FPGA Groups

FPGA Groups (http://www.fpgacentral.com/group/index.php)
-   FPGA (http://www.fpgacentral.com/group/forumdisplay.php?f=14)
-   -   Paper about selecting fixed point bit widths? (http://www.fpgacentral.com/group/showthread.php?t=63081)

[email protected] 10-24-2007 07:56 PM

Paper about selecting fixed point bit widths?
 
Hello all,

Can anyone point me to a good general purpose paper about selecting
appropriate bit-widths for a fixed point implementation of a signal
processing algorithm? I've looked around and haven't found anything
that describes a general methodology to use. I have implemented a
design and it isn't performing as well as I would like. I suspect it
has to do with some poor selection of bit slicing when it comes to
multipiers and accumulators.

Thanks for your help!


Duane Clark 10-24-2007 10:12 PM

Re: Paper about selecting fixed point bit widths?
 
[email protected] wrote:
> Hello all,
>
> Can anyone point me to a good general purpose paper about selecting
> appropriate bit-widths for a fixed point implementation of a signal
> processing algorithm? I've looked around and haven't found anything
> that describes a general methodology to use. I have implemented a
> design and it isn't performing as well as I would like. I suspect it
> has to do with some poor selection of bit slicing when it comes to
> multipiers and accumulators.


I don't know any papers. Personally, I do an implementation of an
algorithm in Matlab, first with floating point numbers, and then with
integers. I can then make quick changes and graph the two
implementations to compare the results. That allows me to determine how
many bits I need for various portions of the algorithm to get acceptable
results.

By the way, I tried the Matlab fixed point package, but found that the
performance was too poor to use on processing even a modest sized data
set, so I stick with using ordinary integers. Typically I stick a
comment at the end of most lines indicating the scaling of that
particular parameter.

Florian Stock 10-25-2007 12:44 PM

Re: Paper about selecting fixed point bit widths?
 

[email protected] writes:

> Can anyone point me to a good general purpose paper about selecting
> appropriate bit-widths for a fixed point implementation of a signal
> processing algorithm? I've looked around and haven't found anything
> that describes a general methodology to use.


Here is also some Matlab based toolbox, I used
http://users.ece.utexas.edu/~bevans/...ter/index.html
.. Not very usefull for me, because it didnt convert automaticly / and
sqrt, and so I also started manual transformation.

From Washington came the Precise Framework, look for it here:
http://fsweb.olin.edu/%7Emchang/publ...ssertation.pdf

And if I remember right, there was a chapter about word length
optimization in a book from some of
the Lononder Imperial College (...some google-seconds later...):
George Constantinides, Peter Cheung und Wayne Luk
"Synthesis and Optimization of DSP Algorithms"

Florian

Marc Reinig 10-25-2007 09:48 PM

Re: Paper about selecting fixed point bit widths?
 
Try:
http://personal.bellsouth.net/y/a/yatesc/fp.pdf
http://personal.bellsouth.net/y/a/yatesc/fir.pdf

--
Marco
________________________
Marc Reinig
UCO/Lick Observatory
Laboratory for Adaptive Optics

<[email protected]> wrote in message
news:[email protected] ups.com...
> Hello all,
>
> Can anyone point me to a good general purpose paper about selecting
> appropriate bit-widths for a fixed point implementation of a signal
> processing algorithm? I've looked around and haven't found anything
> that describes a general methodology to use. I have implemented a
> design and it isn't performing as well as I would like. I suspect it
> has to do with some poor selection of bit slicing when it comes to
> multipiers and accumulators.
>
> Thanks for your help!
>





All times are GMT +1. The time now is 06:51 AM.

Powered by vBulletin® Version 3.8.0
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.
Search Engine Friendly URLs by vBSEO 3.2.0
Copyright 2008 @ FPGA Central. All rights reserved