DRM Software Radio Forums  

Go Back   DRM Software Radio Forums > DRM Software Radio - User Forums > General Topics
User Name
Password
FAQ Members List Calendar Search Today's Posts Mark Forums Read

Reply
 
Thread Tools Search this Thread Display Modes
Old 17-11-2004, 18:47   #136
cesco
Registered User
 
Join Date: May 2004
Posts: 11
Quote:
Originally posted by carknue


Why should the reception of a 20khz signal not be possible with a normal soundcard? I already received 20 khz DRM signals with a quite normal soundblastercard.


Because most probably your receiver will have a +-6khz filter. Byebye 20 khz.....
cesco is offline   Reply With Quote
Old 17-11-2004, 21:33   #137
dk8cb
Roland
 
Join Date: Mar 2004
Location: Munich, Germany
Posts: 2,748
1.1.6cvs - AM mode

Hi,

in AM mode, Dream's new 1.1.6cvs version now employs new filters that are implemented as windowed FFT filters.

The result is a much better stop band attenuation. Heterodynes originating from stations 10 kHz up or down are no longer audible when receiving AM with a wide IF filter in the signal path.

Roland
dk8cb is offline   Reply With Quote
Old 19-11-2004, 01:21   #138
VE3MEO
Registered User
 
Join Date: Oct 2002
Location: Cobourg, Canada
Posts: 289
Re: 1.1.6cvs - AM mode

Quote:
Originally posted by dk8cb
in AM mode, Dream's new 1.1.6cvs version now employs new filters

They sound very good, have steep skirts, no growling or buzzing artefacts when the skirt is on a carrier the way SDRadio does, and now the clicking noise as you slide the bandwidth control is eliminated. Nice!

73, Tom
VE3MEO is offline   Reply With Quote
Old 20-11-2004, 11:05   #139
DigiBC
Registered User
 
Join Date: May 2003
Location: near Cologne, Germany (51°00'N / 6°49'E)
Posts: 36
hamlib 1.2.3

Hi!

A new version of the hamlib library (1.2.3) is available.

There are no problems compiling Dream 1.1.6cvs with the updated library.
Don't forget to use the new hamlib DLL's too.

For more information visit: http://www.tu-darmstadt.de/fb/et/uet...tallation.html

Regards - DigiBC
DigiBC is offline   Reply With Quote
Old 26-11-2004, 19:08   #140
dk8cb
Roland
 
Join Date: Mar 2004
Location: Munich, Germany
Posts: 2,748
Hi,

the new CVS version of Dream features a Weighted Modulation Error Ratio (WMER) plot, that shows the weighted MER as a function of the carrier index.

There used to be some information about the significance of the MER in this paper but unfortunately, the document is currently unavailable as I am writing this.
Some information can also be found in this ETSI standard.

Attached is a screenshot of the new plot.

Last edited by dk8cb : 27-11-2004 at 09:58.
dk8cb is offline   Reply With Quote
Old 27-11-2004, 09:56   #141
dk8cb
Roland
 
Join Date: Mar 2004
Location: Munich, Germany
Posts: 2,748
What the WMER plot can be used for

Hi,

let's assume, there is interference from a signal that is weaker than the DRM signal. You would not normally be able to see this interfering signal in the spectrum plot, because it is buried somewhere under the DRM signal.
However, in case the interference is not of a wideband type but more of a narrowband type, certain carriers will nevertheless be affected more than others by the interfering signal. The new plot will reveal which carrier or which range of carriers is affected by the interference signal.
One might even calculate the frequency on which this interference occurs.

Roland

Last edited by dk8cb : 27-11-2004 at 11:34.
dk8cb is offline   Reply With Quote
Old 27-11-2004, 14:18   #142
dk8cb
Roland
 
Join Date: Mar 2004
Location: Munich, Germany
Posts: 2,748
Practical example

Hi,

here is an example of such a WMER plot taken under special conditions. It was taken while receiving VT-Digital on 9875 kHz with another DRM signal from Kuwait on 9880 kHz.
It can be clearly seen that half of the 9875 kHz carriers are affected by the interference and that there is substantial "noise" on them, degrading the overall decoding result.

Roland
dk8cb is offline   Reply With Quote
Old 28-11-2004, 07:39   #143
carknue
Registered User
 
carknue's Avatar
 
Join Date: Dec 2002
Location: Frankfurt, Germany
Posts: 1,321
Thumbs up

The snr spectrum is really a nice feature! This shot was taken from a vor transmission on 15780 khz. There was a fat data carrier on 15777 khz that could not always been seen in the normal spectrum, but clearly in the snr spectrum.
__________________
regards, Carsten
AOR 7030 + Wellbrook ALA 1530P & DWT + Degen DE31 + TG33

Free DRM Tools for analysing, controling and automating your DRM reception and logging
carknue is offline   Reply With Quote
Old 28-11-2004, 16:52   #144
dk8cb
Roland
 
Join Date: Mar 2004
Location: Munich, Germany
Posts: 2,748
Quote:
Originally posted by carknue
The snr spectrum is really a nice feature! This shot was taken from a vor transmission on 15780 khz. There was a fat data carrier on 15777 khz that could not always been seen in the normal spectrum, but clearly in the snr spectrum.

Here is a similar screenshot, taken while receiving 7320 kHz today. The cause of the interference was a station using frequency shift keying (FSK) with 800 Hz shift. This FSK signal was completely invisible on the spectrum plot. The averaged SNR should have been good enough for perfect reception, however there were dropouts as a result of the interference from the FSK signal.

Roland
dk8cb is offline   Reply With Quote
Old 28-11-2004, 20:01   #145
tacitus-ms
Registered User
 
tacitus-ms's Avatar
 
Join Date: Apr 2004
Location: 52°04'N 7°42'E = Münster / Germany
Posts: 40
Thanks for the new WMER plot.

Thanks for the new WMER plot. I think this is the most useful feature to find cochannel interferences. Recently I have got the Wellbrook ALA 1530 and I've mouted it on an antenna rotor. Now I can use the WMER plot to turn this magnetic loop to find the minimum of the interferer. Now here in Münster I can receive BBC 1296 kHz as stabile as local FM stations without any dropouts.

Last edited by tacitus-ms : 28-11-2004 at 21:30.
tacitus-ms is offline   Reply With Quote
Old 28-11-2004, 22:39   #146
tacitus-ms
Registered User
 
tacitus-ms's Avatar
 
Join Date: Apr 2004
Location: 52°04'N 7°42'E = Münster / Germany
Posts: 40
Strange results of the WMER plot in case of DW 3995 kHz.

Strange results of the WMER plot in case of DW 3995 kHz. Normally if all carriers have the same WMER SNR the total SNR of the whole signal is nearly the same too. I saw this in case of BBC World, in case of several Sines transmissions. But now I saw this: All carriers neary about 17 dB, but the SNR of the whole signal was about 23 dB. See screenshot. Can that be true?

tacitus-ms
tacitus-ms is offline   Reply With Quote
Old 29-11-2004, 08:45   #147
corrados
Registered User
 
corrados's Avatar
 
Join Date: Nov 2003
Location: Darmstadt
Posts: 10
The WMER for the "SNR spectrum" is a decision-based algorithm on the MSC cells. Since the MSC cells are usually 64-QAM modulated, the decisions might be incorrect quite often (especially at low SNRs). So the SNR value based on the FAC cells (which are modulated with a QPSK) will always be better. That explains the difference.

The "SNR spectrum" should only be used for determining if interferers are present or not.

Volker
corrados is offline   Reply With Quote
Old 29-11-2004, 19:06   #148
tacitus-ms
Registered User
 
tacitus-ms's Avatar
 
Join Date: Apr 2004
Location: 52°04'N 7°42'E = Münster / Germany
Posts: 40
Automatic filter based on WMER?

When I observed the WMER SNR spectrum I got an idea, but I do not know if it could work. Is the WMER measurement algorithm good enough to control an automatic filter, which removes the carriers with low SNR which cannot be used in case of interferences? Would that make sense and could that be possible? Or is it better to have carriers with low SNR than a gap with nothing? If this needs to much processor power for a software solution, could that be useful for hardware based solutions in the future? Does anyone have an idea?

regards tacitus-ms.

Last edited by tacitus-ms : 29-11-2004 at 19:12.
tacitus-ms is offline   Reply With Quote
Old 30-11-2004, 07:31   #149
corrados
Registered User
 
corrados's Avatar
 
Join Date: Nov 2003
Location: Darmstadt
Posts: 10
See

https://sourceforge.net/forum/forum....orum_id=242204

Volker
corrados is offline   Reply With Quote
Old 03-12-2004, 23:43   #150
dk8cb
Roland
 
Join Date: Mar 2004
Location: Munich, Germany
Posts: 2,748
Hi,

the latest CVS version of Dream contains a new Modified Metric switch, see attachment.

If activated, the channel estimates are weighted with the SNR estimate of the respective carrier to take into account the higher noise variance of the respective carrier that results from an interfering signal. This is supposed to increase resistance against interference.

So far, when running both old and new versions in parallel, I have not noticed any improvement in interference situations, but I still have to do a few more tests.

Roland
dk8cb is offline   Reply With Quote
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is Off
HTML code is On
Forum Jump


All times are GMT. The time now is 00:12.


Powered by vBulletin Version 3.5.3
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.