Max SNR as Indicator of RX Performance?

[VE3MEO]

Is the Max SNR reported by DRM RX software and guessable from the DRMcalc graphing of DReaM software a good indicator of receiver performance? I'm guessing that, when propagation is good and noise and interference are low, the receiver with the higher Max SNR is the better receiver, at least in some respects. That is, it has lower phase noise, better phase-frequency linearity, lower IMD. These are the invisible characteristics while amplitude-frequency linearity is quite visible through the software's spectrum display.

I realise there are a lot of other factors that can affect SNR at any moment - fading, selective fading, man-made and atmospheric noise and interference, signal strength. Curiously, I have had the perception that received signal strength or IF SNR is not a significant determinant in the DRM SNR above a certain threshold. That some radios apparently give much higher DRM SNR than others suggests that these invisible factors are involved.

Perusing quite a few log reports, I have noticed some logs maxing out at 30 dB and others only 20 dB, even when successful audio decoding is at or near 100%. These are logs from different receivers in different locations. Would it be fair to expect that side-by-side on the same antenna that there will be a significant difference in decoding success?

What is the meaning of SNR in DRM? Is it calculated the same way in DRM RX and DReaM?

Tom

[VE3MEO]

35.4dB SNR reported by DF9RB - a record?

Bernd reports a max SNR of 35.4dB in this message. I have surveyed quite a few reports on drmrx.org and I think this is the highest I have seen.

Are there any others achieving over 30dB? Post your report of the highest SNR you have recorded in this thread with a description of your receiving system.

Thanks,
Tom

[Bob]

Hi,

These are results of my comparative test between a Winradio G303i and an RCA AR-88D vintage communications receiver.

The tests were both carried out on a DW-Radio transmission on 13790 KHz today. The first test was using the G303 fed by a Wellbrook LA5030 loop antenna feeding a Dell 8200 desktop v AR-88D/Sat Service Schneider LC mixer with longwire antenna feeding a Dell D600 laptop. For the second test I swapped the PC's and the antenna so hopefully the only variable is the receiver.

Quite interesting results (perhaps I should tell Winradio)!

regards
Bob
[Note : Incorrect DRMCalc txt files originally posted, now corrected see later message - after I deleted the attachment I couldn't reattach it to this message - apologies]

[VE3MEO]

30dB SNR with 60 yr old front end!

Originally posted by Bob
Hi,

These are results of my comparative test between a Winradio G303i and an RCA AR-88D vintage communications receiver.

[snip]
Quite interesting results (perhaps I should tell Winradio)!

regards
Bob

I'll echo that. It's probably in part due to front end overload. Isn't the G303 a wideband receiver with no preselector? And the AR88 probably has the much higher overload immunity of a tube radio with preselector.

73, Tom

[Bob]

Originally posted by Bob
Hi,

These are results of my comparative test between a Winradio G303i and an RCA AR-88D vintage communications receiver.

Corrected attachment (only txt files changed).

regards
Bob

[VE3MEO]

Test table

Anyone interested in logging the highest SNR you have ever achieved with DRMRX/DReaM in a published table something like this?

<table>
<tr>
<B>
<td>Max SNR
<td>By
<td>Receiver
<td>Converter
<td>Antenna
<td>Service
<td>Freq(kHz)
<td>Date yyyy-mm-dd
</B>
</tr>
<tr>
<td>35.4 dB
<td>Bernd DF9RB
<td>TS850 AGC off
<td>
<td>
<td>DW Sines
<td>13790
<td>2004-09-07
</tr>

<tr>
<td>34.5 dB
<td>carknue
<td>AOR-7030
<td>?
<td>Wellbrook ALA1530P
<td>VoR
<td>15780
<td>2004-08-20
</tr>

<tr>
<td>30.0 dB
<td>Bob
<td>AR88D (60 yrs young!)
<td>Sat-Svc Scheider LC
<td>long wire
<td>DW Sines
<td>13790
<td>2004-09-10
</tr>

</table>

73, Tom

[VE3MEO]

Higher SNR not necessarily better decoding

Attached is an zip file with two screen shots of DReaM 1.1.1cvs - one with higher SNR and no audio decoded, the other with lower SNR and successful audio, both from the same test signal and system. It shows that the calculated DRM SNR is not an ideal indicator of receiver performance and optimal tuning. A receiver that results in a slightly higher reported SNR may not actually decode audio as successfully as the lesser one!

In the examples, the receiver is a Radio Shack DX-394A operating with no AGC into a NE612 downconverter with two 15kHz filters cascaded and the test transmitter is a NE612 upconverter with the DReaM transmitter. The only difference between the two is the tuning of the DX-394 - 10,252.4 kHz for lower SNR and 10,252.5 kHz for higher SNR but no audio.

This shift in frequency crosses a boundary that shifts the desired signal 5kHz at the 1st IF of 45MHz. The 5kHz shift at the 1st IF repositions the DRM signal in the passband of the relatively broad 45 MHz ceramic filter, resulting in the tilt of the spectrum changing between slightly downward and pronounced upward.

Apparently the algorithm for calculating SNR is more generous with the upward tilt or the consequential slight lifting of the central peak. But estimated channel impulse response shows a spreading or delay nearly double and audio is not decoded.

So higher SNR is not the be all and end all to highest success rate for audio decoding. Other factors must be considered. Apparently, impulse response is one of them - hence the interest in group delay distortion, to the extent that it affects impulse response.

73, Tom

[VE3MEO]

Re: Test table

Highest SNR achieved with DRMRX/DReaM:

<table>
<tr>
<B>
<td>Max SNR
<td>By
<td>Receiver
<td>Converter
<td>Antenna
<td>Service
<td>Freq(kHz)
<td>Date yyyy-mm-dd
</B>
</tr>
<tr>
<td>35.4 dB
<td>Bernd DF9RB
<td>TS850 AGC off
<td>
<td>
<td>DW Sines
<td>13790
<td>2004-09-07
</tr>

<tr>
<td>34.5 dB
<td>carknue
<td>AOR-7030
<td>?
<td>Wellbrook ALA1530P
<td>VoR
<td>15780
<td>2004-08-20
</tr>

<tr>
<td>31.4 dB
<td>VE3MEO
<td>Radio Shack DX-394A no AGC nor 2nd IF fltr
<td>NE612 protoboard
<td>G5RV
<td>CBC Sackville
<td>6140
<td>2004-09-29
</tr>

<tr>
<td>30.0 dB
<td>Bob
<td>AR88D (60 yrs young!)
<td>Sat-Svc Scheider LC
<td>long wire
<td>DW Sines
<td>13790
<td>2004-09-10
</tr>

</table>

73, Tom

[VE3MEO]

Max SNR table

Highest SNR achieved with DRMRX/DReaM:

<table>

<tr>
<td>Max SNR
<td>By
<td>Receiver
<td>Converter
<td>Antenna
<td>Service
<td>Freq(kHz)
<td>Date yyyy-mm-dd
</tr>


<tr>
<td>35.6 dB
<td>arniek
<td>FRG-100
<td>Sat-Service Scheider
<td>homemade loop
<td>DW Sines
<td>5980
<td>2004-10-08
</tr>

<tr>
<td>35.4 dB
<td>Bernd DF9RB
<td>TS850 AGC off
<td>
<td>
<td>DW Sines
<td>13790
<td>2004-09-07
</tr>

<tr>
<td>34.5 dB
<td>carknue
<td>AOR-7030
<td>?
<td>Wellbrook ALA1530P
<td>VoR
<td>15780
<td>2004-08-20
</tr>

<tr>
<td>33.4 dB
<td>dn0758
<td>elektor
<td>
<td>long wire
<td>DW Sines
<td>5980
<td>2004-08-20
</tr>

<tr>
<td>31.4 dB
<td>VE3MEO
<td>Radio Shack DX-394A no AGC nor 2nd IF fltr
<td>NE612 protoboard
<td>G5RV
<td>CBC Sackville
<td>6140
<td>2004-09-29
</tr>

<tr>
<td>30.0 dB
<td>Bob
<td>AR88D (60 yrs young!)
<td>Sat-Svc Scheider LC
<td>long wire
<td>DW Sines
<td>13790
<td>2004-09-10
</tr>

</table>

73, Tom

[arniek]

Hi all,

Even with the best RX performance you will never see high SNR values if you live in an area with much man-made noise.

My results in DRM-reception are best when i rotate the loop antenna for maximum SNR.

Using a longwire won't allow for such high SNR values in most cases as you're unable to rotate it ;-)

The better signal to noise ratio of loop antenna compensates for lower signal strength.

73

Arne

[VE3MEO]

Re: Max SNR table

Highest SNR achieved with DRMRX/DReaM:

<table align=LEFT border="1"><tr><td>Max SNR<td>By<td>Receiver<td>Converter<td>Antenna<td>S ervice<td>Freq(kHz)<td>Date
<tr><td>35.6 dB
<td>arniek
<td>FRG-100
<td>Sat-Schneider
<td>homemade loop
<td>DW Sines
<td>5980
<td>2004-10-08
<tr>
<td>35.4 dB
<td>Bernd DF9RB
<td>TS850 AGC off
<td>
<td>
<td>DW Sines
<td>13790
<td>2004-09-07
<tr>
<td>34.5 dB
<td>carknue
<td>AOR-7030
<td>?
<td>Wellbrook ALA1530P
<td>VoR
<td>15780
<td>2004-08-20
<tr>
<td>33.4 dB
<td>dn0758
<td>elektor
<td>
<td>long wire
<td>DW Sines
<td>5980
<td>2004-08-20
<tr>
<td>31.4 dB
<td>VE3MEO
<td>Radio Shack DX-394A no AGC nor 2nd IF fltr
<td>NE612 protoboard
<td>G5RV
<td>CBC Sackville
<td>6140
<td>2004-09-29
<tr>
<td>30.2 dB
<td>simone
<td>FRG8800
<td>Sat-Schneider
<td>whip
<td>DW Sines
<td>15435
<td>2004-10-09
<tr>
<td>30.0 dB
<td>Bob
<td>AR88D (60 yrs young!)
<td>Sat-Schneider LC
<td>long wire
<td>DW Sines
<td>13790
<td>2004-09-10
</table>


73, Tom

[VE3MEO]

Re: Re: Max SNR table

Highest SNR's over 30 dB achieved with DRMRX/DReaM:

<table align=LEFT border="1">
<tr><th>Max SNR
<th>By
<th>Receiver
<th>Converter
<th>Antenna
<th>Service
<th>Freq(kHz)
<th>Date</th>
<tr><td>35.6 dB
<td>arniek
<td>FRG-100
<td>Sat-Schneider
<td>homemade loop
<td>DW Sines
<td>5980
<td>2004-10-08
<tr><td>35.4 dB
<td>Bernd DF9RB
<td>TS850 AGC off
<td>
<td>
<td>DW Sines
<td>13790
<td>2004-09-07
<tr><td>34.5 dB
<td>carknue
<td>AOR-7030
<td>?
<td>Wellbrook ALA1530P
<td>VoR
<td>15780
<td>2004-08-20
<tr><td>33.4 dB
<td>dn0758
<td>elektor
<td>
<td>long wire
<td>DW Sines
<td>5980
<td>2004-08-20
<tr><td>31.7 dB
<td>hseppo
<td>AD8099, AD9834 DDS TUF-3 mxr
<td>?
<td>?
<td>DW Sines
<td>15440
<td>2004-10-10
<tr><td>31.4 dB
<td>VE3MEO
<td>Radio Shack DX-394A no AGC nor 2nd IF fltr
<td>NE612 protoboard
<td>G5RV
<td>CBC Sackville
<td>6140
<td>2004-09-29
<tr><td>30.2 dB
<td>simone
<td>FRG8800
<td>Sat-Schneider
<td>whip
<td>DW Sines
<td>15435
<td>2004-10-09
<tr><td>30.0 dB
<td>Bob
<td>AR88D (60 yrs young!)
<td>Sat-Schneider LC
<td>long wire
<td>DW Sines
<td>13790
<td>2004-09-10</td></tr>
</table>

73, Tom

[VE3MEO]

Re: Re: Re: Max SNR table

Highest SNR's over 30 dB achieved with DRMRX/DReaM

If you have results that leap the 30 dB hurdle, e-mail me with the details and I'll add it to the list.

73, Tom

<table align=LEFT border="1">
<tr><th>Max SNR
<th>By
<th>Receiver
<th>Converter
<th>App
<th>Antenna
<th>Service
<th>Freq(kHz)
<th>Date</th>
<tr><td>35.6 dB
<td>arniek
<td>FRG-100
<td>Sat-Schneider
<td>DReaM 1.1.1c
<td>homemade loop
<td>DW Sines
<td>5980
<td>2004-10-08
<tr><td>35.4 dB
<td>Bernd DF9RB
<td>TS850 AGC off
<td>
<td>DReaM 1.1.1c
<td>
<td>DW Sines
<td>13790
<td>2004-09-07
<tr><td>34.5 dB
<td>carknue
<td>AOR-7030
<td>?
<td>DRMSWR 2.0.34
<td>Wellbrook ALA1530P
<td>VoR
<td>15780
<td>2004-08-20
<tr><td>33.4 dB
<td>dn0758
<td>elektor
<td>
<td>DRMSWR 2.0.34
<td>long wire
<td>DW Sines
<td>5980
<td>2004-08-20
<tr><td>31.8 dB
<td>VE3MEO
<td>Radio Shack DX-394A no AGC nor 2nd IF fltr
<td>LF-D6 filter/NE612 protoboard
<td>DReaM 1.1.1c
<td>G5RV
<td>CBC Sackville
<td>6140
<td>2004-10-19
<tr><td>31.7 dB
<td>hseppo
<td>AD8099, AD9834 DDS TUF-3 mxr
<td>?
<td>DReaM 1.1.1c
<td>?
<td>DW Sines
<td>15440
<td>2004-10-10
<tr><td>30.7 dB
<td>simone
<td>FRG8800
<td>Sat-Schneider
<td>DRMSWR 2.0.34
<td>whip
<td>DW Sines
<td>15435
<td>2004-10-10
<tr><td>30.0 dB
<td>Bob
<td>AR88D (60 yrs young!)
<td>Sat-Schneider LC
<td>DRMSWR 2.0.34
<td>long wire
<td>DW Sines
<td>13790
<td>2004-09-10</td></tr>
</table>

73, Tom

[VE3MEO]

Max SNR table

Highest SNR's over 30 dB achieved with DRMRX/DReaM

If you have results that leap the 30 dB hurdle, e-mail me with the details and I'll add it to the list.

73, Tom

<table align=LEFT border="1">
<tr><th>Max SNR
<th>By
<th>Receiver
<th>Converter
<th>App
<th>Antenna
<th>Service
<th>Freq(kHz)
<th>Date</th>
<tr><td>37.4 dB
<td>Bernd DF9RB
<td>TS850 AGC inactive
<td>SA612 ceramic resonator
<td>DReaM 1.1.6c
<td>3-el ham Yagi
<td>DW Sines
<td>11830
<td>2004-11-24
<tr><td>36.5 dB
<td>CT1ADT
<td>TS-870 unmodified
<td>11,3 kHz out
<td>DRM 2.0.34
<td>8m vert or 24m dipole
<td>VOR
<td>15780
<td>2004-01-07
<tr><td>35.7 dB
<td>hseppo
<td>AD8099, AD9834 DDS TUF-3 mxr
<td>?
<td>DReaM 1.1.4c
<td>5m wire
<td>DW Sines
<td>17800
<td>2004-11-07
<tr><td>35.7 dB
<td>simone
<td>FRG100 AGC off
<td>Sat-Schneider
<td>DRMSWR 2.0.34
<td>loop
<td>DW Sines
<td>9685
<td>2004-11-23
<tr><td>35.6 dB
<td>arniek
<td>FRG-100
<td>Sat-Schneider
<td>DReaM 1.1.1c
<td>homemade loop
<td>DW Sines
<td>5980
<td>2004-10-08
<tr><td>35.0 dB
<td>carknue
<td>AOR-7030
<td>Sat-Schneider xtal
<td>DRMSWR 2.0.34
<td>Wellbrook ALA1530P
<td>VoR
<td>15780
<td>2004-01-10
<tr><td>34.9 dB
<td>dn0758
<td>elektor
<td>
<td>DRMSWR 2.0.34
<td>Sony AN-1
<td>DW Sines
<td>17800
<td>2004-11-20
<tr><td>34.4 dB
<td>chemclouds
<td>Icom IC-R75 AGC off
<td>
<td>DRMSWR 2.0.34
<td>30m inverted-L
<td>CBC Sackville
<td>9800
<td>2004-11-27
<tr><td>34.3 dB
<td>FritzWue DK6QI
<td>FT840
<td>Sat-Schneider
<td>DReaM 1.1.6c
<td>2x 40m LW
<td>DW-Sines
<td>17800
<td>2004-12-18
<tr><td>33.4 dB
<td>df8uo
<td>FT757GX
<td>Sat-Schneider xtal
<td>DReaM 1.1.1c
<td>dipole
<td>VoR
<td>15780
<td>2004-11-06
<tr><td>33.2 dB
<td>G3VRF
<td>AD9852 DDS, AD831 Mixer, 2x AD603 IF
<td>12kHz DRM xtal ctld
<td>DReaM 1.1.6c
<td>35cm tuned loop in attic 11m a.g.
<td>DW Sines
<td>11830
<td>2004-11-24
<tr><td>31.8 dB
<td>VE3MEO
<td>Radio Shack DX-394A no AGC nor 2nd IF fltr
<td>LF-D6 filter/NE612 protoboard
<td>DReaM 1.1.1c
<td>G5RV
<td>CBC Sackville
<td>6140
<td>2004-10-19
<tr><td>30.7 dB
<td>simone
<td>FRG8800
<td>Sat-Schneider
<td>DRMSWR 2.0.34
<td>whip
<td>DW Sines
<td>15435
<td>2004-10-10
<tr><td>30.4 dB
<td>TimoHannes
<td>elektor
<td>?
<td>DRMSWR 2.0.34
<td>active
<td>DW Sines
<td>17800
<td>2004-11-14
<tr><td>30.0 dB
<td>Bob
<td>AR88D (60 yrs young!)
<td>Sat-Schneider LC
<td>DRMSWR 2.0.34
<td>long wire
<td>DW Sines
<td>13790
<td>2004-09-10</td></tr>
</table>

73, Tom

[VE3MEO]

Re: Max SNR table

Highest SNR's over 30 dB achieved with DRMRX/Dream

Congrats to TimoHannes, chemcloud, df8uo, CT1ADT, hseppo for making it onto or climbing up the list since it was last posted.

Rather than re-posting it each time, I've been editing the list, so the latest results will always have the same URL.

If you have results that leap the 30 dB hurdle, e-mail me with the details and I'll add it to the list.

If you are on the list but some details are missing, please send me the needed info.

73, Tom