How do I Calculate CNR or SNR from Spectrum Plot?

[DRCars]

I have been using satsignature.com for looking at spectrum plots. The spectrum is displayed in dBm. Can I estimate a peaks CNR (or maybe it's SNR) by dividing the height of a peak by an estimate of the height of the noise near it?

For instance, a peak maxes out at -65 dBm starting from a baseline of -75 dBm for a difference of 10 dBm. Near this peak there is noise that varies from -74 to - 76 dBm for a difference of 2 dBm. Would a crude estimate of the CNR be 10 / 2 = 5? Even If this is correct I'm not sure whether it's a CNR or SNR.

Furthermore, I would like to be able to decide what size dish they are using on a particular satellite by maybe comparing an estimate of my CNR (or SNR) to what I see on their dish. That way I could possibly decide whether I should be able to receive a particular feed on my puny 7.5' dish by looking at their plot. Any way to maybe do that?

For instance, if it turns out that I estimate they have a 10' dish (twice the area of mine) then I think I would subtract 3 dB from any CNR/SNR that I calculate from their spectrum plots to crudely approximate my expected CNR/SNR. Or am I just way out in left field?

Your take on this would be appreciated.

DRCars

 

[Cham]

Best you can do is estimate, too many variables to know what SNR you will get since you don't have a local noise level to use in the equasion. Subtracting 3dB from their signal level would be a start I suppose, and using their noise level and maybe adding a dB or two for good measure... Also the particular equipment used is rather important in the reception chain; feedhorn type, LNB (LNBF), cabling, aiming accuracy; all add to or subtracts from SNR. Only way to know for sure is to set up your dish and take the measurments.

 

[DRCars]

Thanks for the reply. I have a dish set up already for many years now. Yeah, I'm aware there are a lot of variables involved. Just curious if this is at all practical.

Here is what I have done this morning.

On G16 there are 2 identical feeds side by side. BTW, they are the last 2 shown in the spectrum plots at satsignature.us (occasional signals). I guess I can't say more than that here in the open. One is receivable with my 7.5' dish the other is not. They are both DVB-S2 8PSK with FEC=3/4. According to a list I found that would require 7.9 dB CNR to receive.

Minimum CNRs
DVB-S QPSK
1/2 2.7
2/3 4.4
3/4 5.5
5/6 6.5
7/8 7.2
DVB-S2 QPSK
1/4 -2.4
1/3 -1.2
2/5 0
1/2 1
3/5 2.2
2/3 3.1
3/4 4
4/5 4.6
5/6 5.2
8/9 6.2
9/10 6.5
DVB-S2 8PSK
3/5 5.5
2/3 6.6
3/4 7.9
5/6 9.4
8/9 10.6
9/10 11
DVB-S2 16APSK
2/3 9
3/4 10.2
4/5 11
5/6 11.6
8/9 12.9
9/10 13.1
DVB-S2 32APSK
3/4 12.6
4/5 13.6
5/6 14.3
8/9 15.7
9/10 16.1​

The 1st feed gave me a 59-60% quality on my receiver and would not stay locked. The 2nd gave me 60-61% and maintains lock...barely. Doing the crude math I described above in 1st post over at satsignature.com gave me 9.9 dB CNR on 1st feed and 11.1 on the 2nd. Unless this is just blind luck coincidence, it appears they use a 10' dish, since my 7.5' dish would produce predicted CNR/SNRs of 6.9 and 8.1, respectively, for the 2 feeds. One below threshold the other above. Which is what I see.

Blind luck ? Coincidence?

Edit: Those signals are gone now...for today. Should be back tomorrow around 12:00 PM CDT.

DRCars

 

[Cham]

You should be able to pick up the ABC feeds at 4120Mhz, symb rate (try 30000) but they are 9/10 FEC and really strong, always present wether there is programming running or not. Should be able to use those for a reference signal on your 7.5 dish.

 

[DRCars]

Thanks Cham. Yes, those are strong, even on my puny dish. All I really hope to accomplish with this crude method is to know whether or not to try to tweak and peak a signal to get it above threshold. Kind of like asking if other people are able to get a signal, except I would ask satsignature.us.

Found an occasional feed on G3 where the "little white ball being knocked about" feeds are. They usually use DVB-S2 8PSK and I think FEC 3/4. The crude satsignature CNR analysis showed it to be about 10 dB (maybe 10.5). If they use a 10' vs. my 7.5' dish then 10-3=7 dB (at most 7.5 dB). Blind scan found it but didn't have enough quality to show it. To get it I would need at least 7.9 dB if it's a FEC 3/4 feed. If higher FEC it would require even more.

This crude technique seems to be producing somewhat predictable results. At least I know I probably won't be getting this G3 feed even by tweaking (which I did to no avail).

Edit: I've been referring to satsignature.com part of the time. It will have you go to satsignature.us .

DRCars

 

[DodgeDart]

Thanks Cham. Yes, those are strong, even on my puny dish. All I really hope to accomplish with this crude method is to know whether or not to try to tweak and peak a signal to get it above threshold. Kind of like asking if other people are able to get a signal, except I would ask satsignature.us.

Found an occasional feed on G3 where the "little white ball being knocked about" feeds are. They usually use DVB-S2 8PSK and I think FEC 3/4. The crude satsignature CNR analysis showed it to be about 10 dB (maybe 10.5). If they use a 10' vs. my 7.5' dish then 10-3=7 dB (at most 7.5 dB). Blind scan found it but didn't have enough quality to show it. To get it I would need at least 7.9 dB if it's a FEC 3/4 feed. If higher FEC it would require even more.

This crude technique seems to be producing somewhat predictable results. At least I know I probably won't be getting this G3 feed even by tweaking (which I did to no avail).

Edit: I've been referring to satsignature.com part of the time. It will have you go to satsignature.us .

DRCars

I think you just subtract the carrier level from the noise next to it. You subtract because everything is in dB(m)

 

[DodgeDart]

I think you just subtract the carrier level from the noise next to it. You subtract because everything is in dB(m)

 

[DodgeDart]

I checked satsignature.us and saw the signals you and other posters were talking about. I think you can have satsignature.us lock to a certain carrier by texting your parameters and provide you with carrier parameters as well as antenna size they used to lock. Will try to find that text number and share it.

 

[DRCars]

I think 323-452-6022 for "Carrier Monitoring" and 323-673-2478 for "MMS Plts" , but I don't have a cell phone, believe it or not, so I don't know how/if they work.

DRCars

 

[DRCars]

I think you just subtract the carrier level from the noise next to it. You subtract because everything is in dB(m)

I think you mean subtract the noise from the carrier level ? That would make sense.

I also was not accounting for the ~3 dB loss in EIRP for c-band G16 and G3 from here to the west coast (I think that's where they are). Also, I have been told that they are probably 4.5 m (14.6') dishes. I think the 3 dB EIRP difference was made up for by assuming ~3 dB less antenna size (10' instead of 14.6'). Compensating errors .

I think I've starting polishing an anvil, but I am getting a feel for what I will likely be able to lock on my dish vs. what the satsignature.us plots are showing.

Edit: Chated with satsignature.us "John" and he says the G16 and G3 dishes are 4.5 m and are in L.A., CA. All their dishes are at least 4.5 m and they have some in Phoenix, AZ also.

Thanks guys,

DRCars