8-VSB - A Few Good Words: An Interview With Glenn Reitmeier, Sarnoff Research Center

controversy has broken out over the utility of the single carrier 8-VSB transmission system. Sinclair Broadcast Group has petitioned the FCC to include the COFDM modulation in the US standard. COFDM is the multi-carrier system favored in Europe and elsewhere. Proponents of 8-VSB say that despite poor first generation receivers their system does meet all the broadcast requirements. Billions of dollars have been spent for its development but little has been spent by the US consumer to acquire it.

In contrast, the UK's ONdigital is a COFDM pay DTV terrestrial service provider that secured 552,000 subscribers in one year. This was mostly to meet the demand for an expansion of channels rather than a transition to HDTV, The Financial Times of January 11, 2000, reported a surge in sales of widescreen digital television for Christmas, however, with many retailing for over $1000. CEA reported just over 100,000 DTV-ready sets have been sold in the USA. No one knows how many ATSC decoders were also sold, though the common industry guesstimate is under 25,000 .

Without a clear way forward the broadcasters in the U.S.A. could end up being less relevant in the digital revolution then they once had hoped they would be. With Time-Warner being absorbed by AOL, the spotlight remains on the wired and connected world. Some think the mobile wireless connected world is next, and is as significant as is the wired world, making then broadcasting an enriched sector if they participate beyond the their traditionally stationary audience. This view of broadcasting using its core competency in terrestrial transmission services for entering a broader market with a new set of applications has itself become the centerpeice of the controversy over the modulation system. Some in broadcasting say they are better off focusing upon the programming--certainly exemplified with the tremendous success of "Who Wants To Be A Millionaire" than they are in banking on a business notion (enabled by digital) that doesn't even exist yet.

But the Sinclair challenge, which says COFDM is superior to 8-VSB for all of today's business models in broadcasting, has created a near-paralysis in the consumer and professional manufacturing circles. "Can we afford to go forward if there is a derailment ahead?" That question haunts manufacturers with whom have we spoken at the CES last January 6-9. The greatest fear is that the FCC will open the standard with first a fresh call for Comments to the Sinclair Petition. Many worry that such a move would lead to an inevitable opening of a door that cannot hope to be shut for at least three years. Such a pause would halt the entry of broadcasting into the digital revolution--some say permanently. "That is the objective of Sinclair." is the denfense line from CEA. Sinclair says that a fast track through the FCC process is possible if just everyone one comes to their senses at the same time. Lynn Claudy at the NAB concedes that a year is the shortest possible time for evaluating and testing which COFDM system should be tapped--the DVB-T version modified to 6MHz from Europe, or the more recent entry from Japan.

Larger manufacturers, such as Panasonic, have invested heavily in 8-VSB but also make COFDM products for other markets. Panasonic said to HDTV News Online at the CES that a market-led change to COFDM at this early stage would not be overly consequential. While the ATSC in Washington, D.C. has a very clear responsibility to stick with 8-VSB, many members we approached appear less fixed. Zenith has the most too gain or loose, but they too are not without the understanding that a robust consumer market for DTV products is better than one strangled to death by public and professional confusions. Zenith's new financial partner--Lucky Goldstar--is also a COFDM product supplier.

Zenith had been asked by the ATSC to show definitively that 8-VSB works to meet the stated requirements. Sinclair can be given credit for showing that the 1st generation receivers were bad, but, according to Zenith and the ATSC's leadership, they have failed completely to make the case that a bad first generation of receivers equals a bad transmission system. Zenith has answered with demonstrations in Las Vegas and in a wintery two-day session in Washington last week using improved receivers. What they have shown indicates that the 8-VSB works well--perhaps works even better than does the international challenger--COFDM.

Zenith demonstrated this first in room 464 of the Hilton Hotel to invited guests, the press, and the FCC. The demonstration, complete with documenation handouts, offered a clear indication that second generation 8-VSB receivers are substantially improved over the first generation. The superiority of 8-VSB in reach, power usage, and impulse noise tolerance was shown in a controlled closed loop system rather than real world (impossible to set u in Vegas). The apparently lesser performance from the challenger left most visitors believing again in 8-VSB.

Skeptics and COFDM advocates reacted, however, to the demonstrations by stepping-up their attacks on the 8-VSB standard. They claimed in E-mail after E-mail that these "tests" were faulty, and highly deceptive (rigged). While those advocates were not convinced the new Zenith tests managed to calm some of the anxious broadcasters who were genuinely fearful of the worst, i.e., that ATSC had made a terrible mistake in selecting 8-VSB. Others disagreed, and sent to the FCC encouragement to seek comments. Field reports from those who have already installed transmitters for HDTV are very mixed with few having the same complaints as Sinclair. Motives for Sinclair's actions have been questioned and requestioned by those not experiencing the same grief.

The more zealot supporters of COFDM think that a world standard may at last come with the sweeping in of COFDM in the USA. That, they claim, will lower costs for the whole of the global television market and, thus, speed the digital transition around the world. Others disagree, saying that economies of scale do not have an infinite curve, and the US market is large enough to arrive at any economies needed for achieving the lowest-cost in manufacturing.

The following interview is with one of the bright stars of the HDTV movement--Glenn Reitmeier. The interview may shed some light on 8-VSB for you, and show where it stands in the Pantheon of the digital transmission standards.

Before you make any "final decisions" (hello Regis) on this serious controversy I urge you to read this interview. Glenn has done as much research and development on the subject as anyone. His company--David Sarnoff Research Center--has played a central role in television and communications electronics since its founding in the 1930s.

HDTV NEWS: What is the American case for staying the course with 8-VSB modulation?

Glenn: From the very beginning we have seen these two different modulation schemes. In theory we may say they have both improved a little from their theoretical limits. Given what we know now would 8-VSB be chosen again if the process were to be repeated? Everyone was certainly aware of COFDM technology during the Grand Alliance process. Even in the Advisory Committee OFDM was studied and looked at extensively by a committee of experts headed up by John Henderson, who did a very thorough study of OFDM.

All of us in the Grand Alliance were certainly aware of the work going on in Europe and the attributes of OFDM. Many of us in different places in our corporations and partners had experience with OFDM solutions. If you look at the selection process there were a dozen different criteria that governed the selection for the right modulation technique to use.

One of the primary selection criteria was the replication of the current

• System threshold carrier to noise -- one of the primary considerations.
  
• Interference into existing service --another primary consideration
  
• The data rate capability was still another
  
• Robustness to impulse noise was another important characteristic
  
• Robustness to adjacent channel interference from both analog and digital systems (since we are tightly packing the spectrum) was also very important one. The requirements there on tuner performance and costs were a characteristic
  
• Robustness to co-channel interference -- both analog and digital -- were additional criteria that all had to be met with satisfactory performance.
  
  

In the very first competitive days in 1990 and 1991 our proposal of Advanced Digital HDTV as well as the Zenith proposal (Digital Spectrum Compatible HDTV) had sort of a two tiered scaleable system. We (Advanced Digital HDTV) had a fixed bit rate and two carriers. Zenith switched between 2 and 4 VSB modulation with a variable data rate.

One of the things that were brought home clearly with the broadcasters' evaluation of competiting systems is that they were not willing to pay any penalty -- ANY penalty for graceful degradation or tiered services. Actually, the uniformity of a digital system throughout the entire coverage area was the key thing they were looking for as that was important to "advertising." As we formed the Grand Alliance system the two-tiered notion--very clever and innovative solutions--was actually abandoned as the Grand Alliance moved forward. We still settled on the pioneering aspects of Advanced Digital HDTV (a little plug here for my company). In reading the proponent documents you would easily see that layered architecture and MPEG packetization and data broadcasting, as well as multiple formats, was the legacy of Sarnoff/AD-HDTV. We are tremendously pleased that those aspects were so widely recognized and valued by the industry, and became such an important part of the Grand Alliance system.

HDTV NEWS:: This current cry for that....why do you think it is there?

I don't want to speculate on motives. I think there are some different assumptions that get made in terms of how people use technology in the computing industry vs the television industry. In general, if you look at the computing industry, there is a product put out with the expectation that it gets better next year, and then again better the next year, etc. So, the capability (of computer products) evolves over a period of time.

With consumer electronics it is the opposite way. In consumer electronics the standard itself last for a long time. Why? Because consumers demand longevity in that marketplace. What happens in consumer electronics is that the product gets cheaper.

Remember the $1500 VCRs and the $2000 CD players? What tends to happen in the CE industry is that the technology that advances is applied to reducing costs. You have a very different product evolution velocity in the computer industry vs the traditional consumer electronics industry.

Let me use the example of the decoders and receiver solutions that have resulted from the joint efforts of Motorola and Sarnoff. As you know we collaborated to design a two chip chip set. Part of that is an innovative decoder, which uses a reduced memory decoding approach to break the cost barriers of high-definition decoding. That is a perfect example of new inventions and new algorithms, and new decoding techniques that have been invested in and made very specifically to reduce costs.

HDTV NEWS:: The internet came in full force in the middle of all of this. Some people say that, "well this standard is very much an analog legacy and doesn't represent the future."

I am sorry I am laughing. In the beginning when we went digital we were accused of being nuts by the analog guys. Now the computer folks call me an analog guy.

HDTV NEWS:: Here we are in a world where the IP is a big deal. Is the ATSC standard up to snuff there?

It is one of the most powerful thing we could give to the broadcasters to enable them to compete in the digital world. I say this for the following reasons. This goes back to the tremendous flexibility in the ATSC standard to combine standard definition and to have high-definition services in other times of the day. We have data broadcasting because of the packetization. Let's not get hung up about what the modulation is. Let's think of the services that we are going to offer. Let's think of how this wireless broadcasting complements the world of streaming media in the Internet. You know the Internet is not going to go away. It is going to grow. The companies that are going to be successful are going to be those who figure out how to take their content and corporate assets and apply them in both places.

The technologies involved are not competitive, they are tremendously complimentary. When you think about the niche and narrowcasting capabilities and the ability to individualize content that happens on the Internet...when you think about the economies of scale and the power of mass communications that happens in broadcasting. We know there have to be very powerful ways to combine those two things and have the best of both. What it takes is having an open mind and looking for the business opportunities and not getting hung up on some of the ferver that often colors our opinion about whether one is better than the other. Broadcasting is a different thing than the Internet.

HDTV NEWS: Whatever broadcasting can do to compliment the Internet is incorporated in the ATSC standard?

It sure is. That has a lot to do with a lot of those things we pioneered many years ago.

HDTV NEWS: In waking up in the middle of the night and thinking it all over, what would you like to have incorporated in the standard today that is not in it now? Or, do you have complete satisfaction in it?

You can always think of those little things where you think you could have done a little better. You think, gee, this number would have been a little better of this better. But those are second or third order things. I think what we have in the ATSC standard is such a tremendously powerful and flexible standard that we have positioned the broadcast industry to compete in the digital world. Frankly, some will compete and survive and adopt new business models and figure out to compliment the web and streaming media with their traditional broadcasting bits and others won't get it and go by the wayside.

Regardless of any digital system you have to remember that inherent--nothing to do with ATSC or any other systems-we could throw it all away and start with DVB tomorrow - but inherently the Internet is a two way interactive medium. Broadcasting is a one way thing. Again, let's figure out how to combine those things, but let us not get confused and think that the standard we have in ATSC is not Internet friendly. It is extremely Internet friendly. It is a digital system. It is a layered system, its a flexible systems full of headers and descriptors and flexibility to let broadcasters do very powerful things with their data streams. It is eminently recordable. You can broadcast it to PCs. You can broadcast it to the family rooms. Think about the broadcaster's assets---20 million bits per second going by every home in America.

Thank you.