Missing functions in gnuradio?


#1

Hello,

I am relatively new to gnuradio and plan to write a weatherfax decoder
in gnuradio.

I want to try different schemes for decoding the signal (coherent /
noncoherent / short-time DFT)

As there is phasing in the signal I am most interested in the
non-coherent approach. But there I am really lacking a square-root block
in gnuradio. It should not be difficult for me to implement it - but I
wondered why such a block is not present in the library.

Should such a block (or a general “power of” block) included in the
library?

Best regards,

Horst


#2

You may be able to find an existing weatherfax decoder that can be
ported
more easily. This worked for the multi-mon project. Somebody just took
the
dos code that had been around for a while and got it to use data from
the
gnu radio instead of a sound card like it did originally.

-Clark

I want to try different schemes for decoding the signal (coherent /
Best regards,

Horst


Discuss-gnuradio mailing list
removed_email_address@domain.invalid
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


On the road to retirement? Check out MSN Life Events for advice on how
to
get there! http://lifeevents.msn.com/category.aspx?cid=Retirement


#3

On Sat, Apr 15, 2006 at 11:03:23AM +0200, Horst Eyermann wrote:

in gnuradio. It should not be difficult for me to implement it - but I
wondered why such a block is not present in the library.

Should such a block (or a general “power of” block) included in the
library?

Hi Horst,

We do have gr.rms_cf and gr.rms_ff which compute estimates of RMS
power for complex and float signals respectively.

There’s also gr.complex_to_mag for complex signals.

We generally avoid continuously computing the square root (it’s
expensive). In our examples so far we’ve generally wanted to compare
power to a threshold, and in that case it was easier to factor the
(absence of) square root into the threshold setting than to burn the
cycles computing sqrt on every sample.

Eric