Field-programmable gate array implementation of efficient deep neural network architecture

Deep neural network (DNN) comprises multiple stages of data processing sub-systems with one of the primary sub-systems is a fully connected neural network (FCNN) model. This fully connected neural network model has multiple layers of neurons that need to be implemented using arithmetic units with su...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:International journal of electrical and computer engineering (Malacca, Malacca) Ročník 14; číslo 4; s. 3863
Hlavní autoři: Kumar Reddy, Pottipati Dileep, Ramanaiah, Kota Venkata
Médium: Journal Article
Jazyk:angličtina
Vydáno: 01.08.2024
ISSN:2088-8708, 2722-2578
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Deep neural network (DNN) comprises multiple stages of data processing sub-systems with one of the primary sub-systems is a fully connected neural network (FCNN) model. This fully connected neural network model has multiple layers of neurons that need to be implemented using arithmetic units with suitable number representation to optimize area, power, and speed. In this work, the network parameters are analyzed, and redundancy in weights is eliminated. A pipelined and parallel structure is designed for the fully connected network information. The proposed FCNN structure has 16 inputs, 3 hidden layers, and an output layer. Each hidden layer consists of 4 neurons and describes how the inputs are connected to hidden layer neurons to process the raw data. A hardware description language (HDL) model is developed for the proposed structure and the verified model is implemented on Xilinx field-programmable gate array (FPGA). The modified structure comprises registers, demultiplexers, weight registers, multipliers, adders, and read-only memory lookup table (ROM/LUT). The modified architecture implemented on FPGA is estimated to reduce area by 87.5% and improve timing by 3x compared with direct implementation methods.
AbstractList Deep neural network (DNN) comprises multiple stages of data processing sub-systems with one of the primary sub-systems is a fully connected neural network (FCNN) model. This fully connected neural network model has multiple layers of neurons that need to be implemented using arithmetic units with suitable number representation to optimize area, power, and speed. In this work, the network parameters are analyzed, and redundancy in weights is eliminated. A pipelined and parallel structure is designed for the fully connected network information. The proposed FCNN structure has 16 inputs, 3 hidden layers, and an output layer. Each hidden layer consists of 4 neurons and describes how the inputs are connected to hidden layer neurons to process the raw data. A hardware description language (HDL) model is developed for the proposed structure and the verified model is implemented on Xilinx field-programmable gate array (FPGA). The modified structure comprises registers, demultiplexers, weight registers, multipliers, adders, and read-only memory lookup table (ROM/LUT). The modified architecture implemented on FPGA is estimated to reduce area by 87.5% and improve timing by 3x compared with direct implementation methods.
Author Kumar Reddy, Pottipati Dileep
Ramanaiah, Kota Venkata
Author_xml – sequence: 1
  givenname: Pottipati Dileep
  orcidid: 0009-0008-2349-867X
  surname: Kumar Reddy
  fullname: Kumar Reddy, Pottipati Dileep
– sequence: 2
  givenname: Kota Venkata
  orcidid: 0009-0007-5371-3628
  surname: Ramanaiah
  fullname: Ramanaiah, Kota Venkata
BookMark eNot0M9OAyEQBnBiNLHWvgMvQAUWFvZoGqsmTbzoxQuBYbai-y_sVtO3d209fZPJN3P43ZDLru-QECr4Wghdibv0iYDrb6GSWg9DYcuCFdboC7KQRkomtbGX88ytZdZwe01W45gCV8oobkq9IO_bhE1kQ-732betDw3SvZ-Q-pz9kaZ2aLDFbvJT6jva1xTrOkGaNzQiDrTDQ_bNHNNPn7_mK_hIE8J0yHhLrmrfjLj6zyV52z68bp7Y7uXxeXO_YyDKSjMoqliZiOBlxBAArEYpdCFjkAGFitxKoYKoECOHkkOFZY3Ge9BzC2yxJPb8F3I_jhlrN-TU-nx0grsTkzsxuROTOzO5P6biF2rqY5U
ContentType Journal Article
DBID AAYXX
CITATION
DOI 10.11591/ijece.v14i4.pp3863-3875
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList CrossRef
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 2722-2578
ExternalDocumentID 10_11591_ijece_v14i4_pp3863_3875
GroupedDBID .4S
.DC
8FE
8FG
AAKDD
AAYXX
ABJCF
ABUWG
AFFHD
AFKRA
ALMA_UNASSIGNED_HOLDINGS
ARAPS
ARCSS
BENPR
BGLVJ
BPHCQ
BVBZV
CCPQU
CITATION
EOJEC
HCIFZ
I-F
K6V
K7-
KWQ
L6V
M7S
OBODZ
OK1
P62
PHGZM
PHGZT
PQGLB
PQQKQ
PROAC
PTHSS
TUS
ID FETCH-LOGICAL-c1695-c39d97deca2debbcc85e21532db2be14d08214b19eed0c60c9e6fe7aac5153c83
ISSN 2088-8708
IngestDate Sat Nov 29 02:39:47 EST 2025
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed false
IsScholarly true
Issue 4
Language English
License http://creativecommons.org/licenses/by-sa/4.0
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c1695-c39d97deca2debbcc85e21532db2be14d08214b19eed0c60c9e6fe7aac5153c83
ORCID 0009-0007-5371-3628
0009-0008-2349-867X
OpenAccessLink http://doi.org/10.11591/ijece.v14i4.pp3863-3875
ParticipantIDs crossref_primary_10_11591_ijece_v14i4_pp3863_3875
PublicationCentury 2000
PublicationDate 2024-08-01
PublicationDateYYYYMMDD 2024-08-01
PublicationDate_xml – month: 08
  year: 2024
  text: 2024-08-01
  day: 01
PublicationDecade 2020
PublicationTitle International journal of electrical and computer engineering (Malacca, Malacca)
PublicationYear 2024
SSID ssib044740765
ssj0000866295
Score 2.2824655
Snippet Deep neural network (DNN) comprises multiple stages of data processing sub-systems with one of the primary sub-systems is a fully connected neural network...
SourceID crossref
SourceType Index Database
StartPage 3863
Title Field-programmable gate array implementation of efficient deep neural network architecture
Volume 14
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources
  customDbUrl:
  eissn: 2722-2578
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssib044740765
  issn: 2088-8708
  databaseCode: M~E
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: https://road.issn.org
  providerName: ISSN International Centre
– providerCode: PRVPQU
  databaseName: Advanced Technologies & Aerospace Database
  customDbUrl:
  eissn: 2722-2578
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000866295
  issn: 2088-8708
  databaseCode: P5Z
  dateStart: 20110901
  isFulltext: true
  titleUrlDefault: https://search.proquest.com/hightechjournals
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Computer Science Database
  customDbUrl:
  eissn: 2722-2578
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000866295
  issn: 2088-8708
  databaseCode: K7-
  dateStart: 20110901
  isFulltext: true
  titleUrlDefault: http://search.proquest.com/compscijour
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: East & South Asia Database
  customDbUrl:
  eissn: 2722-2578
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000866295
  issn: 2088-8708
  databaseCode: BVBZV
  dateStart: 20110901
  isFulltext: true
  titleUrlDefault: https://search.proquest.com/eastsouthasia
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Engineering Database (subscription)
  customDbUrl:
  eissn: 2722-2578
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000866295
  issn: 2088-8708
  databaseCode: M7S
  dateStart: 20110901
  isFulltext: true
  titleUrlDefault: http://search.proquest.com
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: ProQuest Central
  customDbUrl:
  eissn: 2722-2578
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000866295
  issn: 2088-8708
  databaseCode: BENPR
  dateStart: 20110901
  isFulltext: true
  titleUrlDefault: https://www.proquest.com/central
  providerName: ProQuest
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Nj9MwELXKwoE9ID7F8iUfuEVZ6thx7CNCu0JCu9rDglZcItuZSIWSRiVUy4Xfx89ibCdpygrxIXGJoqgdpX1P42f7zZiQ5wKkhJzZ1DImUwG1TVUl6xQKVAsCGW2sCYdNFKen6uJCn81m34damM2yaBp1eanb_wo1PkOwfensX8A9BsUHeI-g4xVhx-sfAX_sPWlp77v6FCqj_FJZYtZr89VXRfaG8UEqQmgi4S0BFUCb-AaXCFsT7eHJdKNhKmR3VxIn_SfiuTpjDwLXnxqRwLbxoVe1J2ZpENZYLtTf7vq-EfkqjgBnq64Lzm9M0Et8x-3elPfeLkxYGXqz6kzyDpqPsdhuXMrIxGik6zNehikP0_M8JmSIzwqcLfu8spOyxYSaYpJ_uYrp8urAkGs_Miw-gIPDDRMLcdi2_tO-uXC-HQwHA8BPY-ToXAxzJoxVhkhliFTGSKWPdI1cz4pce3PhybejIbMJUeDEud9PDhpBSZmFI4HG3zzYzDD4i1-85kQ7TUTQ-W1yq5-90JeRdXfIDJq7ZH_S0_IeeX-Vf9Tzjwb-0V3-0VVNR_5Rzz8a-Ud7_tEp_-6Tt8dH569ep_0BHqljUuep47rSRQXOZBVY65zKASUmzyqbWWCiQv3JhGUahdrcybnTIGsojHGosrlT_AHZa1YNPCTUCq0kkpGr2gjOlUVZL9RcOODcytoeEDb8NWUb-7SUv8Pq0T985zG5uWXuE7LXrb_AU3LDbbrF5_WzAPoPyMCWFQ
linkProvider ISSN International Centre
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=Field-programmable+gate+array+implementation+of+efficient+deep+neural+network+architecture&rft.jtitle=International+journal+of+electrical+and+computer+engineering+%28Malacca%2C+Malacca%29&rft.au=Kumar+Reddy%2C+Pottipati+Dileep&rft.au=Ramanaiah%2C+Kota+Venkata&rft.date=2024-08-01&rft.issn=2088-8708&rft.eissn=2722-2578&rft.volume=14&rft.issue=4&rft.spage=3863&rft_id=info:doi/10.11591%2Fijece.v14i4.pp3863-3875&rft.externalDBID=n%2Fa&rft.externalDocID=10_11591_ijece_v14i4_pp3863_3875
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2088-8708&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2088-8708&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2088-8708&client=summon