Computational Complexity Optimization of Neural Network-Based Equalizers in Digital Signal Processing: A Comprehensive Approach

Experimental results based on offline processing reported at optical conferences increasingly rely on neural network-based equalizers for accurate data recovery. However, achieving low-complexity implementations that are efficient for real-time digital signal processing remains a challenge. This pap...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	Journal of lightwave technology Ročník 42; číslo 12; s. 4177 - 4201
Hlavní autori:	Freire, Pedro, Srivallapanondh, Sasipim, Spinnler, Bernhard, Napoli, Antonio, Costa, Nelson, Prilepsky, Jaroslaw E., Turitsyn, Sergei K.
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	New York IEEE 15.06.2024 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Predmet:	Artificial neural networks Communications systems Complexity computational complexity Data recovery Digital signal processing Equalizers Hardware hardware estimation Inference Measurement Neural networks nonlinear equalizer Optimization Real time Signal processing Software Task analysis Training
ISSN:	0733-8724, 1558-2213
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Abstract	Experimental results based on offline processing reported at optical conferences increasingly rely on neural network-based equalizers for accurate data recovery. However, achieving low-complexity implementations that are efficient for real-time digital signal processing remains a challenge. This paper addresses this critical need by proposing a systematic approach to designing and evaluating low-complexity neural network equalizers. Our approach focuses on three key phases: training, inference, and hardware synthesis. We provide a comprehensive review of existing methods for reducing complexity in each phase, enabling informed choices during design. For the training and inference phases, we introduce a novel methodology for quantifying complexity. This includes new metrics that bridge software-to-hardware considerations, revealing the relationship between complexity and specific neural network architectures and hyperparameters. We guide the calculation of these metrics for both feed-forward and recurrent layers, highlighting the appropriate choice depending on the application's focus (software or hardware). Finally, to demonstrate the practical benefits of our approach, we showcase how the computational complexity of neural network equalizers can be significantly reduced and measured for both teacher (biLSTM+CNN) and student (1D-CNN) architectures in different scenarios. This work aims to standardize the estimation and optimization of computational complexity for neural networks applied to real-time digital signal processing, paving the way for more efficient and deployable optical communication systems.
AbstractList	Experimental results based on offline processing reported at optical conferences increasingly rely on neural network-based equalizers for accurate data recovery. However, achieving low-complexity implementations that are efficient for real-time digital signal processing remains a challenge. This paper addresses this critical need by proposing a systematic approach to designing and evaluating low-complexity neural network equalizers. Our approach focuses on three key phases: training, inference, and hardware synthesis. We provide a comprehensive review of existing methods for reducing complexity in each phase, enabling informed choices during design. For the training and inference phases, we introduce a novel methodology for quantifying complexity. This includes new metrics that bridge software-to-hardware considerations, revealing the relationship between complexity and specific neural network architectures and hyperparameters. We guide the calculation of these metrics for both feed-forward and recurrent layers, highlighting the appropriate choice depending on the application's focus (software or hardware). Finally, to demonstrate the practical benefits of our approach, we showcase how the computational complexity of neural network equalizers can be significantly reduced and measured for both teacher (biLSTM+CNN) and student (1D-CNN) architectures in different scenarios. This work aims to standardize the estimation and optimization of computational complexity for neural networks applied to real-time digital signal processing, paving the way for more efficient and deployable optical communication systems.
Author	Srivallapanondh, Sasipim Turitsyn, Sergei K. Prilepsky, Jaroslaw E. Freire, Pedro Napoli, Antonio Spinnler, Bernhard Costa, Nelson
Author_xml	– sequence: 1 givenname: Pedro orcidid: 0000-0003-3145-1018 surname: Freire fullname: Freire, Pedro email: p.freiredecarvalhosourza@aston.ac.uk organization: Aston Institute of Photonic Technologies, Aston University, Birmingham, U.K – sequence: 2 givenname: Sasipim orcidid: 0009-0001-9465-8801 surname: Srivallapanondh fullname: Srivallapanondh, Sasipim organization: Aston Institute of Photonic Technologies, Aston University, Birmingham, U.K – sequence: 3 givenname: Bernhard orcidid: 0000-0001-9578-0297 surname: Spinnler fullname: Spinnler, Bernhard email: anapoli@infinera.com organization: Infinera R&D, Munich, Germany – sequence: 4 givenname: Antonio orcidid: 0000-0002-9264-9274 surname: Napoli fullname: Napoli, Antonio organization: Infinera R&D, Munich, Germany – sequence: 5 givenname: Nelson orcidid: 0000-0002-8678-5691 surname: Costa fullname: Costa, Nelson email: ncosta@infinera.com organization: Infinera Unipessoal, Carnaxide, Portugal – sequence: 6 givenname: Jaroslaw E. orcidid: 0000-0002-3035-4112 surname: Prilepsky fullname: Prilepsky, Jaroslaw E. organization: Aston Institute of Photonic Technologies, Aston University, Birmingham, U.K – sequence: 7 givenname: Sergei K. orcidid: 0000-0003-0101-3834 surname: Turitsyn fullname: Turitsyn, Sergei K. organization: Aston Institute of Photonic Technologies, Aston University, Birmingham, U.K
BookMark	eNp9kDtPwzAURi0EEuWxMzBYYk7xK3HCVkp5qSpIwBw5yU1xSeNgOzy68NdJGwbEwGRd-XzfvTp7aLs2NSB0RMmQUpKc3k4fh4wwMeQ8juI42kIDGoZxwBjl22hAJOdBLJnYRXvOLQihQsRygL7GZtm0XnltalXh9VTBh_af-K7xeqlXmx9sSjyD1nbEDPy7sS_BuXJQ4Mlrqyq9AuuwrvGFnmvfMQ96vi67tyYH53Q9P8OjTbWFZ6idfgM8ahprVP58gHZKVTk4_Hn30dPl5HF8HUzvrm7Go2mQcy58kCtVCBqFkaJSlJAXXIgsVlQULEtEBhAWTEqQJJFRGYqMJFGY5aXKKGe05BnfRyd9b7f2tQXn04VpbXelSzmRoehcCdFRUU_l1jhnoUxz3bvxVukqpSRdy0472eladvojuwuSP8HG6qWyn_9FjvuIBoBfuEgiKin_BpLujss
CODEN	JLTEDG
CitedBy_id	crossref_primary_10_3390_math12111755 crossref_primary_10_1109_ACCESS_2024_3482005 crossref_primary_10_1109_ACCESS_2025_3584331 crossref_primary_10_1016_j_neucom_2025_131373 crossref_primary_10_1049_gtd2_70102 crossref_primary_10_5194_npg_31_535_2024 crossref_primary_10_1016_j_icte_2025_04_016 crossref_primary_10_1364_AO_545671 crossref_primary_10_1109_JLT_2024_3487204 crossref_primary_10_1016_j_optcom_2024_131418 crossref_primary_10_1109_ACCESS_2024_3413717 crossref_primary_10_1109_JLT_2024_3451242 crossref_primary_10_3103_S1068335624601602 crossref_primary_10_1021_acs_jcim_5c00267 crossref_primary_10_1108_JM2_05_2025_0199 crossref_primary_10_1016_j_future_2025_107777 crossref_primary_10_3390_biomimetics10090617 crossref_primary_10_1109_OJCOMS_2025_3586560 crossref_primary_10_3390_s25154825 crossref_primary_10_1016_j_phycom_2024_102576 crossref_primary_10_1016_j_asr_2024_04_055 crossref_primary_10_1109_LPT_2024_3424495 crossref_primary_10_1109_LAWP_2025_3539500 crossref_primary_10_1016_j_engappai_2025_110395 crossref_primary_10_1016_j_iot_2025_101493 crossref_primary_10_1109_JLT_2024_3519782 crossref_primary_10_1109_JLT_2025_3599665 crossref_primary_10_1109_ACCESS_2024_3491499 crossref_primary_10_1109_JLT_2025_3550277
Cites_doi	10.1109/ACCESS.2021.3079639 10.1007/978-3-7908-1866-6_11 10.1016/j.protcy.2013.12.159 10.5220/0010606100002992 10.3389/frai.2021.676564 10.1016/S0165-1684(00)00030-X 10.1145/3527156 10.1016/S0005-1098(00)00050-9 10.1109/72.279181 10.1364/ofc.2021.m3g.2 10.1364/OFC.2022.Th2A.35 10.23919/OFC49934.2023.10116725 10.1109/ECOC52684.2021.9605870 10.1007/s00163-020-00336-7 10.1049/icp.2023.2276 10.21437/Interspeech.2017-1164 10.1109/JLT.2015.2508502 10.1007/978-3-030-49076-8_3 10.1364/OFC.2023.M1F.5 10.1016/0098-3004(93)90090-R 10.1109/ICASSP.1995.480470 10.1109/JLT.2023.3234327 10.1109/JLT.2023.3272011 10.1016/j.pce.2006.03.020 10.2307/2003354 10.1609/aaai.v35i12.17286 10.1016/j.procs.2020.01.079 10.1109/CVPR.2017.643 10.1109/JLT.2021.3056869 10.48550/arxiv.1811.06965 10.1109/MWSCAS.2017.8053243 10.48550/ARXIV.1706.03762 10.1049/cp.2019.0947 10.1038/s41598-022-12563-0 10.1109/CompComm.2016.7925210 10.1038/s41377-022-00717-8 10.1109/JLT.2021.3096286 10.1137/07070111X 10.3390/jimaging9020046 10.1109/ICCD.2006.4380833 10.1145/3444943 10.1109/IROS.2017.8202133 10.1109/ISVLSI49217.2020.00027 10.1007/BF02457822 10.1109/TC.2010.200 10.1162/neco.1997.9.8.1735 10.1109/CVPR.2019.00881 10.1109/TSMCC.2009.2038279 10.1142/S0218126603001045 10.1364/OE.423747 10.1162/089976600300015015 10.1109/ECOC48923.2020.9333417 10.1109/JSTQE.2009.2035931 10.1109/JLT.2020.2973718 10.1109/18.605580 10.1007/s12200-022-00013-8 10.1109/TNNLS.2017.2766162 10.1109/72.182695 10.1109/JLT.2021.3051609 10.1038/s41467-018-07210-0 10.1109/JLT.2020.2991028 10.1364/OE.26.032765 10.1109/82.298385 10.1364/OE.463450 10.1109/5.720251 10.1109/82.539000 10.1109/CVPRW53098.2021.00268 10.1109/JLT.2023.3337604 10.1109/JPROC.2017.2761740 10.1145/1167350.1167440 10.1109/JLT.2021.3108006 10.1109/JSTQE.2022.3174268 10.1109/JLT.2014.2301492 10.1109/ARITH.2007.24 10.1109/TNNLS.2019.2910073 10.1016/j.ymssp.2020.107398 10.1109/LPT.2014.2375960 10.1109/5.726790 10.1109/MSP.2007.361611 10.1109/CIT.2008.Workshops.65 10.1016/j.neunet.2007.04.016 10.1109/CVPR.2016.90 10.1007/978-3-319-91734-4_4 10.1145/2858965.2814290 10.1109/JLT.2021.3092415 10.1364/OE.27.002387 10.1038/s41586-020-2764-0 10.1007/978-1-4757-2370-0
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024
DBID	97E ESBDL RIA RIE AAYXX CITATION 7SP 7U5 8FD H8D L7M
DOI	10.1109/JLT.2024.3386886
DatabaseName	IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE Xplore Open Access Journals IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef Electronics & Communications Abstracts Solid State and Superconductivity Abstracts Technology Research Database Aerospace Database Advanced Technologies Database with Aerospace
DatabaseTitle	CrossRef Aerospace Database Solid State and Superconductivity Abstracts Technology Research Database Advanced Technologies Database with Aerospace Electronics & Communications Abstracts
DatabaseTitleList	Aerospace Database
Database_xml	– sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Applied Sciences Physics
EISSN	1558-2213
EndPage	4201
ExternalDocumentID	10_1109_JLT_2024_3386886 10496171
Genre	orig-research
GrantInformation_xml	– fundername: EU Horizon 2020 Program – fundername: European Union' Horizon Europe Research and Innovation Programme grantid: 101092766 – fundername: EPSRC Project TRANSNET – fundername: Leverhulme Trust grantid: RP-2018-063 funderid: 10.13039/501100000275 – fundername: Marie Sklodowska-Curie grantid: 813144; 956713
GroupedDBID	-~X 0R~ 29K 4.4 5GY 6IK 85S 8SL 97E AAJGR AARMG AASAJ AAWJZ AAWTH ABAZT ABQJQ ABVLG ACBEA ACGFO ACGFS ACIWK AEDJG AENEX AGQYO AHBIQ AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ATHME ATWAV AYPRP AZSQR BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 D-I DSZJF DU5 EBS ESBDL HZ~ IFIPE IPLJI JAVBF LAI M43 O9- OCL OFLFD OPJBK P2P RIA RIE RNS ROL ROS TN5 TR6 ZCA AAYXX CITATION 7SP 7U5 8FD H8D L7M
ID	FETCH-LOGICAL-c334t-caad41656a174fecd344b8a14d2b94bee5d277e70976f54b0965bcfab1321f3b3
IEDL.DBID	RIE
ISICitedReferencesCount	45
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001263599200010&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0733-8724
IngestDate	Mon Jun 30 10:12:59 EDT 2025 Tue Nov 18 22:06:27 EST 2025 Sat Nov 29 02:11:48 EST 2025 Wed Aug 27 02:05:21 EDT 2025
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	12
Language	English
License	https://creativecommons.org/licenses/by-nc-nd/4.0
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c334t-caad41656a174fecd344b8a14d2b94bee5d277e70976f54b0965bcfab1321f3b3
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0001-9578-0297 0000-0002-3035-4112 0000-0002-8678-5691 0000-0002-9264-9274 0000-0003-0101-3834 0009-0001-9465-8801 0000-0003-3145-1018
OpenAccessLink	https://ieeexplore.ieee.org/document/10496171
PQID	3075415544
PQPubID	85485
PageCount	25
ParticipantIDs	proquest_journals_3075415544 crossref_citationtrail_10_1109_JLT_2024_3386886 ieee_primary_10496171 crossref_primary_10_1109_JLT_2024_3386886
PublicationCentury	2000
PublicationDate	2024-06-15
PublicationDateYYYYMMDD	2024-06-15
PublicationDate_xml	– month: 06 year: 2024 text: 2024-06-15 day: 15
PublicationDecade	2020
PublicationPlace	New York
PublicationPlace_xml	– name: New York
PublicationTitle	Journal of lightwave technology
PublicationTitleAbbrev	JLT
PublicationYear	2024
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	ref57 ref56 ref58 ref53 ref52 Veit (ref106) 2016 Malkoff (ref18) 1989 ref54 Baalen (ref24) 2020; 33 Voon (ref123) 2021; 10 Staff (ref73) 1997; 59 ref51 ref50 ref46 ref45 ref48 ref47 ref42 ref44 ref43 Wu (ref72) 2018 Pedro (ref88) 2024 Gysel (ref20) 2016 ref49 ref8 Padmajarani (ref90) 2015; 2 ref9 ref4 ref3 Jahani (ref70) 2009 ref6 ref5 Nichols (ref59) 2021 ref101 ref40 ref35 ref34 ref37 Hubara (ref115) ref31 ref30 ref33 ref32 You (ref77) 2020 ref38 Li (ref121) 2019 Sun (ref103) 2020 Zhou (ref120) 2017 ref23 Lipton (ref96) 2015 ref26 ref25 ref22 ref21 ref28 ref27 Finn (ref41) 2017 (ref83) 2021 Xu (ref36) 2020 Luo (ref12) 1998 Banner (ref113) 2018; 31 ref13 ref15 ref14 ref97 ref11 ref99 ref10 ref98 ref17 ref16 ref19 Han (ref122) 2015 ref93 ref92 ref95 ref91 ref89 Looks (ref62) 2017 ref86 ref85 ref87 Przewlocka-Rus (ref94) 2022 Goyal (ref119) 2021 Kahrs (ref7) 1998 Ba (ref108) 2016 ref82 ref81 ref84 ref80 ref79 ref78 ref109 Maass (ref29) 1994 ref107 ref75 ref104 ref105 ref102 ref76 Chang (ref117) 2020 ref2 ref1 Taras (ref55) 2018 ref71 ref111 ref112 ref110 ref68 ref67 Li (ref74) 2019 ref69 ref118 ref64 ref63 ref116 ref66 Freire (ref39) 2022 ref65 ref114 Wu (ref100) 2018 ref60 ref61
References_xml	– ident: ref4 doi: 10.1109/ACCESS.2021.3079639 – ident: ref8 doi: 10.1007/978-3-7908-1866-6_11 – ident: ref35 doi: 10.1016/j.protcy.2013.12.159 – year: 2020 ident: ref117 article-title: MSP: An FPGA-specific mixed-scheme, multi-precision deep neural network quantization framework – ident: ref65 doi: 10.5220/0010606100002992 – ident: ref71 doi: 10.3389/frai.2021.676564 – ident: ref13 doi: 10.1016/S0165-1684(00)00030-X – ident: ref60 doi: 10.1145/3527156 – ident: ref34 doi: 10.1016/S0005-1098(00)00050-9 – ident: ref95 doi: 10.1109/72.279181 – year: 2016 ident: ref108 article-title: Layer normalization – ident: ref58 doi: 10.1364/ofc.2021.m3g.2 – year: 2024 ident: ref88 article-title: Code to estimate computational complexity of neural network applications in signal processing – volume-title: Proc. Opt. Fiber Commun. Conf. year: 2022 ident: ref39 article-title: Domain adaptation: The key enabler of neural network equalizers in coherent optical systems doi: 10.1364/OFC.2022.Th2A.35 – start-page: 550 volume-title: Proc. 30th Int. Conf. Neural Inform. Process. Syst. year: 2016 ident: ref106 article-title: Residual networks behave like ensembles of relatively shallow networks – ident: ref54 doi: 10.23919/OFC49934.2023.10116725 – ident: ref75 doi: 10.1109/ECOC52684.2021.9605870 – ident: ref30 doi: 10.1007/s00163-020-00336-7 – ident: ref43 doi: 10.1049/icp.2023.2276 – year: 2015 ident: ref96 article-title: A critical review of recurrent neural networks for sequence learning – start-page: 4466 volume-title: Proc. Int. Conf. Mach. Learn. ident: ref115 article-title: Accurate post training quantization with small calibration sets – ident: ref21 doi: 10.21437/Interspeech.2017-1164 – volume: 10 start-page: 12 issue: 1 year: 2021 ident: ref123 article-title: Performance analysis of CPU, GPU and TPU for deep learning applications publication-title: Int. J. Des., Anal. Tools Intergrated Circuits Syst. – ident: ref14 doi: 10.1109/JLT.2015.2508502 – ident: ref111 doi: 10.1007/978-3-030-49076-8_3 – volume: 31 volume-title: Proc. Adv. Neural Inf. Process. Syst. year: 2018 ident: ref113 article-title: Scalable methods for 8-bit training of neural networks – volume-title: Applied Neural Networks for Signal Processing year: 1998 ident: ref12 – ident: ref42 doi: 10.1364/OFC.2023.M1F.5 – year: 2015 ident: ref122 article-title: Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding – ident: ref47 doi: 10.1016/0098-3004(93)90090-R – ident: ref78 doi: 10.1109/ICASSP.1995.480470 – ident: ref15 doi: 10.1109/JLT.2023.3234327 – ident: ref56 doi: 10.1109/JLT.2023.3272011 – ident: ref64 doi: 10.1016/j.pce.2006.03.020 – year: 2017 ident: ref120 article-title: Incremental network quantization: Towards lossless CNNs with low-precision weights – year: 2019 ident: ref121 article-title: Additive powers-of-two quantization: An efficient non-uniform discretization for neural networks – volume: 33 start-page: 5741 year: 2020 ident: ref24 article-title: Bayesian bits: Unifying quantization and pruning publication-title: Adv. Neural Inf. Process. Syst. – ident: ref110 doi: 10.2307/2003354 – ident: ref112 doi: 10.1609/aaai.v35i12.17286 – ident: ref46 doi: 10.1016/j.procs.2020.01.079 – ident: ref22 doi: 10.1109/CVPR.2017.643 – ident: ref16 doi: 10.1109/JLT.2021.3056869 – ident: ref63 doi: 10.48550/arxiv.1811.06965 – ident: ref99 doi: 10.1109/MWSCAS.2017.8053243 – ident: ref107 doi: 10.48550/ARXIV.1706.03762 – year: 2017 ident: ref62 article-title: Deep learning with dynamic computation graphs – volume-title: Applications of Digital Signal Processing to Audio and Acoustics year: 1998 ident: ref7 – ident: ref52 doi: 10.1049/cp.2019.0947 – ident: ref49 doi: 10.1038/s41598-022-12563-0 – ident: ref82 doi: 10.1109/CompComm.2016.7925210 – ident: ref85 doi: 10.1038/s41377-022-00717-8 – year: 2009 ident: ref70 article-title: ZOT-MK: A new algorithm for big integer multiplication – year: 2020 ident: ref103 article-title: A review of designs and applications of echo state networks – ident: ref28 doi: 10.1109/JLT.2021.3096286 – ident: ref51 doi: 10.1137/07070111X – year: 2022 ident: ref94 article-title: Power-of-two quantization for low bitwidth and hardware compliant neural networks – ident: ref44 doi: 10.3390/jimaging9020046 – volume: 2 start-page: 1 issue: 9 year: 2015 ident: ref90 article-title: FPGA implementation of multiplier using shift and add technique publication-title: Int. J. Adv. Electron. Comput. Sci. – ident: ref69 doi: 10.1109/ICCD.2006.4380833 – ident: ref25 doi: 10.1145/3444943 – ident: ref38 doi: 10.1109/IROS.2017.8202133 – ident: ref57 doi: 10.1109/ISVLSI49217.2020.00027 – ident: ref1 doi: 10.1007/BF02457822 – year: 2021 ident: ref83 article-title: Ultrascale architecture DSP slice – ident: ref91 doi: 10.1109/TC.2010.200 – ident: ref97 doi: 10.1162/neco.1997.9.8.1735 – year: 2020 ident: ref77 article-title: Shiftaddnet: A hardware-inspired deep network – ident: ref114 doi: 10.1109/CVPR.2019.00881 – ident: ref6 doi: 10.1109/TSMCC.2009.2038279 – year: 2019 ident: ref74 article-title: Additive powers-of-two quantization: An efficient non-uniform discretization for neural networks – ident: ref80 doi: 10.1142/S0218126603001045 – year: 2021 ident: ref119 article-title: Fixed-point quantization of convolutional neural networks for quantized inference on embedded platforms – ident: ref84 doi: 10.1364/OE.423747 – ident: ref98 doi: 10.1162/089976600300015015 – ident: ref45 doi: 10.1109/ECOC48923.2020.9333417 – ident: ref50 doi: 10.1109/ECOC52684.2021.9605870 – ident: ref68 doi: 10.1109/JSTQE.2009.2035931 – ident: ref48 doi: 10.1109/JLT.2020.2973718 – ident: ref23 doi: 10.1109/18.605580 – ident: ref87 doi: 10.1007/s12200-022-00013-8 – ident: ref102 doi: 10.1109/TNNLS.2017.2766162 – ident: ref116 doi: 10.1109/72.182695 – start-page: 5363 volume-title: Proc. Int. Conf. Mach. Learn. year: 2018 ident: ref72 article-title: Deep k-means: Re-training and parameter sharing with harder cluster assignments for compressing deep convolutions – ident: ref109 doi: 10.1109/JLT.2021.3051609 – ident: ref9 doi: 10.1038/s41467-018-07210-0 – year: 2018 ident: ref100 article-title: On the statistical challenges of echo state networks and some potential remedies – ident: ref40 doi: 10.1109/JLT.2020.2991028 – ident: ref27 doi: 10.1364/OE.26.032765 – ident: ref79 doi: 10.1109/82.298385 – ident: ref53 doi: 10.1364/OE.463450 – volume: 59 year: 1997 ident: ref73 article-title: Gate count capacity metrics for FPGAS – ident: ref5 doi: 10.1109/5.720251 – ident: ref92 doi: 10.1109/82.539000 – start-page: 248 volume-title: Proc. 2nd Int. Conf. Neural Inform. Process. Syst. year: 1989 ident: ref18 article-title: A neural network for real-time signal processing – ident: ref76 doi: 10.1109/CVPRW53098.2021.00268 – ident: ref118 doi: 10.1109/JLT.2023.3337604 – year: 2018 ident: ref55 article-title: Quantization error as a metric for dynamic precision scaling in neural net training – ident: ref19 doi: 10.1109/JPROC.2017.2761740 – ident: ref33 doi: 10.1145/1167350.1167440 – ident: ref37 doi: 10.1109/JLT.2021.3108006 – start-page: 1 volume-title: Proc. IEEE Opt. Fiber Commun. Conf. Exhib. year: 2020 ident: ref36 article-title: Transfer learning aided neural networks for nonlinear equalization in short-reach direct detection systems – ident: ref66 doi: 10.1109/JSTQE.2022.3174268 – start-page: 1126 volume-title: Proc. Int. Conf. Mach. Learn. year: 2017 ident: ref41 article-title: Model-agnostic meta-learning for fast adaptation of deep networks – ident: ref17 doi: 10.1109/JLT.2014.2301492 – year: 2021 ident: ref59 article-title: A survey and empirical evaluation of parallel deep learning frameworks – year: 2016 ident: ref20 article-title: Hardware-oriented approximation of convolutional neural networks – ident: ref89 doi: 10.3389/frai.2021.676564 – ident: ref93 doi: 10.1109/ARITH.2007.24 – ident: ref31 doi: 10.1109/TNNLS.2019.2910073 – ident: ref3 doi: 10.1016/j.ymssp.2020.107398 – ident: ref11 doi: 10.1109/LPT.2014.2375960 – ident: ref2 doi: 10.1109/5.726790 – ident: ref67 doi: 10.1109/MSP.2007.361611 – start-page: 183 volume-title: Proc. 7th Int. Conf. Neural Inform. Process. Syst. year: 1994 ident: ref29 article-title: On the computational complexity of networks of spiking neurons – ident: ref32 doi: 10.1109/CIT.2008.Workshops.65 – ident: ref104 doi: 10.1016/j.neunet.2007.04.016 – ident: ref105 doi: 10.1109/CVPR.2016.90 – ident: ref10 doi: 10.1007/978-3-319-91734-4_4 – ident: ref61 doi: 10.1145/2858965.2814290 – ident: ref26 doi: 10.1109/JLT.2021.3092415 – ident: ref101 doi: 10.1364/OE.27.002387 – ident: ref86 doi: 10.1038/s41586-020-2764-0 – ident: ref81 doi: 10.1007/978-1-4757-2370-0
SSID	ssj0014487
Score	2.6248076
Snippet	Experimental results based on offline processing reported at optical conferences increasingly rely on neural network-based equalizers for accurate data...
SourceID	proquest crossref ieee
SourceType	Aggregation Database Enrichment Source Index Database Publisher
StartPage	4177
SubjectTerms	Artificial neural networks Communications systems Complexity computational complexity Data recovery Digital signal processing Equalizers Hardware hardware estimation Inference Measurement Neural networks nonlinear equalizer Optimization Real time Signal processing Software Task analysis Training
Title	Computational Complexity Optimization of Neural Network-Based Equalizers in Digital Signal Processing: A Comprehensive Approach
URI	https://ieeexplore.ieee.org/document/10496171 https://www.proquest.com/docview/3075415544
Volume	42
WOSCitedRecordID	wos001263599200010&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: 1558-2213 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014487 issn: 0733-8724 databaseCode: RIE dateStart: 19830101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEA4qCl58i_VFDl48bN1tZptdb_WFiFTBCr0tea0WtJW2CnrxrzuTTYsgCt52IQkL305mJpnvG8YOHMmQZaoZmcTmEZgyjRSUJkoM6Z0rqnuPfbMJ2W5n3W5-G8jqngvjnPPFZ65Oj_4u3w7MKx2VoYVDjh4Xk51ZKZsVWWt6ZYB5hudGSyHQxBswuZOM86Or6w5mgg2oYz7WzIg2_c0H-aYqP3Zi714ulv_5YStsKcSRvFUBv8pmXH-NLYeYkgeLHa2xBV_iaUbr7LNq4BAO_zi9kRjm-J3f4LbxHPiYfFByEuzAEe2qQjw6QUdn-bmnX35gtMh7fX7We6BuI_yu90CLBboBusFj3vJLD91jVRrPW0G1fIPdX5x3Ti-j0H4hMkLAODJKWSBxHoVZS-mMFQA6UwnYhs5BO5fahpROxhjRlClo0pHRplQaE9ykFFpssrn-oO-2GC-dVSpzOrdCg5CxikFDgv5TSJUaSGrsaAJIYYI2ObXIeCp8jhLnBUJYEIRFgLDGDqczXipdjj_GbhBk38ZVaNXY7gT0IljuqMA9L_VBFmz_Mm2HLdLqVC-WpLtsbjx8dXts3ryNe6Phvv8pvwAnO-Cq
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LT9tAEB4hWgSXprzUUCh74MLBYGfH2O4tbUE8QkAiSNysfRkilQTlUalc-OudWW8QEqISN1vaXVv6PDsz3vm-AdhxLEOWq4PIJLaI0FRppLAyUWJY71xx3Xvsm01k3W5-c1NcBrK658I453zxmdvjS3-Wb4dmyr_KyMKxII9Lyc4Hbp0V6FrPhwaUaXh2dCYlGXkLZ6eScbF_2ulRLtjCPcrIDnImTr_wQr6tyqu92DuYo8Y7X-0zfAqRpGjX0C_DnBusQCNElSLY7HgFFnyRpxmvwlPdwiH8_hN8x3KYk7_igjaO-8DIFMNKsGQHjejWNeLRD3J1Vhx6AuYjxYuiPxC_-rfcb0Rc9W95sUA4IEf4XbT90iN3VxfHi3bQLV-D66PD3s_jKDRgiIyUOImMUhZZnkdR3lI5YyWizlWCtqUL1M6ltpVlLosppqlS1Kwko02lNKW4SSW1XIf5wXDgvoConFUqd7qwUqPMYhWjxoQwlJlKDSZN2J8BUpqgTs5NMn6XPkuJi5IgLBnCMkDYhN3nGQ-1Msd_xq4xZC_G1Wg1YXMGehlsd1zSrpf6MAs33pi2DYvHvfNO2Tnpnn2FJX4SV48l6SbMT0ZTtwUfzZ9Jfzz65j_Qf_8E4_M
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=Computational+Complexity+Optimization+of+Neural+Network-Based+Equalizers+in+Digital+Signal+Processing%3A+A+Comprehensive+Approach&rft.jtitle=Journal+of+lightwave+technology&rft.au=Freire%2C+Pedro&rft.au=Srivallapanondh%2C+Sasipim&rft.au=Spinnler%2C+Bernhard&rft.au=Napoli%2C+Antonio&rft.date=2024-06-15&rft.issn=0733-8724&rft.eissn=1558-2213&rft.volume=42&rft.issue=12&rft.spage=4177&rft.epage=4201&rft_id=info:doi/10.1109%2FJLT.2024.3386886&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_JLT_2024_3386886
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0733-8724&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0733-8724&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0733-8724&client=summon