title: arXiv 1808 05377
– ident: ref383
  doi: 10.1109/TII.2018.2842821
– ident: ref126
  doi: 10.1145/2742647.2742663
– start-page: 2938
  year: 2020
  ident: ref204
  article-title: How to backdoor federated learning
  publication-title: Proc Int Conf Artif Intell Statist
– year: 2018
  ident: ref157
  article-title: Federated learning with non-IID data
  publication-title: arXiv 1806 00582
– volume: 64
  start-page: 1
  year: 2020
  ident: ref553
  article-title: Towards 6G wireless communication networks: Vision, enabling technologies, and new paradigm shifts
  publication-title: Sci China Inf Sci
– ident: ref511
  doi: 10.1145/1150402.1150464
– ident: ref483
  doi: 10.1006/ijhc.2001.0499
– ident: ref285
  doi: 10.1109/SiPS.2016.31
– ident: ref387
  doi: 10.1109/ACCESS.2018.2846609
– ident: ref77
  doi: 10.1109/TII.2018.2877217
– ident: ref342
  doi: 10.1145/2938559.2949718
– ident: ref529
  doi: 10.1109/INFCOM.2012.6195845
– ident: ref372
  doi: 10.1109/HPCA.2018.00018
– year: 2018
  ident: ref211
  article-title: Applied federated learning: Improving Google keyboard query suggestions
  publication-title: arXiv 1812 02903
– ident: ref336
  doi: 10.1109/JSSC.2016.2616357
– ident: ref106
  doi: 10.1016/j.adhoc.2018.05.014
– ident: ref105
  doi: 10.1109/VNC.2017.8275621
– ident: ref466
  doi: 10.1109/PerCom45495.2020.9127360
– year: 2017
  ident: ref166
  article-title: Stochastic, distributed and federated optimization for machine learning
  publication-title: arXiv 1707 01155
– year: 2014
  ident: ref541
  article-title: Fingerprinting smart devices through embedded acoustic components
  publication-title: arXiv 1403 3366
– year: 2017
  ident: ref46
  article-title: Accelerating convolutional neural networks for continuous mobile vision via cache reuse
  publication-title: arXiv 1712 01670
– ident: ref62
  doi: 10.1109/ACCESS.2016.2607719
– ident: ref347
  doi: 10.1109/MNET.2018.1700202
– ident: ref177
  doi: 10.1109/VTS-APWCS.2019.8851649
– ident: ref111
  doi: 10.1109/WoWMoM.2019.8792973
– ident: ref412
  doi: 10.1137/0718026
– ident: ref495
  doi: 10.1109/CVPR.2017.195
– ident: ref477
  doi: 10.1145/3123266.3123453
– ident: ref526
  doi: 10.1098/rsta.2019.0155
– ident: ref107
  doi: 10.1109/VNC.2016.7835943
– ident: ref308
  doi: 10.1145/3081333.3081336
– ident: ref507
  doi: 10.1145/2030112.2030164
– ident: ref340
  doi: 10.1145/3081333.3081358
– ident: ref390
  doi: 10.1007/s00521-010-0362-z
– ident: ref52
  doi: 10.1145/2906388.2906395
– ident: ref392
  doi: 10.1186/s40537-014-0008-6
– ident: ref382
  doi: 10.1109/MVT.2018.2883777
– start-page: 1
  year: 2017
  ident: ref66
  article-title: Air and noise pollution monitoring in the city of zagreb by using mobile crowdsensing
  publication-title: Proc 25th Int Conf Softw Telecommun Comput Netw (SoftCOM)
– ident: ref168
  doi: 10.1109/NCA.2017.8171350
– ident: ref304
  doi: 10.1109/EMC2.2018.00012
– ident: ref381
  doi: 10.1002/nem.2146
– ident: ref431
  doi: 10.1109/TPAMI.2013.50
– ident: ref147
  doi: 10.1109/PDP2018.2018.00023
– ident: ref547
  doi: 10.1109/JIOT.2020.3037194
– ident: ref405
  doi: 10.1038/nbt1206-1565
– ident: ref552
  doi: 10.23919/JCC.2020.09.008
– year: 2016
  ident: ref156
  article-title: Communication-efficient learning of deep networks from decentralized data
  publication-title: arXiv 1602 05629
– ident: ref125
  doi: 10.1145/3081333.3081347
– ident: ref484
  doi: 10.1109/MSP.2006.1657817
– ident: ref367
  doi: 10.1016/j.future.2019.01.059
– ident: ref378
  doi: 10.1109/MCC.2018.1081070
– ident: ref138
  doi: 10.1145/2594368.2594388
– ident: ref289
  doi: 10.1109/WACV45572.2020.9093331
– ident: ref103
  doi: 10.1109/MedHocNet.2012.6257116
– ident: ref256
  doi: 10.1145/2964284.2967280
– ident: ref16
  doi: 10.1109/MCOM.2017.1600863
– year: 2018
  ident: ref209
  article-title: Federated learning for mobile keyboard prediction
  publication-title: arXiv 1811 03604
– year: 2015
  ident: ref255
  article-title: Compression of deep convolutional neural networks for fast and low power mobile applications
  publication-title: arXiv 1511 06530
– year: 2014
  ident: ref260
  article-title: FitNets: Hints for thin deep nets
  publication-title: arXiv 1412 6550
– ident: ref302
  doi: 10.1109/CAC.2017.8243585
– ident: ref13
  doi: 10.1109/COMST.2018.2844341
– ident: ref71
  doi: 10.1109/TVT.2013.2291811
– year: 2018
  ident: ref315
  article-title: Accelerating deep neural networks on low power heterogeneous architectures
  publication-title: Proc 11th Int Workshop Programmability Archit Heterogeneous Multicores (MULTIPROG)
– ident: ref434
  doi: 10.1109/BigData.2018.8622396
– year: 2016
  ident: ref12
  publication-title: Cisco Visual Networking Index Global Mobile Data Traffic Forecast Update
– start-page: 25
  year: 2019
  ident: ref61
  article-title: Mobile crowdsensing application of road condition detection
  publication-title: Urdu News Headline Text Classification by Using Different Machine Learning Algorithms
– ident: ref10
  doi: 10.1007/s11042-016-3316-3
– year: 2017
  ident: ref491
  article-title: Science driven innovations powering mobile product: Cloud AI vs. device AI solutions on smart device
  publication-title: arXiv 1711 07580
– year: 2017
  ident: ref167
  article-title: Deep gradient compression: Reducing the communication bandwidth for distributed training
  publication-title: arXiv 1712 01887
– ident: ref519
  doi: 10.1109/ICCV.2019.00140
– ident: ref386
  doi: 10.1109/IC2E48712.2020.00009
– year: 2014
  ident: ref422
  article-title: Recurrent neural network regularization
  publication-title: arXiv 1409 2329
– ident: ref243
  doi: 10.1109/SocialSens.2018.00017
– ident: ref112
  doi: 10.1016/j.agsy.2017.01.023
– ident: ref420
  doi: 10.1016/j.acha.2019.06.004
– ident: ref76
  doi: 10.1109/TITS.2019.2906038
– ident: ref354
  doi: 10.1109/MC.2017.3641638
– ident: ref78
  doi: 10.1109/INFOCOM.2015.7218644
– ident: ref161
  doi: 10.1109/TWC.2019.2961673
– year: 2017
  ident: ref231
  article-title: MobileNets: Efficient convolutional neural networks for mobile vision applications
  publication-title: arXiv 1704 04861
– year: 2017
  ident: ref535
  article-title: CryptoDL: Deep neural networks over encrypted data
  publication-title: arXiv 1711 05189
– ident: ref449
  doi: 10.1145/3243734.3243781
– ident: ref119
  doi: 10.1109/TCSVT.2019.2903421
– start-page: 1467
  year: 2012
  ident: ref196
  article-title: Poisoning attacks against support vector machines
  publication-title: Proc 29th Int Conf Mach Learn (ICML)
– ident: ref179
  doi: 10.1109/ICC40277.2020.9148862
– start-page: 3517
  year: 2017
  ident: ref197
  article-title: Certified defenses for data poisoning attacks
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref220
  doi: 10.1109/TCOMM.2019.2956472
– ident: ref548
  doi: 10.1145/3338501.3357371
– ident: ref275
  doi: 10.1109/ACCESS.2018.2810264
– ident: ref363
  doi: 10.1145/3132211.3134457
– ident: ref318
  doi: 10.1109/ICCE.2016.7430525
– ident: ref490
  doi: 10.1109/CVPR.2018.00745
– ident: ref546
  doi: 10.1145/3378679.3394533
– ident: ref73
  doi: 10.1016/j.adhoc.2016.02.002
– ident: ref57
  doi: 10.1109/TITS.2020.3040557
– ident: ref450
  doi: 10.1109/TII.2019.2952645
– year: 1998
  ident: ref399
  publication-title: What Is Artificial Intelligence
– ident: ref365
  doi: 10.1109/AVSS.2018.8639121
– ident: ref448
  doi: 10.1145/3196494.3196532
– ident: ref251
  doi: 10.1145/3131895
– ident: ref127
  doi: 10.1145/3173162.3173185
– ident: ref294
  doi: 10.1109/CVPR.2016.521
– ident: ref440
  doi: 10.1109/INFOCOM.2016.7524381
– start-page: 1
  year: 2018
  ident: ref92
  article-title: Mobile data repositories at the edge
  publication-title: Proc USENIX Workshop Hot Topics Edge Comput (HotEdge)
– start-page: 2285
  year: 2015
  ident: ref292
  article-title: Compressing neural networks with the hashing trick
  publication-title: Proc Int Conf Mach Learn
– ident: ref142
  doi: 10.1109/SEC.2016.38
– ident: ref93
  doi: 10.1145/3359993.3366644
– volume: 23
  start-page: 2613
  year: 2010
  ident: ref414
  article-title: Double Q-learning
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref371
  doi: 10.1145/3213344.3213345
– ident: ref339
  doi: 10.1145/2906388.2906396
– ident: ref309
  doi: 10.1145/3152127
– year: 2015
  ident: ref293
  article-title: Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding
  publication-title: arXiv 1510 00149 [cs]
– ident: ref337
  doi: 10.1109/JETCAS.2019.2910232
– ident: ref317
  doi: 10.1145/2964284.2973801
– ident: ref496
  doi: 10.1109/CVPR.2017.634
– ident: ref49
  doi: 10.1109/ICIP.2012.6467224
– ident: ref81
  doi: 10.1109/ICCSE.2016.7581655
– year: 2016
  ident: ref416
  article-title: Deep exploration via bootstrapped DQN
  publication-title: arXiv 1602 04621
– ident: ref384
  doi: 10.1109/TII.2017.2679740
– ident: ref542
  doi: 10.1145/3267305.3267505
– volume: 3
  start-page: 1137
  year: 2003
  ident: ref3
  article-title: A neural probabilistic language model
  publication-title: J Mach Learn Res
– year: 2019
  ident: ref366
  article-title: LEP-CNN: A lightweight edge device assisted privacy-preserving CNN inference solution for IoT
  publication-title: arXiv 1901 04100
– ident: ref539
  doi: 10.1109/MCC.2016.124
– ident: ref162
  doi: 10.1109/TCOMM.2020.2979149
– year: 2015
  ident: ref298
  article-title: Neural networks with few multiplications
  publication-title: arXiv 1510 03009 [cs]
– ident: ref90
  doi: 10.1016/j.procs.2015.05.043
– ident: ref79
  doi: 10.1007/s00779-016-0932-x
– ident: ref15
  doi: 10.1109/JIOT.2016.2579198
– ident: ref506
  doi: 10.1109/ICASSP.2013.6638947
– start-page: 128
  year: 2015
  ident: ref67
  article-title: Roads: A road pavement monitoring system for anomaly detection using smart phones
  publication-title: Big Data Analytics in the Social and Ubiquitous Context
– ident: ref192
  doi: 10.1145/2976749.2978318
– ident: ref333
  doi: 10.1145/3020078.3021736
– ident: ref329
  doi: 10.1145/3140582.3081040
– ident: ref44
  doi: 10.1109/JIOT.2020.2984887
– ident: ref385
  doi: 10.1109/TSC.2017.2662008
– ident: ref331
  doi: 10.1109/CVPR.2016.91
– ident: ref180
  doi: 10.1109/ICASSP40776.2020.9054634
– ident: ref360
  doi: 10.1109/INFOCOM.2018.8485905
– ident: ref467
  doi: 10.1145/3397461
– year: 2017
  ident: ref272
  article-title: SquishedNets: Squishing SqueezeNet further for edge device scenarios via deep evolutionary synthesis
  publication-title: arXiv 1711 07459
– ident: ref113
  doi: 10.1109/LCN.2016.039
– ident: ref41
  doi: 10.1109/COMST.2020.3007787
– start-page: 1
  year: 2018
  ident: ref374
  article-title: ECO: Harmonizing edge and cloud with ML/DL orchestration
  publication-title: Proc USENIX Workshop Hot Topics Edge Comput (HotEdge)
– ident: ref554
  doi: 10.1109/MNET.001.1900287
– ident: ref68
  doi: 10.1016/j.aci.2018.05.002
– ident: ref528
  doi: 10.1145/3351260
– ident: ref176
  doi: 10.1109/ACCESS.2020.3038287
– ident: ref173
  doi: 10.1016/j.phycom.2021.101381
– ident: ref36
  doi: 10.1109/MPRV.2017.2940968
– ident: ref475
  doi: 10.1145/2619239.2626296
– ident: ref286
  doi: 10.1016/j.patcog.2020.107461
– ident: ref314
  doi: 10.1145/3089801.3089805
– ident: ref536
  doi: 10.1002/nav.3800010110
– year: 2017
  ident: ref100
  article-title: When cars meet distributed computing: Data storage as an example
  publication-title: arXiv 1711 02014
– ident: ref452
  doi: 10.1109/IUCC/DSCI/SmartCNS.2019.00126
– ident: ref343
  doi: 10.4108/eai.30-11-2016.2267463
– ident: ref550
  doi: 10.1109/TSG.2016.2553647
– ident: ref395
  doi: 10.1109/ICECC.2011.6066743
– ident: ref498
  doi: 10.1109/TIFS.2015.2400395
– ident: ref235
  doi: 10.1109/ACCESS.2018.2843341
– ident: ref222
  doi: 10.1109/GLOBECOM38437.2019.9013587
– ident: ref129
  doi: 10.1145/3241539.3241565
– ident: ref468
  doi: 10.1109/TSE.2017.2685387
– ident: ref463
  doi: 10.1109/ICSCC.2019.8843640
– ident: ref327
  doi: 10.1145/3126555
– ident: ref394
  doi: 10.1016/j.future.2017.03.034
– year: 2016
  ident: ref121
  article-title: Sigma delta quantized networks
  publication-title: arXiv 1611 02024
– start-page: 533
  year: 2018
  ident: ref123
  article-title: EVA2: Exploiting temporal redundancy in live computer vision
  publication-title: Proc ACM/IEEE 45th Annu Int Symp Comput Archit (ISCA)
– ident: ref170
  doi: 10.1109/TWC.2020.3037554
– ident: ref249
  doi: 10.1145/2030613.2030625
– ident: ref428
  doi: 10.1109/COMST.2018.2820021
– start-page: 621
  year: 2016
  ident: ref11
  article-title: Ensemble of feature-based and deep learning-based classifiers for detection of abnormal heart sounds
  publication-title: Proc Comput Cardiol Conf (CinC)
– start-page: 630
  year: 2016
  ident: ref479
  article-title: Identity mappings in deep residual networks
  publication-title: Computer Vision-ECCV 2016
  doi: 10.1007/978-3-319-46493-0_38
– ident: ref487
  doi: 10.1109/ICDCS.2018.00041
– ident: ref353
  doi: 10.1109/JIOT.2019.2897005
– ident: ref364
  doi: 10.1145/3229556.3229562
– ident: ref551
  doi: 10.1109/MVT.2019.2921208
– ident: ref172
  doi: 10.1109/IPDPS47924.2020.00033
– ident: ref134
  doi: 10.1145/3299710.3211336
– start-page: 1
  year: 2018
  ident: ref133
  article-title: Shadow puppets: Cloud-level accurate AI inference at the speed and economy of edge
  publication-title: Proc USENIX Workshop Hot Topics Edge Comput (HotEdge)
– ident: ref397
  doi: 10.1145/3126686.3126739
– ident: ref432
  doi: 10.1007/s11633-017-1054-2
– ident: ref89
  doi: 10.1109/JIOT.2020.2988697
– ident: ref481
  doi: 10.1109/CVPR.2016.308
– year: 2019
  ident: ref194
  article-title: Federated deep reinforcement learning
  publication-title: arXiv 1901 08277
– ident: ref543
  doi: 10.2200/S00737ED1V01Y201610AIM033
– ident: ref485
  doi: 10.3390/s18051532
– ident: ref146
  doi: 10.1109/ICPADS.2017.00069
– ident: ref391
  doi: 10.1049/cje.2016.11.016
– ident: ref459
  doi: 10.1016/j.compag.2020.105233
– ident: ref509
  doi: 10.1038/s41928-018-0059-3
– ident: ref242
  doi: 10.3390/s18072203
– year: 1995
  ident: ref400
  publication-title: Artificial Intelligence A Modern Approach
– year: 2017
  ident: ref193
  article-title: Learning differentially private recurrent language models
  publication-title: arXiv 1710 06963
– year: 2016
  ident: ref165
  article-title: Revisiting distributed synchronous SGD
  publication-title: arXiv 1604 00981
– ident: ref356
  doi: 10.1145/3197231.3197256
– ident: ref451
  doi: 10.1109/TMC.2020.3019988
– ident: ref87
  doi: 10.1109/JIOT.2020.2979691
– year: 2016
  ident: ref330
  article-title: Are very deep neural networks feasible on mobile devices
  publication-title: IEEE Trans Circuits Syst Video Technol
– ident: ref461
  doi: 10.1007/978-981-10-8339-6_6
– ident: ref527
  doi: 10.1109/JIOT.2017.2750180
– ident: ref402
  doi: 10.1126/science.aaa8415
– start-page: 3068
  year: 2014
  ident: ref489
  article-title: Communication-efficient distributed dual coordinate ascent
  publication-title: Proc Adv Neural Inf Process Syst
– year: 2014
  ident: ref291
  article-title: Compressing deep convolutional networks using vector quantization
  publication-title: arXiv 1412 6115
– ident: ref104
  doi: 10.1109/VNC51378.2020.9318400
– ident: ref313
  doi: 10.1109/ICCCBDA51879.2021.9442600
– ident: ref369
  doi: 10.1145/3132211.3134459
– year: 2017
  ident: ref232
  article-title: Hello edge: Keyword spotting on microcontrollers
  publication-title: arXiv 1711 07128
– ident: ref283
  doi: 10.1145/3131672.3131675
– ident: ref254
  doi: 10.1145/3109761.3109804
– volume: 4
  start-page: 71
  year: 2017
  ident: ref517
  article-title: Beating floating point at its own game: Posit arithmetic
  publication-title: Supercomputing Frontiers and Innovations
– ident: ref444
  doi: 10.3390/s20237000
– ident: ref521
  doi: 10.1109/JIOT.2019.2902141
– ident: ref361
  doi: 10.1145/2973750.2973777
– ident: ref476
  doi: 10.1145/3083187.3084016
– year: 2021
  ident: ref218
  article-title: Collaborative federated learning for healthcare: Multi-modal COVID-19 diagnosis at the edge
  publication-title: arXiv 2101 07511
– ident: ref523
  doi: 10.1145/3210240.3210337
– start-page: 382
  year: 1976
  ident: ref515
  article-title: On the construction of Huffman trees
  publication-title: Proc ICALP
– ident: ref33
  doi: 10.1109/MCOM.001.1900103
– start-page: 164
  year: 1993
  ident: ref514
  article-title: Second order derivatives for network pruning: Optimal brain surgeon
  publication-title: Proc Adv Neural Inf Process Syst
– start-page: 2888
  year: 2018
  ident: ref263
  article-title: Moonshine: Distilling with cheap convolutions
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref437
  doi: 10.1109/AINA.2015.254
– volume: 25
  year: 2015
  ident: ref419
  publication-title: Neural Networks and Deep Learning
– ident: ref413
  doi: 10.1007/BF00992698
– ident: ref288
  doi: 10.1109/WACV45572.2020.9093377
– ident: ref350
  doi: 10.1145/3093337.3037698
– ident: ref252
  doi: 10.5244/C.28.88
– start-page: 199
  year: 2013
  ident: ref248
  article-title: You're driving and texting: Detecting drivers using personal smart phones by leveraging inertial sensors
  publication-title: Proc 19th Annu Int Conf Mobile Comput Netw
– year: 2016
  ident: ref296
  article-title: Binarized neural networks: Training deep neural networks with weights and activations constrained to +1 or ?1
  publication-title: arXiv 1602 02830 [cs]
– ident: ref115
  doi: 10.1109/MNET.011.2000180
– ident: ref43
  doi: 10.1109/COMST.2020.2970550
– ident: ref99
  doi: 10.1109/ICWS49710.2020.00073
– ident: ref398
  doi: 10.1109/TMC.2016.2567378
– ident: ref118
  doi: 10.1145/3081333.3089331
– ident: ref14
  doi: 10.1007/978-3-319-57959-7
– ident: ref47
  doi: 10.1145/3241539.3241557
– start-page: 4424
  year: 2017
  ident: ref152
  article-title: Federated multi-task learning
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref215
  doi: 10.1109/MIS.2020.2988604
– ident: ref164
  doi: 10.1109/JIOT.2020.2964162
– year: 2016
  ident: ref522
  article-title: Optimizing performance of recurrent neural networks on GPUs
  publication-title: arXiv 1604 01946
– ident: ref139
  doi: 10.1109/SEC50012.2020.00041
– year: 2018
  ident: ref373
  article-title: Musical chair: Efficient real-time recognition using collaborative IoT devices
  publication-title: arXiv 1802 02138
– ident: ref85
  doi: 10.1016/j.comcom.2019.12.054
– ident: ref233
  doi: 10.1109/ICRA.2019.8794182
– volume: 16
  start-page: 281
  year: 2004
  ident: ref408
  article-title: Learning the k in k-means
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref470
  doi: 10.1109/CVPR.2005.177
– ident: ref174
  doi: 10.1109/ICC.2019.8761315
– ident: ref505
  doi: 10.1056/NEJMsa1204142
– ident: ref310
  doi: 10.1145/3089801.3089803
– ident: ref64
  doi: 10.1145/2639189.2670273
– ident: ref540
  doi: 10.1109/IPSN.2018.00048
– year: 2019
  ident: ref31
  article-title: FedHealth: A federated transfer learning framework for wearable healthcare
  publication-title: arXiv 1907 09173
– year: 2011
  ident: ref426
  publication-title: Computer Architecture A Quantitative Approach
– ident: ref149
  doi: 10.1109/CSTIC.2019.8755642
– year: 2020
  ident: ref435
  article-title: Machine learning at the edge: A data-driven architecture with applications to 5G cellular networks
  publication-title: IEEE Trans Mobile Comput
– ident: ref532
  doi: 10.1109/COMST.2018.2808242
– start-page: 963
  year: 2014
  ident: ref305
  article-title: Expectation backpropagation: Parameter-free training of multilayer neural networks with continuous or discrete weights
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref549
  doi: 10.1109/JIOT.2020.2978286
– ident: ref145
  doi: 10.1145/3213344.3213351
– ident: ref74
  doi: 10.1109/ACCESS.2019.2951587
– ident: ref135
  doi: 10.1145/3278721.3278778
– ident: ref359
  doi: 10.1145/3097895.3097903
– start-page: 2148
  year: 2013
  ident: ref510
  article-title: Predicting parameters in deep learning
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref430
  doi: 10.1016/j.compeleceng.2017.12.009
– volume: 18
  start-page: 7
  year: 2017
  ident: ref508
  article-title: Google home to the Amazon echo: 'Anything you can do...
  publication-title: CNET
– ident: ref72
  doi: 10.1016/j.ins.2019.01.014
– ident: ref23
  doi: 10.1109/SmartCloud.2016.18
– year: 2015
  ident: ref421
  article-title: An introduction to convolutional neural networks
  publication-title: arXiv 1511 08458
– start-page: 1050
  year: 2016
  ident: ref206
  article-title: Dropout as a Bayesian approximation: Representing model uncertainty in deep learning
  publication-title: Proc Int Conf Mach Learn
– ident: ref486
  doi: 10.1109/MNET.001.1900506
– year: 2019
  ident: ref48
  publication-title: Google Street View Image API
– ident: ref276
  doi: 10.1145/3089801.3089806
– ident: ref17
  doi: 10.1145/2831347.2831354
– year: 2020
  ident: ref32
  article-title: 6G white paper on edge intelligence
  publication-title: arXiv 2004 14850
– ident: ref141
  doi: 10.1145/3220192.3220460
– ident: ref324
  doi: 10.1007/978-1-4302-4000-6_9
– ident: ref148
  doi: 10.1109/CIT/IUCC/DASC/PICOM.2015.170
– start-page: 1
  year: 2017
  ident: ref271
  article-title: Inception-v4, inception-ResNet and the impact of residual connections on learning
  publication-title: Proc 31st AAAI Conf Artif Intell
– ident: ref319
  doi: 10.3390/electronics5040088
– year: 2017
  ident: ref160
  publication-title: Co-op Cooperative machine learning from mobile devices
– year: 2019
  ident: ref212
  article-title: Federated learning for emoji prediction in a mobile keyboard
  publication-title: arXiv 1906 04329
– year: 2017
  ident: ref370
  article-title: DeepWear: Adaptive local offloading for on-wearable deep learning
  publication-title: arXiv 1712 03073
– ident: ref163
  doi: 10.1109/TWC.2020.2974748
– volume: 11
  start-page: 1
  year: 2015
  ident: ref18
  publication-title: Mobile edge computing-A key technology towards 5G
– ident: ref362
  doi: 10.1145/1999995.2000000
– ident: ref65
  doi: 10.1145/3191752
– ident: ref274
  doi: 10.24963/ijcai.2018/429
– ident: ref375
  doi: 10.1016/j.jss.2018.03.032
– start-page: 525
  year: 2016
  ident: ref297
  article-title: XNOR-Net: ImageNet classification using binary convolutional neural networks
  publication-title: Computer Vision-ECCV 2016
  doi: 10.1007/978-3-319-46493-0_32
– year: 2017
  ident: ref345
  article-title: DeepRT: Deep learning for peptide retention time prediction in proteomics
  publication-title: arXiv 1705 05368
– ident: ref42
  doi: 10.1109/ICDCS.2018.00139
– ident: ref34
  doi: 10.1109/HotWeb.2015.22
– ident: ref488
  doi: 10.1109/TITS.2021.3081560
– ident: ref175
  doi: 10.1109/JIOT.2020.3036157
– ident: ref247
  doi: 10.1145/2753509.2753518
– ident: ref349
  doi: 10.1145/3194554.3194565
– ident: ref300
  doi: 10.21437/Interspeech.2016-128
– ident: ref54
  doi: 10.1109/ITAIC49862.2020.9338964
– start-page: 1
  year: 2018
  ident: ref265
  article-title: Rocket launching: A universal and efficient framework for training well-performing light net
  publication-title: Proc 32nd AAAI Conf Artif Intell
– ident: ref376
  doi: 10.1109/SMARTCOMP.2018.00087
– ident: ref238
  doi: 10.3233/JIFS-169699
– year: 2021
  ident: ref407
  publication-title: Introduction to Linear Regression Analysis
– ident: ref51
  doi: 10.1109/GLOBECOM38437.2019.9013821
– ident: ref332
  doi: 10.1109/NGCAS.2017.16
– ident: ref181
  doi: 10.1109/TWC.2020.3003744
– ident: ref35
  doi: 10.1145/2594368.2594383
– ident: ref183
  doi: 10.1109/IJCNN48605.2020.9207469
– ident: ref159
  doi: 10.1109/JSAC.2019.2904348
– ident: ref410
  doi: 10.1016/0169-7439(87)80084-9
– ident: ref433
  doi: 10.1109/ACCESS.2014.2325029
– ident: ref19
  doi: 10.1145/2342509.2342513
– ident: ref321
  doi: 10.3390/app7080826
– year: 2019
  ident: ref195
  article-title: SecureBoost: A lossless federated learning framework
  publication-title: arXiv 1901 08755
– volume: 3
  year: 2017
  ident: ref39
  article-title: Federated learning: Collaborative machine learning without centralized training data
  publication-title: Google Research Blog
– volume: 3
  start-page: 261
  year: 2003
  ident: ref516
  article-title: CNN image processing on a Xilinx Virtex-II 6000
  publication-title: Proc ECCTD
– ident: ref269
  doi: 10.1109/EMC2-NIPS53020.2019.00013
– ident: ref460
  doi: 10.1016/j.ecolind.2020.107124
– start-page: 3123
  year: 2015
  ident: ref295
  article-title: BinaryConnect: Training deep neural networks with binary weights during propagations
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref524
  doi: 10.1016/S0043-1648(00)00427-0
– start-page: 448
  year: 2015
  ident: ref478
  article-title: Batch normalization: Accelerating deep network training by reducing internal covariate shift
  publication-title: Proc Int Conf Mach Learn
– ident: ref429
  doi: 10.1109/TMC.2020.2984364
– ident: ref454
  doi: 10.1109/TMC.2018.2815694
– ident: ref25
  doi: 10.1109/JPROC.2019.2922285
– volume: abs 1804 7474
  year: 2018
  ident: ref223
  article-title: DïoT: A self-learning system for detecting compromised IoT devices
  publication-title: CoRR
– ident: ref201
  doi: 10.1145/3154503
– start-page: 5574
  year: 2017
  ident: ref5
  article-title: What uncertainties do we need in Bayesian deep learning for computer vision?
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref184
  doi: 10.1109/TITS.2020.3002712
– ident: ref1
  doi: 10.5244/C.29.41
– year: 1999
  ident: ref411
  publication-title: Survey on Independent Component Analysis
– year: 2018
  ident: ref198
  article-title: Mitigating sybils in federated learning poisoning
  publication-title: arXiv 1808 04866
– ident: ref501
  doi: 10.1109/PCCC.2017.8280500
– ident: ref338
  doi: 10.1109/LCA.2020.2979965
– ident: ref236
  doi: 10.1109/CVPR.2018.00474
– ident: ref530
  doi: 10.1145/2906388.2906405
– year: 2016
  ident: ref155
  article-title: Federated learning: Strategies for improving communication efficiency
  publication-title: arXiv 1610 05492
– ident: ref237
  doi: 10.1109/PERCOMW.2016.7457169
– year: 2016
  ident: ref280
  article-title: Pruning convolutional neural networks for resource efficient inference
  publication-title: arXiv 1611 06440
– ident: ref70
  doi: 10.1109/TMC.2016.2607716
– ident: ref458
  doi: 10.1145/3152130.3152135
– ident: ref98
  doi: 10.1016/j.jpdc.2018.11.006
– ident: ref4
  doi: 10.1145/1390156.1390177
– ident: ref462
  doi: 10.1007/s11042-020-09740-6
– ident: ref328
  doi: 10.1145/2935643.2935650
– ident: ref234
  doi: 10.1109/CVPR.2018.00716
– ident: ref63
  doi: 10.1109/COMSNETS.2018.8328270
– ident: ref124
  doi: 10.1145/2809695.2809711
– ident: ref140
  doi: 10.1145/3240508.3240697
– ident: ref153
  doi: 10.1145/3373087.3375312
– volume: c 20
  start-page: 1578
  year: 1971
  ident: ref518
  article-title: tapered floating point: a new floating-point representation
  publication-title: IEEE Transactions on Computers
  doi: 10.1109/T-C.1971.223174
– ident: ref442
  doi: 10.1109/TPDS.2020.3016344
– ident: ref69
  doi: 10.3390/ijgi8090412
– year: 2018
  ident: ref169
  article-title: Expanding the reach of federated learning by reducing client resource requirements
  publication-title: arXiv 1812 07210
– ident: ref425
  doi: 10.3115/v1/D14-1179
– ident: ref355
  doi: 10.1109/CFEC.2018.8358733
– ident: ref95
  doi: 10.23919/INM.2017.7987297
– ident: ref80
  doi: 10.1109/TITS.2020.3023446
– ident: ref303
  doi: 10.1109/ICASSP.2015.7178146
– start-page: 1
  year: 2020
  ident: ref443
  article-title: Sharing and caring of data at the edge
  publication-title: Proc 3rd USENIX Workshop Hot Topics Edge Comput (HotEdge)
– ident: ref50
  doi: 10.1109/TCOMM.2020.3019527
– ident: ref114
  doi: 10.3390/electronics8080896
– year: 2017
  ident: ref189
  article-title: Differentially private federated learning: A client level perspective
  publication-title: arXiv 1712 07557
– ident: ref240
  doi: 10.1145/3212725.3212728
– start-page: 598
  year: 1990
  ident: ref513
  article-title: Optimal brain damage
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref403
  doi: 10.1002/cem.873
– ident: ref45
  doi: 10.1109/JPROC.2019.2918951
– start-page: 1269
  year: 2014
  ident: ref253
  article-title: Exploiting linear structure within convolutional networks for efficient evaluation
  publication-title: Proc Adv Neural Inf Process Syst
– start-page: 1
  year: 2017
  ident: ref531
  article-title: Cross-platform machine learning characterization for task allocation in IoT ecosystems
  publication-title: Proc IEEE 7th Annu Comput Commun Workshop Conf (CCWC)
– ident: ref75
  doi: 10.1002/wcm.2591
– ident: ref37
  doi: 10.1145/3161174
– ident: ref396
  doi: 10.1145/3240508.3240561
– ident: ref144
  doi: 10.1016/j.pmcj.2017.07.014
– start-page: 1
  year: 2017
  ident: ref261
  article-title: Paying more attention to attention: Improving the performance of convolutional neural networks via attention transfer
  publication-title: Proc ICLR
– ident: ref96
  doi: 10.1109/INFOCOM42981.2021.9488804
– ident: ref282
  doi: 10.1109/CVPR.2017.643
– ident: ref424
  doi: 10.1162/neco.1997.9.8.1735
– year: 2017
  ident: ref203
  article-title: Targeted backdoor attacks on deep learning systems using data poisoning
  publication-title: arXiv 1712 05526
– ident: ref219
  doi: 10.2196/24207
– ident: ref83
  doi: 10.1109/TNSE.2020.3014455
– ident: ref341
  doi: 10.1109/IPSN.2016.7460664
– ident: ref346
  doi: 10.1109/TVLSI.2018.2825145
– ident: ref227
  doi: 10.1609/aaai.v33i01.33014780
– ident: ref389
  doi: 10.1109/TVT.2010.2043968
– year: 2020
  ident: ref137
  article-title: CacheNet: A model caching framework for deep learning inference on the edge
  publication-title: arXiv 2007 01793
– start-page: 2133
  year: 2019
  ident: ref281
  article-title: Gate decorator: Global filter pruning method for accelerating deep convolutional neural networks
  publication-title: Advances in neural information processing systems
– ident: ref393
  doi: 10.1145/2757384.2757397
– start-page: 252
  year: 2017
  ident: ref537
  article-title: Deep neural networks for wild fire detection with unmanned aerial vehicle
  publication-title: Proc IEEE Int Conf Consum Electron (ICCE)
– ident: ref20
  doi: 10.1109/COMST.2019.2904897
– ident: ref268
  doi: 10.1109/CVPR.2016.90
– ident: ref301
  doi: 10.1109/ICACSIS.2017.8355051
– ident: ref436
  doi: 10.1109/PERCOMW.2015.7134002
– start-page: 1008
  year: 2000
  ident: ref418
  article-title: Actor-critic algorithms
  publication-title: Proc Adv Neural Inf Process Syst
– start-page: 874
  year: 2017
  ident: ref228
  article-title: AdaNet: Adaptive structural learning of artificial neural networks
  publication-title: Proc 34th Int Conf Mach Learn
– ident: ref143
  doi: 10.1109/ICCCN.2018.8487352
– ident: ref497
  doi: 10.1109/PerCom45495.2020.9127387
– ident: ref27
  doi: 10.1109/MNET.2019.1700470
– ident: ref492
  doi: 10.1109/IC2E.2018.00042
– ident: ref150
  doi: 10.1109/CHASE.2017.115
– ident: ref334
  doi: 10.1145/3193025.3193041
– ident: ref207
  doi: 10.1145/3161181
– ident: ref480
  doi: 10.1109/5.726791
– ident: ref53
  doi: 10.1109/ICITBS.2019.00104
– ident: ref101
  doi: 10.1145/2633661.2633671
– ident: ref116
  doi: 10.1109/GCWkshps45667.2019.9024587
– ident: ref257
  doi: 10.1145/2994551.2994564
– ident: ref404
  doi: 10.1007/978-3-642-31537-4_13
– year: 2019
  ident: ref208
  article-title: Towards federated learning at scale: System design
  publication-title: arXiv 1902 01046
– start-page: 1
  year: 2020
  ident: ref94
  article-title: Scale-out edge storage systems with embedded storage nodes to get better availability and cost-efficiency at the same time
  publication-title: Proc 3rd USENIX Workshop Hot Topics Edge Comput (HotEdge)
– ident: ref555
  doi: 10.1109/MVT.2018.2848498
– ident: ref120
  doi: 10.1145/3131885.3131906
– ident: ref102
  doi: 10.1109/MCOM.2014.6736756
– ident: ref401
  doi: 10.1080/03071847.2019.1693801
– ident: ref456
  doi: 10.1109/Oceans-Spain.2011.6003654
– ident: ref439
  doi: 10.1109/JSAC.2016.2545413
– ident: ref199
  doi: 10.1007/3-540-45748-8_24
– ident: ref326
  doi: 10.1109/LES.2018.2815954
– ident: ref191
  doi: 10.1007/978-1-4419-5906-5_752
– ident: ref190
  doi: 10.1109/TIFS.2020.2988575
– ident: ref284
  doi: 10.1109/ICCV.2017.298
– ident: ref58
  doi: 10.1016/j.compenvurbsys.2017.12.005
– ident: ref82
  doi: 10.1007/s00779-020-01440-0
– ident: ref187
  doi: 10.1109/MIS.2020.2988525
– ident: ref128
  doi: 10.1145/2906388.2906418
– start-page: 201
  year: 2016
  ident: ref533
  article-title: CryptoNets: Applying neural networks to encrypted data with high throughput and accuracy
  publication-title: Proc Int Conf Mach Learn
– year: 2003
  ident: ref415
  publication-title: Multiple-Goal Reinforcement Learning with Modular Sarsa(0)
– ident: ref26
  doi: 10.1109/MNET.2019.1800286
– year: 2011
  ident: ref299
  article-title: Improving the speed of neural networks on CPUs
  publication-title: Proc Deep Learn Unsupervised Feature Learn Workshop (NIPS)
– year: 2014
  ident: ref472
  article-title: Very deep convolutional networks for large-scale image recognition
  publication-title: arXiv 1409 1556
– volume: 2017
  start-page: 35
  year: 2017
  ident: ref534
  article-title: Privacy-preserving classification on deep neural network
  publication-title: IACR Cryptol ePrint Arch
– ident: ref22
  doi: 10.1109/MC.2016.145
– ident: ref457
  doi: 10.1109/JIOT.2016.2579198
– ident: ref239
  doi: 10.3390/s18113726
– ident: ref504
  doi: 10.1109/JSEN.2006.886995
– ident: ref91
  doi: 10.1109/ACCESS.2019.2918213
– ident: ref108
  doi: 10.1109/ICCNC.2019.8685481
– ident: ref110
  doi: 10.1109/VTCSpring.2018.8417759
– year: 2019
  ident: ref210
  article-title: Federated learning of out-of-vocabulary words
  publication-title: arXiv 1903 10635
– volume: 141
  start-page: 387
  year: 2015
  ident: ref109
  article-title: Analysis of vehicular storage and dissemination services based on floating content
  publication-title: Mobile Networks and Management
  doi: 10.1007/978-3-319-16292-8_28
– ident: ref226
  doi: 10.1109/CVPR.2018.00907
– year: 2019
  ident: ref494
  article-title: A survey on neural architecture search
  publication-title: arXiv 1905 01392
– start-page: 703
  year: 2019
  ident: ref447
  article-title: MAD-GAN: Multivariate anomaly detection for time series data with generative adversarial networks
  publication-title: Artificial Neural Networks and Machine Learning-ICANN 2019 Text and Time Series
  doi: 10.1007/978-3-030-30490-4_56
– ident: ref499
  doi: 10.1109/MC.2018.2381112
– ident: ref38
  doi: 10.1145/2820975.2820980
– ident: ref24
  doi: 10.1109/JPROC.2019.2920341
– ident: ref59
  doi: 10.1109/WF-IoT.2015.7389148
– ident: ref427
  doi: 10.1109/JSAC.2018.2844939
– year: 2015
  ident: ref259
  article-title: Distilling the knowledge in a neural network
  publication-title: ArXiv 1503 02531
– ident: ref357
  doi: 10.1145/3204949.3204975
– ident: ref245
  doi: 10.1145/3212725.3212729
– ident: ref117
  doi: 10.1145/3241539.3241563
– ident: ref307
  doi: 10.1145/3081333.3081359
– ident: ref244
  doi: 10.1145/3038912.3052577
– ident: ref445
  doi: 10.1109/ICSENS.2010.5690033
– year: 2018
  ident: ref230
  article-title: DARTS: Differentiable architecture search
  publication-title: Arxiv 1806 09055
– ident: ref351
  doi: 10.1109/JIOT.2020.2967734
– ident: ref545
  doi: 10.3115/1658616.1658635
– ident: ref379
  doi: 10.1109/ICDCS.2017.226
– ident: ref29
  doi: 10.1109/MC.2018.2381129
– ident: ref250
  doi: 10.1145/2750858.2804262
– start-page: 17
  year: 2017
  ident: ref132
  article-title: Precog: Prefetching for image recognition applications at the edge
  publication-title: Proc 2nd ACM/IEEE Symp Edge Comput
– ident: ref502
  doi: 10.1145/2809695.2809718
– ident: ref525
  doi: 10.1145/3371154
– start-page: 108
  year: 2007
  ident: ref388
  article-title: A survey of artificial intelligence for prognostics
  publication-title: Proc AAAI Fall Symp Artif Intell Prognostics
– ident: ref55
  doi: 10.1002/ett.3942
– ident: ref273
  doi: 10.1109/TNET.2019.2936939
– ident: ref182
  doi: 10.1109/TNSE.2021.3053588
– ident: ref246
  doi: 10.1145/3154448.3154452
– year: 2017
  ident: ref266
  article-title: Data-free knowledge distillation for deep neural networks
  publication-title: arXiv 1710 07535
– ident: ref453
  doi: 10.1109/MCI.2006.329691
– ident: ref474
  doi: 10.1145/2413176.2413189
– start-page: 1
  year: 2018
  ident: ref136
  article-title: MODI: Mobile deep inference made efficient by edge computing
  publication-title: Proc USENIX Workshop Hot Topics Edge Comput (HotEdge)
– year: 2015
  ident: ref2
  article-title: DeepID3: Face recognition with very deep neural networks
  publication-title: arXiv 1502 00873
– ident: ref500
  doi: 10.1145/3090076
– ident: ref21
  doi: 10.1145/3154815
– ident: ref287
  doi: 10.1109/TCSVT.2020.2996231
– ident: ref229
  doi: 10.1109/CVPR.2019.00293
– ident: ref446
  doi: 10.1109/CySWater.2016.7469060
– year: 2014
  ident: ref423
  article-title: Generative adversarial networks
  publication-title: arXiv 1406 2661
– ident: ref186
  doi: 10.1145/3133956.3133982
– ident: ref178
  doi: 10.1109/JIOT.2020.2967772
– year: 2018
  ident: ref225
  article-title: MobiFace: A lightweight deep learning face recognition on mobile devices
  publication-title: arXiv 1811 11080
– ident: ref97
  doi: 10.1109/ICDCS.2019.00106
– ident: ref267
  doi: 10.1109/CVPR.2015.7298594
– start-page: 1135
  year: 2015
  ident: ref277
  article-title: Learning both weights and connections for efficient neural network
  publication-title: Proc Adv Neural Inf Process Syst
– year: 2015
  ident: ref316
  article-title: GPU-based acceleration of deep convolutional neural networks on mobile platforms
  publication-title: Distrib Parallel Clust Comput
– ident: ref278
  doi: 10.1109/CVPR.2018.00171
– ident: ref406
  doi: 10.1007/978-3-642-38652-7_2
– year: 2019
  ident: ref214
  article-title: BrainTorrent: A peer-to-peer environment for decentralized federated learning
  publication-title: arXiv 1905 06731
– ident: ref358
  doi: 10.1109/ISCIT.2017.8261188
– ident: ref88
  doi: 10.1016/j.procs.2015.05.038
– start-page: 92
  year: 2018
  ident: ref213
  article-title: Multi-institutional deep learning modeling without sharing patient data: A feasibility study on brain tumor segmentation
  publication-title: International Workshop on Brainlesion Glioma Multiple Sclerosis Stroke and Traumatic Brain Injuries
– start-page: 1097
  year: 2012
  ident: ref471
  article-title: ImageNet classification with deep convolutional neural networks
  publication-title: Proc Adv Neural Inf Process Syst
– year: 2019
  ident: ref464
  publication-title: Google Glass
– ident: ref224
  doi: 10.1007/978-3-319-97909-0_46
– ident: ref544
  doi: 10.1023/A:1019956318069
SSID ssj0003563
Score 2.6988194
Snippet Edge intelligence refers to a set of connected systems and devices for data collection, caching, processing, and analysis proximity to where data are captured...
SourceID proquest
crossref
ieee
SourceType Aggregation Database
Enrichment Source
Index Database
Publisher
StartPage 1778
SubjectTerms Artificial intelligence
Artificial intelligence (AI)
Caching
Data collection
Data privacy
Data processing
edge caching
Edge computing
Electronic devices
inference
Inference algorithms
model training
offloading
Systematics
Taxonomy
Training data
Title Edge Intelligence: Empowering Intelligence to the Edge of Network
URI https://ieeexplore.ieee.org/document/9596610
https://www.proquest.com/docview/2592624937
Volume 109
WOSCitedRecordID wos000714203700008&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 Electronic Library (IEL)
  customDbUrl:
  eissn: 1558-2256
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0003563
  issn: 0018-9219
  databaseCode: RIE
  dateStart: 19630101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3PS8MwFH7M4UEP_pridEoO3rSuSZum8TbGhnqYQxR2K2maiKDr2Dr_fpO0KxuK4K2QvNJ-6Uvea_K-D-DKRAkypYR6aRBqL4yIlXnRwuOMxNhyTxIRO7EJNhrFkwkfN-CmroVRSrnDZ-rWXrq9_CyXS_urrMupCc5tPdUWY1FZq1XPugGtVNOwcWDjhqsCGZ93H8fPT32TChJsMlRbkuxvLEJOVeXHVOzWl-H-_57sAPaqOBL1yoE_hIaaHsHuGrtgC3qD7E2hhzXOzTs0-JxZWTTTvtGAihyZSBA5i1yjUXk4_Bheh4OX_r1XKSZ4knBaeCmlmduo1NoEUjjTgQ6ZH5GQBUqkltsrMu0hpQxzrf00Da2d8VstRezLNDiB5jSfqlNAJGIBVlhKGYmQScVxJuJMaCWClAqJ24BXECayohO3qhYfiUsrfJ442BMLe1LB3obr2mZWkmn82btlga57Vhi3obMaqaTyt0VikjhiXtPEWme_W53Djr13WUXYgWYxX6oL2JZfxftifuk-pW9kMcPv
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1bS8MwFD6MKagP3qY4ndoH37TapE0vvo2xsemsQybsLaRpIoKuYxd_v0nalQ1F8K2QHNp-6UnOaXK-D-BKRQk8IZjYietJ2_OxlnmRzI4CHCLNPYlZaMQmgjgOR6NoUIGbshZGCGEOn4lbfWn28tOML_SvsruIqOBc11NtaOWsolqrnHddUuimIeXCyhGXJTJOdPcweHluqWQQI5Wj6qJkZ20ZMroqPyZjs8J09v73bPuwW0SSVjMf-gOoiPEh7KzwC9ag2U7fhNVbYd28t9qfEy2MptrXGqx5ZqlY0DIWmbTi_Hj4Ebx22sNW1y40E2yOIzK3E0JSs1UppQqlUCpd6QWOj73AFSzR7F6-avcICVAkpZMknrZTnis5Cx2euMdQHWdjcQIW9gMXCcQ595kXcBGhlIUpk4K5CWEc1QEtIaS8IBTXuhYf1CQWTkQN7FTDTgvY63Bd2kxyOo0_e9c00GXPAuM6NJYjRQuPm1GVxmH1miraOv3d6hK2usOnPu334scz2Nb3yWsKG1CdTxfiHDb51_x9Nr0wn9U3xtnHOA
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=Edge+Intelligence%3A+Empowering+Intelligence+to+the+Edge+of+Network&rft.jtitle=Proceedings+of+the+IEEE&rft.au=Xu%2C+Dianlei&rft.au=Li%2C+Tong&rft.au=Li%2C+Yong&rft.au=Su%2C+Xiang&rft.date=2021-11-01&rft.pub=The+Institute+of+Electrical+and+Electronics+Engineers%2C+Inc.+%28IEEE%29&rft.issn=0018-9219&rft.eissn=1558-2256&rft.volume=109&rft.issue=11&rft.spage=1778&rft_id=info:doi/10.1109%2FJPROC.2021.3119950&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0018-9219&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0018-9219&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0018-9219&client=summon