An Asynchronous IEEE Floating-Point Arithmetic Unit
An asynchronous floating-point arithmetic unit is designed and tested at the transistor level usingCadence software. It uses CMOS (complementary metal oxide semiconductor) and DCVS (differentialcascode voltage switch) logic in a 0.35 µm process using a 3.3 V supply voltage, with dual-rail data ands...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
University of the Philippines
2007-12-01
|
| Series: | Science Diliman |
| Subjects: | |
| Online Access: | http://journals.upd.edu.ph/index.php/sciencediliman/article/view/711 |
| _version_ | 1818288690455117824 |
|---|---|
| author | Joel R. Noche Jose C. Araneta |
| author_facet | Joel R. Noche Jose C. Araneta |
| author_sort | Joel R. Noche |
| collection | DOAJ |
| description | An asynchronous floating-point arithmetic unit is designed and tested at the transistor level usingCadence software. It uses CMOS (complementary metal oxide semiconductor) and DCVS (differentialcascode voltage switch) logic in a 0.35 µm process using a 3.3 V supply voltage, with dual-rail data andsingle-rail control signals using four-phase handshaking.Using 17,085 transistors, the unit handles single-precision (32-bit) addition/subtraction, multiplication,division, and remainder using the IEEE 754-1985 Standard for Binary Floating-Point Arithmetic, withrounding and other operations to be handled by separate hardware or software. Division and remainderare done using a restoring subtractive algorithm; multiplication uses an additive algorithm. Exceptionsare noted by flags (and not trap handlers) and the output is in single-precision.Previous work on asynchronous floating-point arithmetic units have mostly focused on single operationssuch as division. This is the first work to the authors' knowledge that can perform floating-point addition,multiplication, division, and remainder using a common datapath. |
| first_indexed | 2024-12-13T02:00:23Z |
| format | Article |
| id | doaj.art-986a4d47f9a647c5bfaee4378b713da6 |
| institution | Directory Open Access Journal |
| issn | 0115-7809 2012-0818 |
| language | English |
| last_indexed | 2024-12-13T02:00:23Z |
| publishDate | 2007-12-01 |
| publisher | University of the Philippines |
| record_format | Article |
| series | Science Diliman |
| spelling | doaj.art-986a4d47f9a647c5bfaee4378b713da62022-12-22T00:03:17ZengUniversity of the PhilippinesScience Diliman0115-78092012-08182007-12-011921222An Asynchronous IEEE Floating-Point Arithmetic UnitJoel R. NocheJose C. AranetaAn asynchronous floating-point arithmetic unit is designed and tested at the transistor level usingCadence software. It uses CMOS (complementary metal oxide semiconductor) and DCVS (differentialcascode voltage switch) logic in a 0.35 µm process using a 3.3 V supply voltage, with dual-rail data andsingle-rail control signals using four-phase handshaking.Using 17,085 transistors, the unit handles single-precision (32-bit) addition/subtraction, multiplication,division, and remainder using the IEEE 754-1985 Standard for Binary Floating-Point Arithmetic, withrounding and other operations to be handled by separate hardware or software. Division and remainderare done using a restoring subtractive algorithm; multiplication uses an additive algorithm. Exceptionsare noted by flags (and not trap handlers) and the output is in single-precision.Previous work on asynchronous floating-point arithmetic units have mostly focused on single operationssuch as division. This is the first work to the authors' knowledge that can perform floating-point addition,multiplication, division, and remainder using a common datapath.http://journals.upd.edu.ph/index.php/sciencediliman/article/view/711Asynchronous logic circuitsfloating point arithmeticcalculation times |
| spellingShingle | Joel R. Noche Jose C. Araneta An Asynchronous IEEE Floating-Point Arithmetic Unit Science Diliman Asynchronous logic circuits floating point arithmetic calculation times |
| title | An Asynchronous IEEE Floating-Point Arithmetic Unit |
| title_full | An Asynchronous IEEE Floating-Point Arithmetic Unit |
| title_fullStr | An Asynchronous IEEE Floating-Point Arithmetic Unit |
| title_full_unstemmed | An Asynchronous IEEE Floating-Point Arithmetic Unit |
| title_short | An Asynchronous IEEE Floating-Point Arithmetic Unit |
| title_sort | asynchronous ieee floating point arithmetic unit |
| topic | Asynchronous logic circuits floating point arithmetic calculation times |
| url | http://journals.upd.edu.ph/index.php/sciencediliman/article/view/711 |
| work_keys_str_mv | AT joelrnoche anasynchronousieeefloatingpointarithmeticunit AT josecaraneta anasynchronousieeefloatingpointarithmeticunit AT joelrnoche asynchronousieeefloatingpointarithmeticunit AT josecaraneta asynchronousieeefloatingpointarithmeticunit |