[opendtv] Re: Interview: ARE MANY TRANSMITTERS BETTER THAN ONE?

At 11:12 AM -0400 9/18/07, Manfredi, Albert E wrote:


I'm not sure where you're going with this. It sounds like in terms of
the SFN argument alone, it's six of one and half a dozen of the other.
The cost of deploying SFNs is a wash, if you assume synchronization to
be required in both cases.


Where I am going with this in ~1994.
The advisory committee, the FCC and the ATSC were all focused on a DTV transmission system based on the "Big Stick" model employed for NTSC. This was the easiest approach for the FCC to deal with, and the cheapest route for broadcasters. There was NO INTEREST in SFNs in 1994 when COFDM was being evaluated, however, this was being promoted by the DVB folks as a major advantage of their system. So an economic argument was created to mitigate this advantage...
"It's an interesting feature, but economically prohibitive to deploy because of the need for telco land lines to feed the transmitters."
Somehow, the economic argument used in the 1994 time-frame turned out to be completely wrong. So now, 13 years later the ATSC has figured out how to make SFNs work and it is economically viable and politically correct to talk about SFNs. 



Are you saying that he should have included the example of passively
synchronized, small area SFNs, with either two towers or one big stick
and a few much smaller ones?
No. I am saying that it was already obvious that there would be multiple technologies available to solve the synchronization problem in 1994, thus the economic argument against DVB was questionable at best. To be fair, there were other issues as well. For example, there were purported advantages in using an 8K carrier system versus a 2K system, but it took a decade for the 8k version to become economically practical.
When you are designing a standard that is to be used for decades, you need to have a decent grasp of where the enabling technologies are going to come from. This is true for DVB, ATSC, et al.
You have been a staunch supporter of the big stick approach, predicting that the equalization problems would be solved. It took more than a decade, but it turns out that you were right.
I could also add that we provided input to the Advisory committee about the trends in enabling technologies, which were largely ignored. These predictions also turned out to be true. So it all boils down to how technology is misused to gore someone else's Oxen. 




 This effectively forced COFDM to be compared with ATSC using
 only single high powered transmitters. In this comparison, many
 of the advantages of the COFDM approach are lost.


Which is exactly the right analysis. The passively synchronized, small
area SFNs, for which COFDM would have the clear advantage, require
dialing down spectral efficiency, and are not very effective in covering
huge OTA markets. As we have in the US. So it is very appropriate, IMO,
to consider whether the benefits outweight the liabilities.
Right analysis, but your conclusion is wrong. SFNs DO NOT "dial down" spectral efficiency; they SIGNIFICANTLY enhance it.
As we have discussed MANY times before, there are some areas in the U.S. where markets are isolated and the "Big Stick" approach is the best solution. There are ALSO many markets in the U.S. that are closely co-located, where the BIG Stick approach causes severe market-into-market interference, forcing large chunks of spectrum to lie fallow to protect the channels that are being used. This reduces spectral efficiency by nearly half.
As you are no doubt aware, attempts to use these white spaces have been challenged at every turn to protect a distribution infrastructure that is used by less than 15% of the population. If this spectrum were used properly, broadcasters could offer at least 20 multiplexes in every market, which in turn, would allow them to compete effectively with cable and DBS. 

But this is not the U.S. way...
Here, the spectrum is used to provide the political cover so that broadcasters can gain additional compensation from competitors for content that is advertiser supported and delivered in the free and clear via their Big Sticks. 


Now that synchronization is more practical, using GPS rather than
perfectly measured transmission lines, more useful SFNs can be installed
in both systems. Which makes the two more similar, and further pushes
into the past all the silly debates about COFDM vs 8-VSB.


Not really.
Funny how we could not disenfranchise first generation receivers, but now it is perfectly acceptable to develop channel compatible enhancements to the ATSC standard that will require new receivers for the new services. If we had gone with DVB, we would have all of the needed modes of operation from the outset, and using these modes would not disenfranchise exsiting receivers. 

Regards
Craig


----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:
- Using the UNSUBSCRIBE command in your user configuration settings at FreeLists.org 
- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word 
unsubscribe in the subject line.

Other related posts: