{"id":"https://openalex.org/W2946355854","doi":"https://doi.org/10.1109/tvlsi.2019.2913958","title":"High-Performance FPGA-Based CNN Accelerator With Block-Floating-Point Arithmetic","display_name":"High-Performance FPGA-Based CNN Accelerator With Block-Floating-Point Arithmetic","publication_year":2019,"publication_date":"2019-05-16","ids":{"openalex":"https://openalex.org/W2946355854","doi":"https://doi.org/10.1109/tvlsi.2019.2913958","mag":"2946355854"},"language":"en","primary_location":{"id":"doi:10.1109/tvlsi.2019.2913958","doi":"https://doi.org/10.1109/tvlsi.2019.2913958","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tvlsi.2019.2913958","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["ccf-b","cwts-core","dongbi-c","jufo-2","norway-1","poland-100","russia-white-list-1"],"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","raw_type":"journal-article"},"volume":null,"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101872207","display_name":"Xiaocong Lian","orcid":"https://orcid.org/0000-0003-3917-5273","observed_orcids":["https://orcid.org/0000-0003-3917-5273"]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaocong Lian","raw_affiliation_strings":["Department of Automation, Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-3917-5273","affiliations":[{"raw_affiliation_string":"Department of Automation, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111822524","display_name":"Zhenyu Liu","orcid":"https://orcid.org/0000-0001-7559-8519","observed_orcids":["https://orcid.org/0000-0001-7559-8519","https://orcid.org/0000-0001-5077-5777"]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhenyu Liu","raw_affiliation_strings":["Tsinghua National Laboratory for Information Science and Technology, Research Institute of Information Technology, Tsinghua University, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua National Laboratory for Information Science and Technology, Research Institute of Information Technology, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001745189","display_name":"Zhourui Song","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhourui Song","raw_affiliation_strings":["School of Cyberspace Security, Beijing University of Posts and Telecommunications (BUPT), Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Cyberspace Security, Beijing University of Posts and Telecommunications (BUPT), Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089498218","display_name":"Jiwu Dai","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I17145004","display_name":"Northwestern Polytechnical University","ror":"https://ror.org/01y0j0j86","country_code":"CN","type":"education","lineage":["https://openalex.org/I17145004"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiwu Dai","raw_affiliation_strings":["School of Electronics and Information, Northwestern Polytechnical University, Xi\u2019an, China","School of Electronics and Information, Northwestern Polytechnical University, Xi'an, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Electronics and Information, Northwestern Polytechnical University, Xi\u2019an, China","institution_ids":["https://openalex.org/I17145004"]},{"raw_affiliation_string":"School of Electronics and Information, Northwestern Polytechnical University, Xi'an, China","institution_ids":["https://openalex.org/I17145004"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000288828","display_name":"Wei Zhou","orcid":"https://orcid.org/0000-0001-9715-6957","observed_orcids":["https://orcid.org/0000-0001-9715-6957"]},"institutions":[{"id":"https://openalex.org/I17145004","display_name":"Northwestern Polytechnical University","ror":"https://ror.org/01y0j0j86","country_code":"CN","type":"education","lineage":["https://openalex.org/I17145004"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Zhou","raw_affiliation_strings":["School of Electronics and Information, Northwestern Polytechnical University, Xi\u2019an, China","School of Electronics and Information, Northwestern Polytechnical University, Xi'an, China"],"raw_orcid":"https://orcid.org/0000-0001-9715-6957","affiliations":[{"raw_affiliation_string":"School of Electronics and Information, Northwestern Polytechnical University, Xi\u2019an, China","institution_ids":["https://openalex.org/I17145004"]},{"raw_affiliation_string":"School of Electronics and Information, Northwestern Polytechnical University, Xi'an, China","institution_ids":["https://openalex.org/I17145004"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5024401174","display_name":"Xiangyang Ji","orcid":"https://orcid.org/0000-0002-7333-9975","observed_orcids":["https://orcid.org/0000-0002-7333-9975"]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiangyang Ji","raw_affiliation_strings":["Department of Automation, Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-7333-9975","affiliations":[{"raw_affiliation_string":"Department of Automation, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":{"value":2045,"currency":"USD","value_usd":2045},"apc_paid":null,"fwci":10.9725,"has_fulltext":false,"cited_by_count":161,"citation_normalized_percentile":{"value":0.99039461,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":"27","issue":"8","first_page":"1874","last_page":"1885"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.6807000041007996,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.6807000041007996,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10904","display_name":"Embedded Systems Design Techniques","score":0.288100004196167,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11697","display_name":"Numerical Methods and Algorithms","score":0.007400000002235174,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/block-floating-point-arithmetic","display_name":"block floating-point arithmetic","score":0.9599999785423279},{"id":"https://openalex.org/keywords/8-bit-quantization","display_name":"8-bit quantization","score":0.8820000290870667},{"id":"https://openalex.org/keywords/low-precision-arithmetic","display_name":"low-precision arithmetic","score":0.5730000138282776},{"id":"https://openalex.org/keywords/reconfigurable-hardware-accelerators","display_name":"reconfigurable hardware accelerators","score":0.5789999961853027},{"id":"https://openalex.org/keywords/ddr3","display_name":"DDR3 memory","score":0.671999990940094},{"id":"https://openalex.org/keywords/on-chip-buffer","display_name":"on-chip buffer","score":0.7929999828338623},{"id":"https://openalex.org/keywords/energy-efficiency","display_name":"energy efficiency","score":0.871999979019165},{"id":"https://openalex.org/keywords/deep-neural-network-inference","display_name":"deep neural network inference","score":0.7210000157356262}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8040236234664917},{"id":"https://openalex.org/C84211073","wikidata":"https://www.wikidata.org/wiki/Q117879","display_name":"Floating point","level":2,"score":0.5856360793113708},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.5586517453193665},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.5150132179260254},{"id":"https://openalex.org/C136625980","wikidata":"https://www.wikidata.org/wiki/Q663208","display_name":"Rounding","level":2,"score":0.5147154927253723},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.5117718577384949},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.45581597089767456},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.44163069128990173},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4414834678173065},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.434086412191391},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.32578611373901367},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2734004259109497},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.1987011432647705},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.08445987105369568},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tvlsi.2019.2913958","doi":"https://doi.org/10.1109/tvlsi.2019.2913958","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tvlsi.2019.2913958","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["ccf-b","cwts-core","dongbi-c","jufo-2","norway-1","poland-100","russia-white-list-1"],"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"sustainable_development_goals_aurora":[{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.8999999761581421}],"study_designs":[],"awards":[{"id":"https://openalex.org/G1295000622","display_name":null,"funder_award_id":"61836012","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G1449094659","display_name":null,"funder_award_id":"61620106005","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G4779183083","display_name":null,"funder_award_id":"61827804","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G6061191929","display_name":null,"funder_award_id":"61325003","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":34,"referenced_works":["https://openalex.org/W1686810756","https://openalex.org/W1841592590","https://openalex.org/W1903029394","https://openalex.org/W1999085092","https://openalex.org/W2094756095","https://openalex.org/W2097117768","https://openalex.org/W2112796928","https://openalex.org/W2117539524","https://openalex.org/W2119144962","https://openalex.org/W2123736966","https://openalex.org/W2155893237","https://openalex.org/W2163605009","https://openalex.org/W2194775991","https://openalex.org/W2276486856","https://openalex.org/W2289252105","https://openalex.org/W2525740295","https://openalex.org/W2565851976","https://openalex.org/W2585540825","https://openalex.org/W2605487586","https://openalex.org/W2616014673","https://openalex.org/W2623629680","https://openalex.org/W2627042741","https://openalex.org/W2750822049","https://openalex.org/W2759398875","https://openalex.org/W2781078435","https://openalex.org/W2795915628","https://openalex.org/W2899915146","https://openalex.org/W2962786581","https://openalex.org/W2963374099","https://openalex.org/W2963679555","https://openalex.org/W3118608800","https://openalex.org/W4240168186","https://openalex.org/W4242577057","https://openalex.org/W6787972765"],"related_works":["https://openalex.org/W2535419677","https://openalex.org/W2134876043","https://openalex.org/W2476214559","https://openalex.org/W3039927303","https://openalex.org/W4388911419","https://openalex.org/W4322008031","https://openalex.org/W2118569990","https://openalex.org/W2389154333","https://openalex.org/W1989068301","https://openalex.org/W4232872402"],"abstract_inverted_index":{"Convolutional":[0],"neural":[1,53],"networks":[2,54],"(CNNs)":[3],"are":[4,64],"widely":[5],"used":[6],"and":[7,15,36,61,68,80,86,102,109,151,173],"have":[8],"achieved":[9],"great":[10],"success":[11],"in":[12,25,45,55,66,72,121],"computer":[13],"vision":[14],"speech":[16],"processing":[17],"applications.":[18],"However,":[19],"deploying":[20],"the":[21,26,32,73,90,107,127,160],"large-scale":[22],"CNN":[23,116],"model":[24,62,117],"embedded":[27],"system":[28],"is":[29,43,157],"subject":[30],"to":[31,89],"constraints":[33],"of":[34,51,129,133,170],"computation":[35],"memory.":[37],"An":[38],"optimized":[39,100,153],"block-floating-point":[40],"(BFP)":[41],"arithmetic":[42,98],"adopted":[44],"our":[46,122],"accelerator":[47,123,141,166],"for":[48],"efficient":[49],"inference":[50],"deep":[52],"this":[56],"paper.":[57],"The":[58,94,138],"feature":[59],"maps":[60],"parameters":[63],"represented":[65],"16-bit":[67],"8-bit":[69,96],"formats,":[70],"respectively,":[71],"off-chip":[74,81,147],"memory,":[75],"which":[76],"can":[77,118],"reduce":[78],"memory":[79,149],"bandwidth":[82],"requirements":[83],"by":[84,112],"50%":[85],"75%":[87],"compared":[88],"32-bit":[91],"FP":[92],"counterpart.":[93],"proposed":[95,139],"BFP":[97],"with":[99,142],"rounding":[101],"shifting-operation-based":[103],"quantization":[104],"schemes":[105],"improves":[106],"energy":[108],"hardware":[110],"efficiency":[111],"three":[113,143],"times.":[114],"One":[115],"be":[119],"deployed":[120],"without":[124],"retraining":[125],"at":[126],"cost":[128],"an":[130,152],"accuracy":[131],"loss":[132],"not":[134],"more":[135],"than":[136],"0.12%.":[137],"reconfigurable":[140],"parallelism":[144],"dimensions,":[145],"ping-pong":[146],"DDR3":[148],"access,":[150],"on-chip":[154],"buffer":[155],"group":[156],"implemented":[158],"on":[159],"Xilinx":[161],"VC709":[162],"evaluation":[163],"board.":[164],"Our":[165],"achieves":[167],"a":[168,177],"performance":[169],"760.83":[171],"GOP/s":[172],"82.88":[174],"GOP/s/W":[175],"under":[176],"200-MHz":[178],"working":[179],"frequency,":[180],"significantly":[181],"outperforming":[182],"previous":[183],"accelerators.":[184]},"counts_by_year":[{"year":2026,"cited_by_count":7},{"year":2025,"cited_by_count":20},{"year":2024,"cited_by_count":16},{"year":2023,"cited_by_count":19},{"year":2022,"cited_by_count":35},{"year":2021,"cited_by_count":30},{"year":2020,"cited_by_count":24},{"year":2019,"cited_by_count":10}],"updated_date":"2026-10-16T09:59:22.235173","created_date":"2025-10-10T00:00:00"}