The official final DTV Table Of Allotments/channel change thread - Page 243

How is 4 = 3 * 2? Increasing the bits per symbol from 3 to 4 raises the bitrate by 4/3 or 1.333. That gets us to 19.39 * 1.333 = 25.853 Mbps. Then by removing the 2/3 FEC, the bitrate is increased by 3/2 or 1.5. 25.853 * 1.5 = 38.78 Mbps. Another way to look at it is 4/3 * 3/2 = 2.

The 16-VSB numbers can be found here.

http://www.atsc.org/cms/standards/a_54a_with_corr_1.pdf

See Table 8.1

Ron

For QAM-64, the required S/N is 22 to 24 dB and for QAM-256, it's about 28 to 30 dB. So at least 7 dB for QAM-64 and 13 dB for QAM-256.

But as Calaveras points out, QAM and PSK aren't very well suited to OTA transmission. A simplified explanation is that QAM and PSK both use phase modulation to carry the bits. But multipath is essentially a phase shifted version of the main signal. Therefore, with QAM and PSK, multipath directly affects the way the bits are being received. No bueno.

For QAM and PSK to work, you need to take care of the multipath (more technically called inter symbol interference or ISI) somehow. That's were multi-carrier systems like COFDM come into play.

Ron

You can have different FEC coding rates. The most often used rates are 1/2, 2/3, 3/4, 5/6 and 7/8.

However, there's no free lunch. Any increase in bitrate in the same 6 MHz channel will require more signal to noise ratio. IMHO, it will be impossible to convince broadcasters to either increase power to maintain their coverage area or lose coverage area just to increase the delivered bitrate.

Ron

When I mentioned QAM I meant broadcasting OTA using the system that is currently used on U.S. cable, because on RF channels 2-13 it would be compatible with most current tv sets. I was not referring to any system currently used in Europe or elsewhere. I am vehemantly opposed to anything that means another nonbackwardcompatible transition anytime within the next 30 years minimum!

Not every 8VSB set has QAM in it, but every QAM set does 8VSB. Remember 8VSB IS required for digital sets. QAM is not. I have two sets that have analog and 8VSB but no QAM. 8VSB and QAM are modulation types and are not compatible to each other. It isn't as simple as swapping one for the other. There are lots of technical issues that go along with a swap like that.

No CECB that I'm aware handles QAM. I don't think that was even permitted.

I got it. I was being dense. I was confusing the number of bits with the values they represent.

Chuck

Correct. FCC rules did not permit anything but 8VSB for the CECB.

Practically.

Technically, many boxes using chips (tuner/demod) what are support QAM. Updating FW would open the feature.

I don't know how many CECB's have firmware upgrade features. Mine doesn't.

We're talking about boxes mass marketed to J6P. "Practically" is all that counts; "technically" is irrelevant. (And resistance is futile... )

Okay, so somehow the FCC granted WLJC 185 kW on channel 7. Now I'm confused.

- Trip

They want to make sure the wisdom of Jesus reaches Mars.

Largest station exception, according to their technical exhibit. They claim that it takes 185 kW to match the coverage of WYMT Hazard, and apparently, the FCC agrees. Unless the FCC changes its mind in the future and rescinds the grant. Conveniently for WLJC-TV, it's just enough to put the 43 DBU contour completely over Lexington.

But it exceeds the 160 kW limit. They shot down KCWX's request to boost to 82 kW and WBNS's request to boost to 1700 kW on those grounds.

- Trip

My guess (and it's only that) is that the FCC is now looking with favor on VHF. This may set a precedent in order to encourage stations to reconsider VHF over UHF.

Lookee! VHF is the hot new band! We'll give you BIG GIANT power if you give up your UHF assignment for the telecoms.

KCWX didn't claim largest station exception in their application. It's also worth noting that the FCC database shows that application as still active, but ungranted. What the FCC dismissed was KCWX's STA request to begin operations at that power level prior to FCC approval of the Form 301 application, which would have to clear Mexican coordination first. I have a sense, however, KCWX has abandoned its own application, as they have now applied for a replacement DTV translator in the Austin area, to operate on RF channel 8.

While WBNS-TV did claim the largest station exception, their application admitted to 0.66% interference to co-channel WUPX Morehead KY, when calculated using Longley-Rice methodology. That's above the 0.5% permissible limit. WBNS vowed to either secure an agreement from WUPX to accept the interference, or to amend their application to lower interference to WUPX. Again, according to the FCC database, they did neither. Nor do we know if their application ever cleared Canadian coordination; I'm sure Columbus lies within the Canadian coordination zone. In any case, just based on the impermissible interference to WUPX, the FCC would dismiss that application. It would have been helpful if the FCC had published the communication with WBNS concerning this application.

If Low VHF could have 185,000 watts would it be suitable to use then? What is the putential coverage area for a low VHF station with that much power?

Wow - talk about a DXer's delight!!

I just ran a Longley-Rice estimate of how KCWX (RF-5) would look with 185 kW at its HAAT of 1351':

http://www.onairusa.com/185kw/index.html

Compare it with the present coverage with 23.7 kW at that height:

http://www.rabbitears.info/contour.p...-98.6097222222

Ok if I'm reading that correctly the main differnce between 8-VSB and 16-VSB is that 8-VSB has more coding to prevent analog interference. And that extra coding is why its capped at 19 Mbps. Seeing that as of Sept 1 2015 that will not longer be an issue since all analogs will have to shut down by then then it seem that that extra coding will no longer be necessary. That would allow OTA broadcasters to use 16-VSB which means 38 Mbps. Or am I misreading that.

Possibly -
BUT -
you still need to get reception devices out to the viewers that can accept 16VSB.

I think analog interference is the least of the issues. The error correcting also helps with multipath and noise significantly improving the useable SNR. From what I can tell, the minimum SNR for 8VSB with no error correcting is about 24 dB. Someone can correct me if I'm wrong about that. See page 9 in this PDF.

http://www.digitalproceiver.com/mods...ing%208vsb.pdf

It looks like the dots just stay within the lines at 24 dB SNR. Dots that cross the lines are errors that need to be corrected.

In Table 8.1 in the link that Ron posted above, it shows that the 8VSB symbol rate is 10.76 Msymbols/S or 32.28 Mbps. The Data Payload Rate is 19.39 Mbps. If you eliminate the Trellis coding, the Data Payload Rate should go up to 29.09 Mbps, leaving 3.19 Mbps of other overhead.

In the same table it shows 16VSB Data Payload Rate of 38.78 Mbps with no Trellis coding. If you were to add 3/4 coding for OTA use then the Payload Data Rate would go down to 29.09 Mbps.

From what I've seen, it looks like Trellis coding uses 1 bit/symbol, hence 2/3 for 8VSB and I assume 3/4 for 16VSB. There must be some SNR penalty for 3/4 compared to 2/3. 16VSB should require 3 dB higher SNR for the same performance so the transmitters need to increase power by 2X. I'm thinking that the minimum SNR for 16VSB would be on the order of 19-20 dB.

Of course this is just an exercise in "What could be done to improve efficiency?" since it would require another transition which should not happen for a long time.

Chuck

If the FCC takes away 31-51 and stations are focred to share channels it'll have to happen sooner than later.

Besides the FCC can madate that all TVs made after Jan 1st 2013 have 16VSB tuners in them. Have the transition take place in Jan 2018. That's 5 years.

If it was possible to have another transition, then using 16-VSB would be a pretty silly way to go. It's a 90's technology that never saw the light of day.

There have been many advances in error coding and modulation since 1995. Many of these new techniques (like Low Density Parity Check and Bose-Chaudhuri-Hocquenghem multiple error correction binary block codes) are used in the DVB-T2 specification.

From this document, it would seem that a 27 Mbps service could be obtained in a 6 MHz channel at around 16 to 17 dB signal to noise ratio. Or even more trickier, a 20 Mbps service in a 4.5 MHz channel. Then you could have the same number of TV stations delivering an ATSC similar service in 25% less RF bandwidth.

http://www.dvb.org/technology/standa..._Imp_Guide.pdf

Ron

No matter what KCWX does with their ERP at that "remote" TL, it will still deliver the best signal to more cattle and mesquite than to OTA viewers. The long tale of one of perhaps last full-power US commercial Ch 2 NTSCs to be licensed
http://www.qsl.net/wa5iyx/fred-2.htm
That said, my viewing them as a 23-kw DT-5 has been far superior to their 100-kw analog 2 with all the interference it would suffer (lightning, appliances, co-channel from Houston tropo and the rest of country's on Es) at 46 miles away.

Another transition in 2018?! Are you freaking joking?! HELL NO!!! The public should riot in the streets if that happens. Even 2028 or 2038 are way too soon for another nonbackwardcompatible transition. Something of the magnitude of what we went through last decade should be a once in a lifetime thing!

Hi,

Transition in 2018, I will not be around then perhaps but I expect the OTA to Cable / Satellite / Internet transition to be almost completed by then.

What might remain of OTA DTV at that point, look at Channel 1 (ONE). That service with 12 video channels will be the pattern for sure. (First in list)


Even the equipment that KAXT uses is pretty.

http://www.avsforum.com/avs-vb/showt...7#post20406177

HDTV, you think that it will be forever free, ROFLOL.

Remember, we had a "DTV" transition, NOT a "HDTV" transition.

SHF

As it turns out, the ATSC 3.0 folks have already considered using LDPC and BCH error correction codes with 8-VSB.

http://atsc.org/cms/pdf/pt2/02-PT-2-...-Yiyan-V10.pdf

On page 40, it looks like a 24 Mbps service is possible at 15 dB S/N ratio.

Ron

I'm doing my best to understand this. In another link you posted there was a table that said that the ATSC symbol rate is 10.76 Msymbols/S or 32.28 Mbps. Looking at the graph on page 40, that would correspond to a Shannon limit of about 20 dB C/N. This is what I read somewhere else as the essentially error-free C/N requirement with no error correcting.

If I understand this correctly, the Data Rate label is actually the Data Payload Rate, a term used in your other link. Data Payload Rate meaning the actual useable data which is what's left over after all overhead is accounted for; i.e., error correction and whatever else there is.

So it appears that the newer error correction schemes use fewer bits to achieve the same level of error correction as the current ATSC system does thus resulting in a greater Data Payload Rate.

Further it appears that all this error correction has been implemented to reduce the useable C/N from around 20 dB to 15 dB.

On top of greater error correction efficiency with these newer methods, there's also greater data compression efficiency that can be achieved by moving from MPEG2 to MPEG4.

Do I have this correct or have I missed something?

At the end of the presentation it is stated that OTA can never compete with satellite or cable in terms of bandwidth efficiency. I assume this is because the latter two don't need any error correcting. But it is also stated that OTA has its own advantages.

Chuck