{"id":"https://openalex.org/W4283801046","doi":"https://doi.org/10.1145/3546182","title":"Efficient Design of Low Bitwidth Convolutional Neural Networks on FPGA with Optimized Dot Product Units","display_name":"Efficient Design of Low Bitwidth Convolutional Neural Networks on FPGA with Optimized Dot Product Units","publication_year":2022,"publication_date":"2022-07-04","ids":{"openalex":"https://openalex.org/W4283801046","doi":"https://doi.org/10.1145/3546182"},"language":"en","primary_location":{"id":"doi:10.1145/3546182","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3546182","pdf_url":null,"source":{"id":"https://openalex.org/S112809824","display_name":"ACM Transactions on Reconfigurable Technology and Systems","issn_l":"1936-7406","issn":["1936-7406","1936-7414"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Reconfigurable Technology and Systems","raw_type":"journal-article"},"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/A5003178960","display_name":"M\u00e1rio V\u00e9stias","orcid":"https://orcid.org/0000-0001-8556-4507"},"institutions":[{"id":"https://openalex.org/I121345201","display_name":"Instituto de Engenharia de Sistemas e Computadores Investiga\u00e7\u00e3o e Desenvolvimento","ror":"https://ror.org/04mqy3p58","country_code":"PT","type":"nonprofit","lineage":["https://openalex.org/I121345201","https://openalex.org/I4210125590"]},{"id":"https://openalex.org/I203847022","display_name":"Instituto Polit\u00e9cnico de Lisboa","ror":"https://ror.org/04ea70f07","country_code":"PT","type":"education","lineage":["https://openalex.org/I203847022"]}],"countries":["PT"],"is_corresponding":false,"raw_author_name":"M\u00e1rio V\u00e9stias","raw_affiliation_strings":["INESC-ID, ISEL, Instituto Polit\u00e9cnico de Lisboa, Lisbon, Portugal"],"raw_orcid":"https://orcid.org/0000-0001-8556-4507","affiliations":[{"raw_affiliation_string":"INESC-ID, ISEL, Instituto Polit\u00e9cnico de Lisboa, Lisbon, Portugal","institution_ids":["https://openalex.org/I121345201","https://openalex.org/I203847022"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079441335","display_name":"Rui Policarpo Duarte","orcid":"https://orcid.org/0000-0002-7060-4745"},"institutions":[{"id":"https://openalex.org/I121345201","display_name":"Instituto de Engenharia de Sistemas e Computadores Investiga\u00e7\u00e3o e Desenvolvimento","ror":"https://ror.org/04mqy3p58","country_code":"PT","type":"nonprofit","lineage":["https://openalex.org/I121345201","https://openalex.org/I4210125590"]}],"countries":["PT"],"is_corresponding":false,"raw_author_name":"Rui P. Duarte","raw_affiliation_strings":["INESC-ID, Lisbon, Portugal"],"raw_orcid":"https://orcid.org/0000-0002-7060-4745","affiliations":[{"raw_affiliation_string":"INESC-ID, Lisbon, Portugal","institution_ids":["https://openalex.org/I121345201"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043958334","display_name":"Jos\u00e9 T. de Sousa","orcid":"https://orcid.org/0000-0001-7525-7546"},"institutions":[{"id":"https://openalex.org/I121345201","display_name":"Instituto de Engenharia de Sistemas e Computadores Investiga\u00e7\u00e3o e Desenvolvimento","ror":"https://ror.org/04mqy3p58","country_code":"PT","type":"nonprofit","lineage":["https://openalex.org/I121345201","https://openalex.org/I4210125590"]},{"id":"https://openalex.org/I141596103","display_name":"University of Lisbon","ror":"https://ror.org/01c27hj86","country_code":"PT","type":"education","lineage":["https://openalex.org/I141596103"]}],"countries":["PT"],"is_corresponding":false,"raw_author_name":"Jos\u00e9 T. de Sousa","raw_affiliation_strings":["INESC-ID, Instituto Superior T\u00e9cnico, Universidade de Lisboa, Lisbon, Portugal"],"raw_orcid":"https://orcid.org/0000-0001-7525-7546","affiliations":[{"raw_affiliation_string":"INESC-ID, Instituto Superior T\u00e9cnico, Universidade de Lisboa, Lisbon, Portugal","institution_ids":["https://openalex.org/I121345201","https://openalex.org/I141596103"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5062555847","display_name":"Hor\u00e1cio C. Neto","orcid":"https://orcid.org/0000-0002-3621-8322"},"institutions":[{"id":"https://openalex.org/I121345201","display_name":"Instituto de Engenharia de Sistemas e Computadores Investiga\u00e7\u00e3o e Desenvolvimento","ror":"https://ror.org/04mqy3p58","country_code":"PT","type":"nonprofit","lineage":["https://openalex.org/I121345201","https://openalex.org/I4210125590"]},{"id":"https://openalex.org/I141596103","display_name":"University of Lisbon","ror":"https://ror.org/01c27hj86","country_code":"PT","type":"education","lineage":["https://openalex.org/I141596103"]}],"countries":["PT"],"is_corresponding":false,"raw_author_name":"Hor\u00e1cio Neto","raw_affiliation_strings":["INESC-ID, Instituto Superior T\u00e9cnico, Universidade de Lisboa, Lisbon, Portugal"],"raw_orcid":"https://orcid.org/0000-0002-3621-8322","affiliations":[{"raw_affiliation_string":"INESC-ID, Instituto Superior T\u00e9cnico, Universidade de Lisboa, Lisbon, Portugal","institution_ids":["https://openalex.org/I121345201","https://openalex.org/I141596103"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.5873,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.65470059,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":97},"biblio":{"volume":"16","issue":"1","first_page":"1","last_page":"36"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998999834060669,"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":0.9998999834060669,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.996999979019165,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.9965000152587891,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8426734209060669},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.7109282612800598},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.6714528799057007},{"id":"https://openalex.org/keywords/adder","display_name":"Adder","score":0.5599427223205566},{"id":"https://openalex.org/keywords/dot-product","display_name":"Dot product","score":0.5245409607887268},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.46447479724884033},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4527745842933655},{"id":"https://openalex.org/keywords/digital-signal-processing","display_name":"Digital signal processing","score":0.44649261236190796},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.4446036219596863},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.43296411633491516},{"id":"https://openalex.org/keywords/multiplier","display_name":"Multiplier (economics)","score":0.4111191928386688},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.39694949984550476},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.39018580317497253},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.32481908798217773},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3201479911804199},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.13168662786483765},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.1282845437526703},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.08943486213684082}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8426734209060669},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.7109282612800598},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.6714528799057007},{"id":"https://openalex.org/C164620267","wikidata":"https://www.wikidata.org/wiki/Q376953","display_name":"Adder","level":3,"score":0.5599427223205566},{"id":"https://openalex.org/C32900221","wikidata":"https://www.wikidata.org/wiki/Q181365","display_name":"Dot product","level":2,"score":0.5245409607887268},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.46447479724884033},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4527745842933655},{"id":"https://openalex.org/C84462506","wikidata":"https://www.wikidata.org/wiki/Q173142","display_name":"Digital signal processing","level":2,"score":0.44649261236190796},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.4446036219596863},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.43296411633491516},{"id":"https://openalex.org/C124584101","wikidata":"https://www.wikidata.org/wiki/Q1053266","display_name":"Multiplier (economics)","level":2,"score":0.4111191928386688},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.39694949984550476},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.39018580317497253},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.32481908798217773},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3201479911804199},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.13168662786483765},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.1282845437526703},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.08943486213684082},{"id":"https://openalex.org/C139719470","wikidata":"https://www.wikidata.org/wiki/Q39680","display_name":"Macroeconomics","level":1,"score":0.0},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3546182","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3546182","pdf_url":null,"source":{"id":"https://openalex.org/S112809824","display_name":"ACM Transactions on Reconfigurable Technology and Systems","issn_l":"1936-7406","issn":["1936-7406","1936-7414"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Reconfigurable Technology and Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.44999998807907104,"display_name":"Industry, innovation and infrastructure","id":"https://metadata.un.org/sdg/9"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":46,"referenced_works":["https://openalex.org/W2117539524","https://openalex.org/W2183341477","https://openalex.org/W2194775991","https://openalex.org/W2276486856","https://openalex.org/W2466675884","https://openalex.org/W2521740029","https://openalex.org/W2565125333","https://openalex.org/W2585560244","https://openalex.org/W2606722458","https://openalex.org/W2699539367","https://openalex.org/W2737762472","https://openalex.org/W2742152118","https://openalex.org/W2752782242","https://openalex.org/W2761757071","https://openalex.org/W2762374354","https://openalex.org/W2799843241","https://openalex.org/W2884367402","https://openalex.org/W2887782013","https://openalex.org/W2892053304","https://openalex.org/W2901214301","https://openalex.org/W2901848761","https://openalex.org/W2904282303","https://openalex.org/W2904741511","https://openalex.org/W2905006747","https://openalex.org/W2907463061","https://openalex.org/W2917710201","https://openalex.org/W2935524202","https://openalex.org/W2946355854","https://openalex.org/W2947350915","https://openalex.org/W2985030833","https://openalex.org/W2997106510","https://openalex.org/W3008408165","https://openalex.org/W3028015399","https://openalex.org/W3034939442","https://openalex.org/W3075593146","https://openalex.org/W3101086546","https://openalex.org/W3105385999","https://openalex.org/W3116460607","https://openalex.org/W3137147200","https://openalex.org/W3157721836","https://openalex.org/W3158020960","https://openalex.org/W3212505503","https://openalex.org/W4205928000","https://openalex.org/W4211085526","https://openalex.org/W4287254789","https://openalex.org/W6910613894"],"related_works":["https://openalex.org/W4390550886","https://openalex.org/W3217463396","https://openalex.org/W2516396101","https://openalex.org/W2790557758","https://openalex.org/W3204929712","https://openalex.org/W4295102875","https://openalex.org/W2300671402","https://openalex.org/W2015457513","https://openalex.org/W4312888585","https://openalex.org/W2533938775"],"abstract_inverted_index":{"Designing":[0],"hardware":[1,57],"accelerators":[2],"to":[3,45,76,151,160,199],"run":[4],"the":[5,27,33,42,51,56,62,68,115,152,158,162,173],"inference":[6,63,75],"of":[7,26,32,155,157,184],"convolutional":[8,96],"neural":[9,28,97],"networks":[10,98],"(CNN)":[11],"is":[12,37,145],"under":[13],"intensive":[14],"research.":[15],"Several":[16],"different":[17],"architectures":[18],"have":[19],"been":[20,72],"proposed":[21,174],"along":[22],"with":[23,136,187],"hardware-oriented":[24],"optimizations":[25,36],"network":[29],"models.":[30],"One":[31],"most":[34],"used":[35],"quantization":[38],"since":[39],"it":[40],"reduces":[41],"memory":[43,52],"requirements":[44,54],"store":[46],"weights":[47],"and":[48,55,67,94,137,147,170],"layer":[49],"maps,":[50],"bandwidth":[53],"complexity.":[58],"As":[59],"a":[60,106,120,181,188],"consequence,":[61],"throughput":[64],"has":[65,71],"improved":[66],"computing":[69],"cost":[70],"reduced,":[73],"allowing":[74],"be":[77,149],"executed":[78],"on":[79,99],"embedded":[80,139],"devices.":[81],"In":[82,132],"this":[83],"work,":[84],"we":[85],"propose":[86],"highly":[87],"efficient":[88],"dot-product":[89,103,117],"arithmetic":[90],"units":[91],"for":[92,193],"ternary":[93,116],"non-ternary":[95,102],"FPGA.":[100,206],"The":[101,143,178],"unit":[104,118,123],"uses":[105,119],"fused":[107],"multiply-add":[108],"that":[109],"avoids":[110],"expensive":[111],"adder":[112,129],"trees,":[113],"while":[114],"dual":[121],"product":[122,176],"followed":[124],"by":[125],"an":[126],"optimized":[127],"conditional":[128],"tree":[130],"structure.":[131],"both":[133],"cases,":[134],"designs":[135],"without":[138],"DSP":[140],"are":[141],"considered.":[142],"solution":[144],"configurable":[146],"can":[148],"adapted":[150],"available":[153],"number":[154],"resources":[156],"FPGA":[159],"achieve":[161],"best":[163],"efficiency.":[164],"A":[165],"CNN":[166],"architecture":[167],"was":[168],"developed":[169],"characterized":[171],"using":[172],"dot":[175],"units.":[177],"results":[179],"show":[180],"performance":[182],"improvement":[183],"1.8":[185],"\u00d7":[186],"2\u00d7":[189],"more":[190],"area":[191],"efficiency":[192],"low":[194],"bit-width":[195],"quantizations":[196],"when":[197],"compared":[198],"previous":[200],"works":[201],"running":[202],"large":[203],"CNNs":[204],"in":[205]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":3},{"year":2022,"cited_by_count":1}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}