PWL-Based Architecture for the Logarithmic Computation of Floating-Point Numbers
In this brief, we propose a logarithmic converter for floating-point numbers based on the piecewise linear (PWL) approximation method. The proposed method is applicable to any customized floating-point format with a mantissa length of 16-23 bits and a maximum absolute error (MAE) larger than 10 −6 ....
Gespeichert in:
| Veröffentlicht in: | IEEE transactions on very large scale integration (VLSI) systems 2021-07, Vol.29 (7), p.1470-1474 |
|---|---|
| Hauptverfasser: | , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 1474 |
|---|---|
| container_issue | 7 |
| container_start_page | 1470 |
| container_title | IEEE transactions on very large scale integration (VLSI) systems |
| container_volume | 29 |
| creator | Lyu, Fei Mao, Zhelong Zhang, Jin Wang, Yu Luo, Yuanyong |
| description | In this brief, we propose a logarithmic converter for floating-point numbers based on the piecewise linear (PWL) approximation method. The proposed method is applicable to any customized floating-point format with a mantissa length of 16-23 bits and a maximum absolute error (MAE) larger than 10 −6 . The logarithmic function is automatically segmented into several maximal subsections by a software-based segmentation scheme with the restriction of a predefined MAE and a fractional word length for the computing units. Then, we make a tradeoff between the piecewise number and the fractional word length. Based on the results of the segmentor, our design is coded in the Verilog hardware description language. The synthesized results show that our design consumes less area, time, and power without compromising accuracy compared to existing techniques based on the COordinate Rotation Digital Computer (CORDIC) and PWL methods. |
| doi_str_mv | 10.1109/TVLSI.2021.3081572 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_RIE</sourceid><recordid>TN_cdi_ieee_primary_9444119</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>9444119</ieee_id><sourcerecordid>2546719515</sourcerecordid><originalsourceid>FETCH-LOGICAL-c295t-8bbf03ce8f7c6c698dea4f0ba7006143de13bd0c32142590270e71511e380b4f3</originalsourceid><addsrcrecordid>eNo9kMtOwzAQRSMEEqXwA7CxxDplxo8kXpaKQqUIKlFgaSWO3aZq4uI4C_6elFbMZu7inhnpRNEtwgQR5MPqM39fTChQnDDIUKT0LBqhEGkshzkfMiQszijCZXTVdVsA5FzCKFouv_L4sehMRaZeb-pgdOi9IdZ5EjaG5G5d-DpsmlqTmWv2fShC7VriLJnv3JDbdbx0dRvIa9-UxnfX0YUtdp25Oe1x9DF_Ws1e4vzteTGb5rGmUoQ4K0sLTJvMpjrRicwqU3ALZZECJMhZZZCVFWhGkVMhgaZgUhSIhmVQcsvG0f3x7t677950QW1d79vhpaKCJylKgWJo0WNLe9d13li193VT-B-FoA7m1J85dTCnTuYG6O4I1caYf0ByzhEl-wWsZWmX</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2546719515</pqid></control><display><type>article</type><title>PWL-Based Architecture for the Logarithmic Computation of Floating-Point Numbers</title><source>IEEE Electronic Library (IEL)</source><creator>Lyu, Fei ; Mao, Zhelong ; Zhang, Jin ; Wang, Yu ; Luo, Yuanyong</creator><creatorcontrib>Lyu, Fei ; Mao, Zhelong ; Zhang, Jin ; Wang, Yu ; Luo, Yuanyong</creatorcontrib><description>In this brief, we propose a logarithmic converter for floating-point numbers based on the piecewise linear (PWL) approximation method. The proposed method is applicable to any customized floating-point format with a mantissa length of 16-23 bits and a maximum absolute error (MAE) larger than 10 −6 . The logarithmic function is automatically segmented into several maximal subsections by a software-based segmentation scheme with the restriction of a predefined MAE and a fractional word length for the computing units. Then, we make a tradeoff between the piecewise number and the fractional word length. Based on the results of the segmentor, our design is coded in the Verilog hardware description language. The synthesized results show that our design consumes less area, time, and power without compromising accuracy compared to existing techniques based on the COordinate Rotation Digital Computer (CORDIC) and PWL methods.</description><identifier>ISSN: 1063-8210</identifier><identifier>EISSN: 1557-9999</identifier><identifier>DOI: 10.1109/TVLSI.2021.3081572</identifier><identifier>CODEN: IEVSE9</identifier><language>eng</language><publisher>New York: IEEE</publisher><subject>Approximation ; Approximation algorithms ; Converters ; Digital computers ; Floating point arithmetic ; Hardware ; Hardware description languages ; Logarithmic converter ; maximum absolute error (MAE) ; Measurement ; piecewise linear (PWL) approximation ; Power consumption ; Quantization (signal) ; Segmentation ; Signal processing algorithms ; Very large scale integration</subject><ispartof>IEEE transactions on very large scale integration (VLSI) systems, 2021-07, Vol.29 (7), p.1470-1474</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2021</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c295t-8bbf03ce8f7c6c698dea4f0ba7006143de13bd0c32142590270e71511e380b4f3</citedby><cites>FETCH-LOGICAL-c295t-8bbf03ce8f7c6c698dea4f0ba7006143de13bd0c32142590270e71511e380b4f3</cites><orcidid>0000-0002-4450-066X ; 0000-0003-2282-1574 ; 0000-0003-0497-8906</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/9444119$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>314,776,780,792,27903,27904,54736</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/9444119$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Lyu, Fei</creatorcontrib><creatorcontrib>Mao, Zhelong</creatorcontrib><creatorcontrib>Zhang, Jin</creatorcontrib><creatorcontrib>Wang, Yu</creatorcontrib><creatorcontrib>Luo, Yuanyong</creatorcontrib><title>PWL-Based Architecture for the Logarithmic Computation of Floating-Point Numbers</title><title>IEEE transactions on very large scale integration (VLSI) systems</title><addtitle>TVLSI</addtitle><description>In this brief, we propose a logarithmic converter for floating-point numbers based on the piecewise linear (PWL) approximation method. The proposed method is applicable to any customized floating-point format with a mantissa length of 16-23 bits and a maximum absolute error (MAE) larger than 10 −6 . The logarithmic function is automatically segmented into several maximal subsections by a software-based segmentation scheme with the restriction of a predefined MAE and a fractional word length for the computing units. Then, we make a tradeoff between the piecewise number and the fractional word length. Based on the results of the segmentor, our design is coded in the Verilog hardware description language. The synthesized results show that our design consumes less area, time, and power without compromising accuracy compared to existing techniques based on the COordinate Rotation Digital Computer (CORDIC) and PWL methods.</description><subject>Approximation</subject><subject>Approximation algorithms</subject><subject>Converters</subject><subject>Digital computers</subject><subject>Floating point arithmetic</subject><subject>Hardware</subject><subject>Hardware description languages</subject><subject>Logarithmic converter</subject><subject>maximum absolute error (MAE)</subject><subject>Measurement</subject><subject>piecewise linear (PWL) approximation</subject><subject>Power consumption</subject><subject>Quantization (signal)</subject><subject>Segmentation</subject><subject>Signal processing algorithms</subject><subject>Very large scale integration</subject><issn>1063-8210</issn><issn>1557-9999</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>RIE</sourceid><recordid>eNo9kMtOwzAQRSMEEqXwA7CxxDplxo8kXpaKQqUIKlFgaSWO3aZq4uI4C_6elFbMZu7inhnpRNEtwgQR5MPqM39fTChQnDDIUKT0LBqhEGkshzkfMiQszijCZXTVdVsA5FzCKFouv_L4sehMRaZeb-pgdOi9IdZ5EjaG5G5d-DpsmlqTmWv2fShC7VriLJnv3JDbdbx0dRvIa9-UxnfX0YUtdp25Oe1x9DF_Ws1e4vzteTGb5rGmUoQ4K0sLTJvMpjrRicwqU3ALZZECJMhZZZCVFWhGkVMhgaZgUhSIhmVQcsvG0f3x7t677950QW1d79vhpaKCJylKgWJo0WNLe9d13li193VT-B-FoA7m1J85dTCnTuYG6O4I1caYf0ByzhEl-wWsZWmX</recordid><startdate>20210701</startdate><enddate>20210701</enddate><creator>Lyu, Fei</creator><creator>Mao, Zhelong</creator><creator>Zhang, Jin</creator><creator>Wang, Yu</creator><creator>Luo, Yuanyong</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SP</scope><scope>8FD</scope><scope>L7M</scope><orcidid>https://orcid.org/0000-0002-4450-066X</orcidid><orcidid>https://orcid.org/0000-0003-2282-1574</orcidid><orcidid>https://orcid.org/0000-0003-0497-8906</orcidid></search><sort><creationdate>20210701</creationdate><title>PWL-Based Architecture for the Logarithmic Computation of Floating-Point Numbers</title><author>Lyu, Fei ; Mao, Zhelong ; Zhang, Jin ; Wang, Yu ; Luo, Yuanyong</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c295t-8bbf03ce8f7c6c698dea4f0ba7006143de13bd0c32142590270e71511e380b4f3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Approximation</topic><topic>Approximation algorithms</topic><topic>Converters</topic><topic>Digital computers</topic><topic>Floating point arithmetic</topic><topic>Hardware</topic><topic>Hardware description languages</topic><topic>Logarithmic converter</topic><topic>maximum absolute error (MAE)</topic><topic>Measurement</topic><topic>piecewise linear (PWL) approximation</topic><topic>Power consumption</topic><topic>Quantization (signal)</topic><topic>Segmentation</topic><topic>Signal processing algorithms</topic><topic>Very large scale integration</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Lyu, Fei</creatorcontrib><creatorcontrib>Mao, Zhelong</creatorcontrib><creatorcontrib>Zhang, Jin</creatorcontrib><creatorcontrib>Wang, Yu</creatorcontrib><creatorcontrib>Luo, Yuanyong</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998-Present</collection><collection>IEEE Electronic Library (IEL)</collection><collection>CrossRef</collection><collection>Electronics & Communications Abstracts</collection><collection>Technology Research Database</collection><collection>Advanced Technologies Database with Aerospace</collection><jtitle>IEEE transactions on very large scale integration (VLSI) systems</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Lyu, Fei</au><au>Mao, Zhelong</au><au>Zhang, Jin</au><au>Wang, Yu</au><au>Luo, Yuanyong</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>PWL-Based Architecture for the Logarithmic Computation of Floating-Point Numbers</atitle><jtitle>IEEE transactions on very large scale integration (VLSI) systems</jtitle><stitle>TVLSI</stitle><date>2021-07-01</date><risdate>2021</risdate><volume>29</volume><issue>7</issue><spage>1470</spage><epage>1474</epage><pages>1470-1474</pages><issn>1063-8210</issn><eissn>1557-9999</eissn><coden>IEVSE9</coden><abstract>In this brief, we propose a logarithmic converter for floating-point numbers based on the piecewise linear (PWL) approximation method. The proposed method is applicable to any customized floating-point format with a mantissa length of 16-23 bits and a maximum absolute error (MAE) larger than 10 −6 . The logarithmic function is automatically segmented into several maximal subsections by a software-based segmentation scheme with the restriction of a predefined MAE and a fractional word length for the computing units. Then, we make a tradeoff between the piecewise number and the fractional word length. Based on the results of the segmentor, our design is coded in the Verilog hardware description language. The synthesized results show that our design consumes less area, time, and power without compromising accuracy compared to existing techniques based on the COordinate Rotation Digital Computer (CORDIC) and PWL methods.</abstract><cop>New York</cop><pub>IEEE</pub><doi>10.1109/TVLSI.2021.3081572</doi><tpages>5</tpages><orcidid>https://orcid.org/0000-0002-4450-066X</orcidid><orcidid>https://orcid.org/0000-0003-2282-1574</orcidid><orcidid>https://orcid.org/0000-0003-0497-8906</orcidid></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1063-8210 |
| ispartof | IEEE transactions on very large scale integration (VLSI) systems, 2021-07, Vol.29 (7), p.1470-1474 |
| issn | 1063-8210 1557-9999 |
| language | eng |
| recordid | cdi_ieee_primary_9444119 |
| source | IEEE Electronic Library (IEL) |
| subjects | Approximation Approximation algorithms Converters Digital computers Floating point arithmetic Hardware Hardware description languages Logarithmic converter maximum absolute error (MAE) Measurement piecewise linear (PWL) approximation Power consumption Quantization (signal) Segmentation Signal processing algorithms Very large scale integration |
| title | PWL-Based Architecture for the Logarithmic Computation of Floating-Point Numbers |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-24T06%3A57%3A46IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_RIE&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=PWL-Based%20Architecture%20for%20the%20Logarithmic%20Computation%20of%20Floating-Point%20Numbers&rft.jtitle=IEEE%20transactions%20on%20very%20large%20scale%20integration%20(VLSI)%20systems&rft.au=Lyu,%20Fei&rft.date=2021-07-01&rft.volume=29&rft.issue=7&rft.spage=1470&rft.epage=1474&rft.pages=1470-1474&rft.issn=1063-8210&rft.eissn=1557-9999&rft.coden=IEVSE9&rft_id=info:doi/10.1109/TVLSI.2021.3081572&rft_dat=%3Cproquest_RIE%3E2546719515%3C/proquest_RIE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2546719515&rft_id=info:pmid/&rft_ieee_id=9444119&rfr_iscdi=true |