Unifying bit-width optimisation for fixed-point and floating-point designs
作者： A.A. Gaffar , O. Mencer , W. Luk , P.Y.K. Cheung
DOI: 10.1109/FCCM.2004.59
关键词:
摘要: This paper presents a method that offers uniform treatment for bit-width optimisation of both fixed-point and floating-point designs. Our work utilises automatic differentiation to compute the sensitivities outputs various operands in design. sensitivity analysis enables us explore compare implementation particular As result, we can automate selection optimal number representation each variable design optimize area performance. We implement our BitSize tool targeting reconfigurable architectures, which takes user-defined constraints direct procedure. illustrate approach using applications such as ray-tracing function approximation.
ic.ac.uk 
acm.org 
sci-hub.se 参考文章(21)
Chee Yap, T Dube, The exact computation paradigm World Scientific. pp. 452- 492 ,(1995)
Rahul Razdan, PRISC: programmable reduced instruction set computers Harvard University. ,(1994)
Dong-U Lee, Wayne Luk, John Villasenor, Peter Y. K. Cheung, Non-uniform Segmentation for Hardware Function Evaluation field-programmable logic and applications. pp. 796- 807 ,(2003) , 10.1007/978-3-540-45234-8_77
Mihai Budiu, Majd Sakr, Kip Walker, Seth C. Goldstein, BitValue Inference: Detecting and Exploiting Narrow Bitwidth Computations european conference on parallel processing. pp. 969- 979 ,(2000) , 10.1007/3-540-44520-X_137
Paul David Fiore, Berthold K. Horn, A custom computing framework for orientation and photogrammetry Massachusetts Institute of Technology. ,(2000)
A.A. Gaffar, O. Mencer, W. Luk, P.Y.K. Cheung, N. Shirazi, Floating-point bitwidth analysis via automatic differentiation field-programmable technology. pp. 158- 165 ,(2002) , 10.1109/FPT.2002.1188677
Altaf Abdul Gaffar, Wayne Luk, Peter Y.K. Cheung, Nabeel Shirazi, James Hwang, Automating Customisation of Floating-Point Designs field programmable logic and applications. pp. 523- 533 ,(2002) , 10.1007/3-540-46117-5_55
Barry Lee, Neil Burgess, A Dual-Path Logarithmic Number System Addition/Subtraction Scheme for FPGA field-programmable logic and applications. pp. 808- 817 ,(2003) , 10.1007/978-3-540-45234-8_78
James Hwang, Brent Milne, Nabeel Shirazi, Jeffrey D. Stroomer, System Level Tools for DSP in FPGAs Field-Programmable Logic and Applications. pp. 534- 543 ,(2001) , 10.1007/3-540-44687-7_55
Pavle Belanović, Miriam Leeser, A Library of Parameterized Floating-Point Modules and Their Use field programmable logic and applications. pp. 657- 666 ,(2002) , 10.1007/3-540-46117-5_68