{"id":"https://openalex.org/W4400229846","doi":"https://doi.org/10.1109/iscas58744.2024.10558440","title":"High Accuracy and Low Latency Mixed Precision Neural Network Acceleration for TinyML Applications on Resource-Constrained FPGAs","display_name":"High Accuracy and Low Latency Mixed Precision Neural Network Acceleration for TinyML Applications on Resource-Constrained FPGAs","publication_year":2024,"publication_date":"2024-05-19","ids":{"openalex":"https://openalex.org/W4400229846","doi":"https://doi.org/10.1109/iscas58744.2024.10558440"},"language":"en","primary_location":{"id":"doi:10.1109/iscas58744.2024.10558440","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iscas58744.2024.10558440","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE International Symposium on Circuits and Systems (ISCAS)","raw_type":"proceedings-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/A5113274220","display_name":"Wei Soon Ng","orcid":"https://orcid.org/0000-0003-0476-350X"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Wei Soon Ng","raw_affiliation_strings":["Nanyang Technological University,School of Electrical and Electronic Engineering,Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanyang Technological University,School of Electrical and Electronic Engineering,Singapore","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042682339","display_name":"Wang Ling Goh","orcid":"https://orcid.org/0000-0001-7466-8941"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Wang Ling Goh","raw_affiliation_strings":["Nanyang Technological University,School of Electrical and Electronic Engineering,Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanyang Technological University,School of Electrical and Electronic Engineering,Singapore","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5075642035","display_name":"Yuan Gao","orcid":"https://orcid.org/0000-0002-0832-2982"},"institutions":[{"id":"https://openalex.org/I115228651","display_name":"Agency for Science, Technology and Research","ror":"https://ror.org/036wvzt09","country_code":"SG","type":"government","lineage":["https://openalex.org/I115228651"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Yuan Gao","raw_affiliation_strings":["Institue of Microelectronics, Agency for Science, Technology and Research (A*STAR),Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institue of Microelectronics, Agency for Science, Technology and Research (A*STAR),Singapore","institution_ids":["https://openalex.org/I115228651"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.8328,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.87226217,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9803000092506409,"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"}},"topics":[{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9803000092506409,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.95660001039505,"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/T13579","display_name":"Image and Video Stabilization","score":0.9419000148773193,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.7890247106552124},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7449720501899719},{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.6770057678222656},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.6510931849479675},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5919554233551025},{"id":"https://openalex.org/keywords/low-latency","display_name":"Low latency (capital markets)","score":0.4651298224925995},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.4307568669319153},{"id":"https://openalex.org/keywords/real-time-computing","display_name":"Real-time computing","score":0.34256404638290405},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.20459088683128357},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.14688432216644287},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.09662851691246033}],"concepts":[{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.7890247106552124},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7449720501899719},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.6770057678222656},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.6510931849479675},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5919554233551025},{"id":"https://openalex.org/C46637626","wikidata":"https://www.wikidata.org/wiki/Q6693015","display_name":"Low latency (capital markets)","level":2,"score":0.4651298224925995},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.4307568669319153},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.34256404638290405},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.20459088683128357},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.14688432216644287},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.09662851691246033},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C74650414","wikidata":"https://www.wikidata.org/wiki/Q11397","display_name":"Classical mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iscas58744.2024.10558440","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iscas58744.2024.10558440","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE International Symposium on Circuits and Systems (ISCAS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":20,"referenced_works":["https://openalex.org/W2067523571","https://openalex.org/W2810704618","https://openalex.org/W2963122961","https://openalex.org/W2963367920","https://openalex.org/W2981341989","https://openalex.org/W3003453708","https://openalex.org/W3082058622","https://openalex.org/W4211085526","https://openalex.org/W4211090839","https://openalex.org/W4225136883","https://openalex.org/W4291963193","https://openalex.org/W4295127407","https://openalex.org/W4319870530","https://openalex.org/W4360831992","https://openalex.org/W4367016390","https://openalex.org/W6729126992","https://openalex.org/W6791603202","https://openalex.org/W6797184042","https://openalex.org/W6838877803","https://openalex.org/W6842065451"],"related_works":["https://openalex.org/W2096844293","https://openalex.org/W2363944576","https://openalex.org/W2351041855","https://openalex.org/W2570254841","https://openalex.org/W3205411230","https://openalex.org/W4286899009","https://openalex.org/W9168048","https://openalex.org/W4300849822","https://openalex.org/W4376480820","https://openalex.org/W3155891479"],"abstract_inverted_index":{"Recent":[0],"advancements":[1],"in":[2,123],"mixed":[3,27,63,106],"precision":[4,28,64,88],"quantization":[5,74],"algorithms":[6],"hold":[7],"significant":[8],"promise":[9],"for":[10,109],"Tiny":[11,159],"Machine":[12],"Learning":[13],"(TinyML)":[14],"applications":[15],"that":[16,84],"face":[17],"tight":[18],"resource":[19,124],"constraints.":[20],"However,":[21],"to":[22,60,76,95,128,142],"leverage":[23],"the":[24,72,110,129,143,149],"benefits":[25],"of":[26,36,45,152],"NNs,":[29],"specialized":[30],"hardware":[31],"is":[32],"required.":[33],"The":[34],"reconfigurability":[35],"Field":[37],"Programmable":[38],"Gate":[39],"Arrays":[40],"(FPGAs)":[41],"allows":[42],"tensor":[43],"computations":[44,66],"any":[46],"precision.":[47],"In":[48],"this":[49],"work,":[50],"we":[51,70],"propose":[52],"a":[53,105,115],"streaming":[54],"architecture":[55],"with":[56,120],"layer-specific":[57],"processing":[58],"elements":[59],"support":[61,85],"interlayer":[62],"NN":[65,108],"on":[67,99,114],"FPGAs.":[68,101],"Additionally,":[69],"extend":[71],"supported":[73],"scheme":[75],"intralayer":[77,86],"level":[78],"by":[79,155],"introducing":[80],"multiply-accumulation":[81],"(MAC)":[82],"units":[83,92],"dynamic":[87],"quantization.":[89],"Bit-serial":[90],"MAC":[91],"are":[93],"chosen":[94],"accommodate":[96],"more":[97],"PEs":[98],"resource-constrained":[100],"We":[102],"successfully":[103],"implemented":[104],"quantized":[107],"keyword":[111],"spotting":[112],"task":[113],"FPGA-only":[116,131],"Digilent":[117],"A7-35T":[118],"platform":[119],"62%":[121],"reduction":[122],"consumption":[125],"when":[126],"compared":[127,141],"previous":[130],"implementations.":[132],"Our":[133],"implementation":[134,146],"also":[135],"achieves":[136],"2":[137],"times":[138],"speedup":[139],"as":[140],"state-of-the-art":[144],"ASIC":[145],"while":[147],"meeting":[148],"accuracy":[150],"requirement":[151],"90%":[153],"set":[154],"MLPerf<sup":[156],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[157],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">\u00ae</sup>":[158],"Benchmark.":[160]},"counts_by_year":[{"year":2025,"cited_by_count":6}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}