discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] Big vector problem


From: Martin Braun
Subject: Re: [Discuss-gnuradio] Big vector problem
Date: Mon, 6 Jun 2011 10:27:31 +0200
User-agent: Mutt/1.5.20 (2009-06-14)

On Mon, Jun 06, 2011 at 09:33:52AM +0200, Michael Höin wrote:
> Hi all
> 
> I have a problem with a big vector. The input/output of my entity is
> like:
> : gr_block ("sub_channel_vff",
>              gr_make_io_signature2 (2, 2, sizeof (float)*3072, sizeof
> (char)),
>              gr_make_io_signature2 (1, 2, sizeof (float)*5376))
> 
> And my forecast is like:
> ninput_items_required[i] = 76/4;
> 
> The problem is that one input Frame has a size about 76*3072= 233472
> values and I cut four parts (at different positions) of size 5376 out of
> it.
> If I run the simulation I get an error:
> .....is requesting more input data than we can provide.
>   ninput_items_required = 19
>   max_possible_items_available = 4....
> 
> If I reduce the ninput_items_required value the simulation works, but
> the value reflectes not the truth and in the gr_block file is written:
> "The estimate doesn't have to be exact, but should be close". Knows
> anybody a solution or a detour to work with this big vectors?
> 
> Any idea? Any help will be appreciated


Hi Michael,

without knowing what you're block's supposed to do, two solutions come
to mind:

1) You re-think the functionality of the block. Does it really need
these massive input vectors?

2) You re-write the block to operate on streams. This should not be a
problem since you've already chosen a gr_block instead of a
gr_sync_block. You might need some internal buffers (that operate
outside the buffers provided by GNU Radio) and some state variables to
remember where you are in the stream.
Are you simply cutting stuff out of a stream? In that case, a re-write
should be fairly simple.

MB

-- 
Karlsruhe Institute of Technology (KIT)
Communications Engineering Lab (CEL)

Dipl.-Ing. Martin Braun
Research Associate

Kaiserstraße 12
Building 05.01
76131 Karlsruhe

Phone: +49 721 608-43790
Fax: +49 721 608-46071
www.cel.kit.edu

KIT -- University of the State of Baden-Württemberg and
National Laboratory of the Helmholtz Association

Attachment: pgpBv30ERbHaa.pgp
Description: PGP signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]