Automatic Floating-Point to Fixed-Point Transformations

Han, Kyungtae; Olson, A.G.; Evans, Lance S.

doi:10.1109/acssc.2006.356588

Cited by 12 publications

(9 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, we have computed the coefficients of several low-pass and high-pass FIR filters for different number of taps using Matlab. Then, the floating-point version of the filters, considering the direct-form structure, has been optimized for fixed-point computation using "FloatingPoint to Fixed-Point Transformation Toolbox" [7]. For a given error threshold, this Matlab toolbox optimizes the worth-length combination of the given DSP data-path to minimize the estimated area cost of the circuit when implemented it on FPGA.…”

Section: Results and Comparisonmentioning

confidence: 99%

Optimizing DSP circuits by a new family of arithmetic operators

Hormigo

Villalba

2014

2014 48th Asilomar Conference on Signals, Systems and Computers

View full text Add to dashboard Cite

show abstract

Section: Results and Comparisonmentioning

confidence: 99%

Optimizing DSP circuits by a new family of arithmetic operators

Hormigo

Villalba

2014

2014 48th Asilomar Conference on Signals, Systems and Computers

View full text Add to dashboard Cite

show abstract

“…The linear fitting method is introduced in reference [12] to implement the Logarithmic 2 module. It divides (1,2] span into 128 segments, while each segment corresponds to a slope and a bias. Computation begins with shifting input to (1,2] span, then the section is chosen correspondingly to get the result.…”

Section: A a Look-up Table Methods On Logarithm 1) Arithmetic Descripmentioning

confidence: 99%

“…According to extreme value theory, maximum and minimum are converged to Gumbel distribution. Maximum is got by (2).…”

Section: ) Extreme Value Theory Approachmentioning

confidence: 99%

A Novel Fixed-Point Simulation Library for the Design of FPGAs

Liu¹,

Zhang²,

Chang³

et al. 2010

IJIEEB

View full text Add to dashboard Cite

Fixed-point simulation is extremely important in the design of fixed-point FPGAs. Float-point simulation is used to verify the arithmetic, while the fixed-point simulation is adopted to evaluate the performance and verify the implementation. The design of high-precision system based on FPGAs must focus on the fixed-point simulation to reduce the error acceptable even to zero. Basing on the analysis of fixed-point simulation approaches, especially the approach with MATLAB Fixed-Point Toolbox, we propose a novel fixed-point simulation library consisting of four modules. The library works under VC environment and its basic definition module imitates operating principle of MATLAB Fixed-Point Toolbox. The aim of library is to assit setting up fixed-point simulation conveniently, easily and quickly. Finally, simulation shows effect of the library. Index Terms -fixed-point simulation, library, zero error, floatpoint to fixed-point conversion

show abstract

“…However, this work was limited to monoobjective optimization only. More recently, Han et al [49] reported EA-based multiobjective wordlength optimization for a filtering application. This work, however, considered only linear objective functions and lacked postsynthesis validation.…”

Section: Discussionmentioning

confidence: 99%

Multiobjective Optimization for Reconfigurable Implementation of Medical Image Registration

Dandekar

Plishker

Bhattacharyya

et al. 2008

International Journal of Reconfigurable Computing

View full text Add to dashboard Cite

In real-time signal processing, a single application often has multiple computationally intensive kernels that can benefit from acceleration using custom or reconfigurable hardware platforms, such as field-programmable gate arrays (FPGAs). For adaptive utilization of resources at run time, FPGAs with capabilities for dynamic reconfiguration are emerging. In this context, it is useful for designers to derive sets of efficient configurations that trade off application performance with fabric resources. Such sets can be maintained at run time so that the best available design tradeoff is used. Finding a single, optimized configuration is difficult, and generating a family of optimized configurations suitable for different run-time scenarios is even more challenging. We present a novel multiobjective wordlength optimization strategy developed through FPGA-based implementation of a representative computationally intensive image processing application: medical image registration. Tradeoffs between FPGA resources and implementation accuracy are explored, and Pareto-optimized wordlength configurations are systematically identified. We also compare search methods for finding Pareto-optimized design configurations and demonstrate the applicability of search based on evolutionary techniques for identifying superior multiobjective tradeoff curves. We demonstrate feasibility of this approach in the context of FPGA-based medical image registration; however, it may be adapted to a wide range of signal processing applications.

show abstract

Automatic Floating-Point to Fixed-Point Transformations

Cited by 12 publications

References 19 publications

Optimizing DSP circuits by a new family of arithmetic operators

Optimizing DSP circuits by a new family of arithmetic operators

A Novel Fixed-Point Simulation Library for the Design of FPGAs

Multiobjective Optimization for Reconfigurable Implementation of Medical Image Registration

Contact Info

Product

Resources

About