Forum Jump: 
 2Likes
  • 1 Post By Rory B
  • 1 Post By holl_ands
 
Thread Tools
post #1 of 11 Old 01-07-2019, 03:11 PM - Thread Starter
Senior Member
 
tustinfarm's Avatar
 
Join Date: Mar 2004
Location: Athens, GA
Posts: 355
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 28
Maximum S/N Ratio for ATSC 1.0 transmitters?

This is purely a curiosity question, and may have already been answered...

I vaguely recall reading somewhere that the maximum S/N ratio an ATSC transmitter can achieve is somewhere around 35-40 dB. This is different than the noise margins in TVFool reports, which can be well in excess of that value, of course, but reflects the limitation of the transmitter hardware itself. What I am curious about is how the 35 dB S/N upper limit comes about at the transmitter - is it simply due to the number of bits used to drive the D/A converter in the output for a digital transmitter? Or alternatively, is the maximum S/N limited by the receiver's A/D converter?
tustinfarm is offline  
Sponsored Links
Advertisement
 
post #2 of 11 Old 01-07-2019, 04:27 PM
AVS Forum Special Member
 
Calaveras's Avatar
 
Join Date: Jan 2007
Location: Mountain Ranch
Posts: 5,794
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 1636 Post(s)
Liked: 348
The transmitter S/N and the TV Fool Noise Margin number are completely different things. TV Fool Noise Margin is telling you the signal strength you receive above the minimum needed to decode the signal (15.2 dB S/N). The TV transmitter S/N is just that. Some amount of noise is generated in all transmitters. Sorry, I'm not versed enough in ATSC transmitters to tell you all the sources of noise.

I have read that most ATSC transmitters have S/Ns in the mid 30's but I have read one report about a new solid state transmitter measuring in the mid 40's. That's incredibly good.

Rarely will you see S/N's at the TV above 30 dB because some tiny amount of multipath is present to lower the S/N a little bit.
Calaveras is online now  
post #3 of 11 Old 01-07-2019, 06:45 PM
Newbie
 
Join Date: Jan 2019
Location: Fair Oaks, California
Posts: 1
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 1
Quote:
Originally Posted by Calaveras View Post
The transmitter S/N and the TV Fool Noise Margin number are completely different things. TV Fool Noise Margin is telling you the signal strength you receive above the minimum needed to decode the signal (15.2 dB S/N). The TV transmitter S/N is just that. Some amount of noise is generated in all transmitters. Sorry, I'm not versed enough in ATSC transmitters to tell you all the sources of noise.

I have read that most ATSC transmitters have S/Ns in the mid 30's but I have read one report about a new solid state transmitter measuring in the mid 40's. That's incredibly good.

Rarely will you see S/N's at the TV above 30 dB because some tiny amount of multipath is present to lower the S/N a little bit.
The SNR of the transmitter is a measure of how close to perfect the signal is at the output of the transmitter.

The transmitter consists of what is called the exciter which generates a low power version of the ATSC modulated signal. The exciter drives power amplifiers that increase the signal level to whatever the output power needs to be. These amplifiers are analog devices and are thus not perfect in as much as they have non-linearity, frequency response, and group delay imperfections. The transmitter power output feeds a low pass filter to get rid of any harmonics that may be present. The low pass filter feeds what is called the mask filter which in turn produces the signal that goes to the broadcast antenna. The mask filter's job is to filter out out of channel RF energy near the actual broadcast channel. A sample of the Mask filter's output signal is fed back to the exciter which pre-distorts the signal it produces to cancel out the amplifier and filter imperfections.

The ultimate transmitter SNR value is a measure of how well the pre-distortion is cancelling out signal imperfections.

Rory
JHBrandt likes this.
Rory B is offline  
Sponsored Links
Advertisement
 
post #4 of 11 Old 01-09-2019, 12:11 PM
AVS Forum Special Member
 
rabbit73's Avatar
 
Join Date: Apr 2008
Location: S.E. VA
Posts: 3,050
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 775 Post(s)
Liked: 318
Quote:
Originally Posted by tustinfarm View Post
What I am curious about is how the 35 dB S/N upper limit comes about at the transmitter - is it simply due to the number of bits used to drive the D/A converter in the output for a digital transmitter?
The 8VSB DTV signal is actually an analog signal (amplitude modulation, single sideband suppressed carrier) that carries digital information.
Quote:
Or alternatively, is the maximum S/N limited by the receiver's A/D converter?
The demodulator in the receiver has to deal with the SNR of the transmitted signal, ambient noise, multipath reflections, and other impairments.

These are the factors that affect the SNR of the transmitted signal:
Tektronix
Technical Brief


Signal to Noise Relationships In 8-VSB

http://www.google.com/url?sa=t&rct=j...DNJjSx_qMikteV

Related paper:
Transmitter SNR for Maximum Coverage
by Oded Bendov

Abstract
The generally accepted notion that DTV coverage will not be materially improved if the transmitter's SNR were raised from 27 dB to 32 dB is shown to be true only for perfect links. In real world links a significant improvement in coverage would materialize.

I don't have a link right now

27 dB is the minimum required transmitted SNR by the FCC

If you can not measure it, you can not improve it.
Lord Kelvin, 1883
www.megalithia.com/elect/aerialsite/dttpoorman.html

Last edited by rabbit73; 01-09-2019 at 12:41 PM.
rabbit73 is offline  
post #5 of 11 Old 01-09-2019, 12:55 PM - Thread Starter
Senior Member
 
tustinfarm's Avatar
 
Join Date: Mar 2004
Location: Athens, GA
Posts: 355
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 28
Quote:
Originally Posted by rabbit73 View Post
The 8VSB DTV signal is actually an analog signal (amplitude modulation, single sideband suppressed carrier) that carries digital information.

The demodulator in the receiver has to deal with the SNR of the transmitted signal, ambient noise, multipath reflections, and other impairments.

These are the factors that affect the SNR of the transmitted signal:
Tektronix
Technical Brief


Signal to Noise Relationships In 8-VSB

http://www.google.com/url?sa=t&rct=j...DNJjSx_qMikteV

Related paper:
Transmitter SNR for Maximum Coverage
by Oded Bendov

Abstract
The generally accepted notion that DTV coverage will not be materially improved if the transmitter's SNR were raised from 27 dB to 32 dB is shown to be true only for perfect links. In real world links a significant improvement in coverage would materialize.

I don't have a link right now

27 dB is the minimum required transmitted SNR by the FCC
Thanks all for the excellent information - and I am looking forward to reading that technical paper. I guess it makes sense that the S/N ratio at the transmitter output is not nearly as big of a factor as it would be for analog radio/TV systems, since you don't depend on hearing/seeing a full "quieting" signal....just need to have enough S/N at the antenna output to "tease out" the data bits, and error correct as needed. Fascinating, and I just knew that avsforum would be the only place to get such good answers from the experts!
tustinfarm is offline  
post #6 of 11 Old 01-09-2019, 01:50 PM
AVS Forum Addicted Member
 
Trip in VA's Avatar
 
Join Date: Jun 2003
Location: Alexandria, VA, US
Posts: 16,200
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 1246 Post(s)
Liked: 420
Send a message via AIM to Trip in VA
I remember hearing somewhere, and I don't remember where anymore, that the theoretical maximum SNR is something like 43 dB. With a manual correction exciter, getting above 27 dB can be a challenge (I did manual correction on WDBJ in 2009 and was pleased to get it above 29 dB) but the auto-correcting exciters now easily hit 35 or 36 dB, and sometimes get up to 40 dB.


- Trip

N4MJC

Comments are my own and not that of the FCC (my employer) or anyone else.

RabbitEars

"Ignorance and prejudice and fear walk hand in hand..." - Rush "Witch Hunt"

Trip in VA is offline  
post #7 of 11 Old 01-09-2019, 03:01 PM
AVS Forum Special Member
 
Calaveras's Avatar
 
Join Date: Jan 2007
Location: Mountain Ranch
Posts: 5,794
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 1636 Post(s)
Liked: 348
Quote:
Originally Posted by rabbit73 View Post
Signal to Noise Relationships In 8-VSB

http://www.google.com/url?sa=t&rct=j...DNJjSx_qMikteV
I'll be amazed if anyone comes on here and says they completely understood that document.

My takeaway is that there are a number of factors that can degrade the transmitted SNR and they combine in the same way that multiple signals combine; i.e., one factor can dominate all the others or several can combine to degrade the transmitted SNR. Unless you're an ATSC transmitter design engineer, understanding all the contributing factors doesn't get you much. It's good enough to know that there are factors that sum up to degrade the SNR.


Quote:
Originally Posted by rabbit73 View Post
Related paper:
Transmitter SNR for Maximum Coverage
by Oded Bendov

Abstract
The generally accepted notion that DTV coverage will not be materially improved if the transmitter's SNR were raised from 27 dB to 32 dB is shown to be true only for perfect links. In real world links a significant improvement in coverage would materialize.
Plenty of links to it but the article is behind a paywall. Too bad because I'd like to know the rationale behind that statement.
Calaveras is online now  
post #8 of 11 Old 01-09-2019, 03:13 PM
AVS Forum Special Member
 
Calaveras's Avatar
 
Join Date: Jan 2007
Location: Mountain Ranch
Posts: 5,794
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 1636 Post(s)
Liked: 348
Quote:
Originally Posted by Trip in VA View Post
I remember hearing somewhere, and I don't remember where anymore, that the theoretical maximum SNR is something like 43 dB. With a manual correction exciter, getting above 27 dB can be a challenge (I did manual correction on WDBJ in 2009 and was pleased to get it above 29 dB) but the auto-correcting exciters now easily hit 35 or 36 dB, and sometimes get up to 40 dB.


- Trip

From the KMAX chief engineer on their new channel 21 transmitter:

"The Gates installer measured a S/N of 42.1 db. This is an absolutely awesome number. Credit the new solid state exciter and amplifiers."

http://01900888.com/forum/45-lo...l#post55023574
Calaveras is online now  
post #9 of 11 Old 01-09-2019, 05:11 PM
AVS Forum Special Member
 
rabbit73's Avatar
 
Join Date: Apr 2008
Location: S.E. VA
Posts: 3,050
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 775 Post(s)
Liked: 318
Quote:
Originally Posted by Calaveras View Post

Plenty of links to it but the article is behind a paywall. Too bad because I'd like to know the rationale behind that statement.
I have a fair copy
Attached Thumbnails
Click image for larger version

Name:	BendovTrSNR1_1.png
Views:	20
Size:	378.3 KB
ID:	2508426   Click image for larger version

Name:	BendovTrSNR2_1.png
Views:	16
Size:	125.2 KB
ID:	2508428   Click image for larger version

Name:	BendovTrSNRfig1.png
Views:	13
Size:	179.5 KB
ID:	2508532   Click image for larger version

Name:	BendovTrSNRfig2.png
Views:	15
Size:	162.7 KB
ID:	2508534  

If you can not measure it, you can not improve it.
Lord Kelvin, 1883
www.megalithia.com/elect/aerialsite/dttpoorman.html

Last edited by rabbit73; 01-09-2019 at 08:36 PM.
rabbit73 is offline  
post #10 of 11 Old 01-09-2019, 09:49 PM
AVS Forum Special Member
 
Calaveras's Avatar
 
Join Date: Jan 2007
Location: Mountain Ranch
Posts: 5,794
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 1636 Post(s)
Liked: 348
Quote:
Originally Posted by rabbit73 View Post
I have a fair copy
Thanks! I didn't know what he meant by "coverage area." I see what he means is that under certain conditions of noise and/or multipath the signal could be decoded if the TX SNR was substantially above the minimum of 27 dB where otherwise it wouldn't decode. I think for us in the real world that it would be very hard to recognize such a situation. I'd be willing to bet that there are few cases where low TX SNR is causing someone to not decode the station.

I've only recognized one example where a transmitter SNR was seriously degraded. Some years ago KTNC, a station that was almost always SNR 31 dB, suddenly got stuck at 25 dB. I read they had a partial transmitter failure. Sometime after they returned to 31 dB.
Calaveras is online now  
post #11 of 11 Old 01-10-2019, 05:00 AM
AVS Forum Special Member
 
holl_ands's Avatar
 
Join Date: Jun 2004
Location: San Diego, CA
Posts: 5,874
Mentioned: 28 Post(s)
Tagged: 1 Thread(s)
Quoted: 843 Post(s)
Liked: 412
Bear in mind that Dr. Obed Bendov was the Chief Engineer at Andrews Antennas, who has provided many (most?) of the current ATSC Broadcast Antennas....and hence was Primarily concerned with Calculating "Coverage" for their Antenna Systems. His paper cited above concluded that improving Xmtr Output SNR by 5 dB [from 27 dB to 32 dB] would result in a "significant" 1.6 dB improvement in SNR at the [collective] Receivers [and diminishing improvement beyond 32 dB]. Which in turn would (theoretically) increase the Total Number of Viewers within the "Coverage Pattern" by a "significant" amount....as calculated by the FCC's Statistical Longley-Rice Coverage Prediction Program [which includes Co-Channel Interference]....and is best understood by reading his Other Papers:
http://www.sonic.net/~aluther/paper1999.html
http://www.researchgate.net/publica...rmance_indices [May require Free Membership???]

So when Dr. Bendov says it's a "significant" improvement", he means in the context of improving the Statistical Viewer Numbers that pop out of the FCC's Coverage Prediction Program.

But for a particular Viewer, an improvement of just 1.6 dB would be Very Difficult to Notice....esp. IF Multipath is causing the signal to jump around by say 10 dB.

Dr. Bendov also tried to Quantify the amount of SNR "aka Noise Figure" degradation there would be to the Transmitted Signal due to SWR degradation on the Antenna Feedline [and Antenna due to snow/ice]....which I have cited many times wrt keeping Receive Antenna Design SWR below say 2.7. For cable lengths longer than the Symbol Time, reflections up and down (and up/down again) result in Inter-Symbol Interference [NOT NF Degradation]....one more good reason to use a Mast-Mounted PreAmp:
http://www.google.com/url?sa=t&rct=...hGHr2vvXTldLm4
Calaveras likes this.

Last edited by holl_ands; Today at 02:53 AM.
holl_ands is offline  
Sponsored Links
Advertisement
 
Reply HDTV Technical



Forum Jump: 

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

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off