[opendtv] "It can't be done"

At 1:47 PM -0800 12/26/06, Dale Kelly wrote:

IMO, Bert does indeed have a grasp of signal and receiver dynamics which
impact DTV reception and does also generally understand issues related to
SFNs. However, he nor you have ever dealt with the practical and financial
aspects of designing an RF system that is FCC compliant and which can
overcome such difficult coverage problems. However, if your maps had the
necessary information, he could likely give it a good start.
Obviously Bert has no interest in the challenge. His pre-conceived notions about the viability of SFNs versus continuing to waste valuable spectrum with the current high powered ATSC big stick approach are keeping him from using his imagination and engineering skills.
So be it. The challenge was a good natured attempt to get Bert thinking about alternatives to the failing U.S. OTA broadcast system. I would add that Dale also seems to be looking at all of this via a rear-view mirror. But I believe Dale may be more open to innovative new solutions, if he can get past the age old problem of "we tried that and it didn't work.
A brief aside. Several times during my career, I have confronted what i consider to be one of the major roadblocks to innovation and progress, as it relates to designing any kind of product or infrastructure. There is a built in (and highly biased) database in most engineers/companies/industries, which springs from real world experience. The response one hears is typically something like: we tried that years ago and it can't be done. Every industry builds up such a legacy mind set, which makes it difficult to solve problems in new and innovative ways.
I experienced this at the Grass Valley Group when I helped to lead the team that built the Model 100. The hardest task was to overcome all of the things that various people around the project said could not be done. Perhaps the best single example was the need to change the company colors - prior to the Model 100 every switcher was Grass Valley beige, with DS (diaper shit) orange silk screens. The most useful new feature was the ability to get the entire video path onto a single board, so that video alignment was simplified, and we could do a switcher alignment via a board exchange.
Helping companies to develop the new computer-based tools that now dominate video production was a much more difficult undertaking. Here we were fighting the way things were always done, and an analog mindset, that was not open to the idea that we are just processing rasters of samples, not analog waveforms complete with sync, blanking and subcarrier. The shift to digital technology re-wrote the rulebook. Many things that could not be done, were quite easy when we started manipulating pixels.
The same is true for the biggest laggards of them all, TV broadcasters. They still don't understand that they are in the business of managing data. They haven't a clue that they have all the technology they need to compete effectively, developing relationships with viewers rather than audiences. And the worst aspect of this is what Dale identified in his post. The need to keep an outdated government bureaucracy happy, rather than working to get rid of it and the excess baggage that keeps broadcasters from using the spectrum to develop a competitive OTA service.
I fully understand where Dale is coming from. I enjoyed great success in my career, acting as the expediter to help companies overcome some of these mental hurdles. That is, until I got involved in the U.S. DTV process. For the past 14 years I have been amazed at the power of key industry players to block and delay, and to throw logic and good engineering principles to the wind. Eventually I let go, after beating my head into a wall too many times. At least now I have the satisfaction of beating a nail or two, and actually building things that work.
I ask Dale, and others to simply consider alternatives, without the legacy bias that say you can't do that. When OTA broadcasting dies someone will come behind and do it. Actually, companies are already starting to do it. But they are highly constrained, due to the lack of spectrum available to build new digital broadcast infrastructure that can provide a viable option to the broadcast and multichannel oligopoly that has us stuck in the past today.
So please consider the possibilities, without the constraints of legacy thinking.
As a starting point, I did a bit of research yesterday into some basic OFDM design constraints. The best resource I found is a pdf White paper published by MediaFlo, the Qualcom subsidiary that is building out an OFDM-based mobile TV infrastructure in recovered 700Mhz spectrum.Here is the link: 

http://www.qualcomm.com/mediaflo/news/pdf/MediaFLO_SFN_Whitepaper.pdf
The paper is highly informative, and includes several maps detailing how the network will be built in several areas of the country. One interesting take away is that it will take only 30 transmission sites to completely cover the NE corridor, and this includes both national channels and local channels in several markets. You can zoom in on the maps and see exactly how they plan to do it.
Based on this document, it appears that they are using a cell spacing of 27.7 km, or about 17.7 miles. Keep in mind that this is designed to provide a base and enhancement layer for fixed, portable and mobile reception in a six MHz channel. I consider this to be the worst case scenario - but it serves as an excellent example of what broadcasters COULD do. Given this spacing, it appears that Dale's market could be served with 5-6 transmission sites, probably two a bit closer together to provide optimal service in the more populous Santa Barbara area. My guess is that this would leave only a few isolated pockets where On-channel repeaters would be needed to fill in terrain blocked areas.
This should be enough to get things moving on a useful discussion. I'll add more ideas as we develop this thread. 





----------------------------------------------------------------------
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: