Re: [Discuss-gnuradio] signal to noise ratio and multiply const

[Martin Braun (CEL)]

On Wed, Oct 30, 2013 at 01:18:53PM +0100, Baier wrote:
> Hi all,
> 
> I have simulated ofdm transmitter with 200kbps in grc. Without
> function multiply const I got a very bad spectrum mask of the
> transmiitter. If I insert the block multiply const with the 0.05 the
> spectrum mask and SNR will be much better (see attached pictures).
> How can I explain it? The dynamic range of the ofdm signal will be
> reduced if I insert the multiply const? That could be better for the
> amplifier (PAPR reducing)?
> Thanks

Hi AB,

PAPR is part of the reason. Without such a multiplier, your signal
clips.

In gr-digital/examples/ofdm/tx_ofdm.grc you will find such a multiplier
and a scope sink (this example was recently updated, you should pull it
from current master). Use the scope sink to make sure your amplitude
stays smaller than 1.0 (lots smaller!).

This does mean you're not using all the bits of the D/A converter (I
guess that's what you mean with dynamic range reduction), but you don't
need them all to produce a clean OFDM signal. On the contrary, you want
to stay away from saturating the D/A range in order to avoid signal
distortions.

As for your comment on SNR: When clipping, you're increasing the
out-of-band emissions. Technically, that's a kind of SNR but it's not
what people usually think of.

MB

PS: If you have a copy of Kammeyer's book, he explains it quite well.

-- 
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

pgp1mzIczMnSZ.pgp
Description: PGP signature