How to keep DTV from going astray

by Gary Sgrignoli, TV Newsday

Gary Sgrignoli may know as much about DTV transmission -- its strengths, its weaknesses, its quirks --  as any person alive.

For nearly 30 years, he worked at Zenith Electronics (now LG Electronics) where he was deeply involved in the development and testing of the 8-VSB transmission scheme, which was ultimately adopted as the ATSC standard.

Now, a partner in the Meintel, Sgrignoli & Wallace consulting firm, Sgrignoli has emerged as a DTV transmission guru, keeping a sharp watch for problems as they arise and figuring out how to fix them.

Over the past 10 years, he has shared his wisdom with broadcast engineers through more than 85 all-day seminars on 8-VSB.

In this interview with TVNewsday, Sgrignoli talks about what broadcasters -- and viewers -- need to do to make sure viewers can find the DTV signals and keep the pictures sharp and steady.

An edited transcript:

I have been hearing lately that the DTV signal is not everything that it should be, that the reception is poor even in areas where the field strength meters are saying it should be strong. What's going on?

I have not heard a great deal of clamoring regarding this since a third of the TV stations turned off analog by Feb 17. Now we've quietly heard that there are certain areas where they're questioning whether high-VHF signals are being received properly. My concern is that there may not be enough radiated transmitter power in some cases, and that people are using rabbit ears that are not very good in other cases. Also, getting the signal indoors sometimes is a problem with high-VHF, if the only decent path to the inside is through the windows, which typically are not large enough for these high-VHF signals to pass through easily. Of course, it's worse for low-VHF signals, but there are only roughly 40 stations in the whole country that are going to end up in that band.

There are plenty of areas where people are pleased with the coverage. Unfortunately, you don't hear much from them because when people are happy, they don't bother to call in and write in and so forth. So, obviously, I believe it's slighted towards the negative side.

That said, there are three areas where problems can creep in -- the transmit side, the propagation side, and then the receive side. The broadcasters are working hard on the transmit system. They're learning, they're gaining more test equipment, and they're learning how to use it and so forth, and so we've seen some really good things there.

The propagation side: Hey, you can't change the terrain or the buildings that are there. Hopefully, the broadcasters have optimized where they've placed their transmit antennas and the height of the antennas and their patterns and all that.

That leaves the receive side, which I believe is the biggest concern. That's where the industry is trying to catch up now. They have to educate the viewers on what kind of antennas they need and how to use them.

One of the problems is with "managed expectations." Viewers were told to just get a DTV set or a converter box, hook up some kind of antenna either indoors or outdoors, and assured that it would work. While that can work in some cases, it does not work all of the time. That is a problem that can only be solved with more viewer education.

Let's talk about the broadcast side first. What's going on or not going on at the transmitter site that may be causing problems? And how do you rectify it?

The broadcasters are pretty much sending out a decent VSB signal in terms of signal quality. They've learned to do linear and nonlinear pre-correction, which means to make the signal as pristine as possible before it's sent up the transmission line to the antenna. They've worked out some of the issues with average power measurement, FCC emission mask compliance, phase noise and things like that, but there may still be some issues with clock jitter.

Clock jitter?

In the early days, we saw that sometimes the transport clock going into the exciter was jittering enough that it caused the symbol clock of the 8-VSB to also jitter to the extent that it affected the receivers out in the field. This jitter can come in anywhere -- from the studio to the transmitter. So that's just one thing to be careful about. The other thing is the actual symbol clock frequency. In the early days, the symbol clock frequency was sometimes not within the recommended 2.8 parts per million tolerance. Test equipment now is in place to at least inform them about the frequency tolerance of the symbol clock.

Lip syncing has apparently been a chronic problem. What's being done to correct that?

It's something that they're working on within the broadcast industry. It has nothing to do with transmission, of course. It's a product of the video and audio processing. Because the video requires far more processing than the audio, it delays it longer than the audio. What they do is try to delay the sound to match it, but every time it goes through a different set of processers, they've got to be careful to do that again, and that's what they're working on now. I believe that they will solve this problem in due time.

Is the transmitter less forgiving in the digital world than it is in the analog world? In other words, do you have to take better care in calibrating and maintaining the transmitter than you did in analog?

I guess that's fair to say. But in some ways things are a little bit easier now because digital and modern technology have caught up with or advanced to the point where it's not so hard to maintain this.

On the other hand, the transmission system for 8-VSB is far more efficient than the old analog system. You're sending five times the amount of video information in digital and using less radiated power to do so. Since you're using the channel much more efficiently, you have to be more careful about the creation of the 8-VSB signal and its transmission. But the broadcasters, I think, have that down fairly well. The equipment is there to do what is necessary to create this pristine signal.

That assumes the broadcasters are paying attention and doing their job.

Right. That's not so much a problem. As a general rule, they've come a long way. Some of them have been doing this for ten years. The reason it's being focused on now is because we've had a third of the stations turn off their analog signal and station management realizes that digital is all they've got now. There's nothing else to fall back on. So, obviously, they want viewers to have a good experience with their free, over-the-air signal.

Has the test and measurement equipment caught up with the technology yet?

Yeah, essentially it's out there. However, I can't say with certainty that every station has all the equipment that they really should have because it's a money thing. A lot of times test equipment is on the lower end of the list of things to buy, especially in a bad economy like we have right now.

What kind of investment do you need to make sure you have the proper test and measurement gear?

That's a hard one since it can vary greatly. It depends on the accuracy of the equipment. For instance, you can get a reasonably low-cost spectrum analyzer -- low cost is in the order of $10,000-$15,000 -- but you might have to pay $40,000 or $50,000 for a VSB analyzer that's of instrument grade. So we're talking $50,000-$60,000 just for those two pieces of equipment. Of course, commercial grade test equipment is also available for less money, but with less accuracy. Then you need a power meter, but you probably have that from the analog days. It is also a good idea to buy an MPEG transport stream analyzer so that you can analyze the actual data packets. So it depends on how far back in the chain you want to go.

Let's talk environment issues -- buildings, terrain and other forms of interference. What are you learning about how they affect the signal?

What we're finding out is that interference from multiple different signals is still an unknown at this point. There had been the assumption in the laboratory and in the FCC planning that there would be one primary interferer. We think we need to go further and start looking at multiple interferers. Obviously, the focus is going to be on digital-into-digital interference as analog turns off.

We've also recently had some problems with FM radio stations. The second harmonic of those stations fall into the high-VHF television band and are causing problems and interference with broadband amplifiers. The broadband amplifiers could be inside the antenna or they could be external to the antenna on the mast or they could be distribution amplifiers in the home. So, if you're near a strong FM station and there's no FM trap in that broadband amplifier, you may have trouble.

We also think there may be other instances where other UHF stations may intermod together and fall into a third television channel and cause a problem. Where does the intermod come from? Well, mostly in broadband amplifiers and potentially in the front end of the DTV tuners.

You said that broadcasters have to spend more time educating people about antennas. What does the average consumer need to know?

Well, first of all, that you do need an antenna. Some people didn't realize that you needed an antenna with the new digital converter boxes. That may sound kind of amazing to you and me, but think about it. Do you have a clock radio sitting on your nightstand in your bedroom? Does it have an external antenna? No, it typically has an internal antenna or one that uses the AC power cord as an antenna. But it doesn't have a separate antenna that the user must connect. So, sometimes, it's just as straightforward as that: You need an antenna.

Now many viewers have no idea that the digital channels use the exact same frequency bands as analog, and therefore there is no such thing as a DTV or HDTV antenna. They also don't realize that the television channels are broken up into three bands: low-VHF, channels two to six; high-VHF, channels seven to thirteen; and then, of course, the UHF band, which will be channels 14 to 51 after June 12. People need to know that so they pick the right antenna when buying a new one.

If you've got an all-band antenna on your roof, you're fine, assuming that it is still in good condition. An antenna situated outdoors enduring the weather for the last 10 or 15 years may not be in as good a shape as the viewer might expect. But many people, we found out, have UHF-only antennas that were sold to them when they first heard that DTV was coming a number of years ago. But what has happened is some stations have turned off their analog and moved back into the high-VHF band and viewers are having a hard time because they have UHF-only antennas.

Sometimes there is also a problem with the use of virtual channels. In general, virtual channels used in the standard allows broadcasters to continue branding themselves using their old analog channel numbers regardless of what their physical digital channel is -- you know, Action Seven News, Sports Center Five and stuff like that. So viewers who want to watch Action Seven News may go out and buy a high-VHF antenna thinking the digital signal is on channel 7. The problem is Action Seven News may actually be on channel 51. They would really need a UHF antenna.

What kind of antenna? You can go to www.antennaweb.org, type in your address, and it will tell you the types of outdoor antennas that will get good reception at your home by determining the distance and the terrain between your home to each transmit antenna, calculating the signal strength for all the channels that it believes you should be able to receive and then providing you with a list of suitable antennas.

Viewers should use outdoor antennas whenever possible. The whole system was predicated -- all the channels were allocated by the FCC --  assuming outdoor reception with an antenna 30 feet above ground level. People often don't realize that there are no FCC planning factors for indoor reception. But if you have to use an indoor antenna, you always want to put it near a window that's facing a transmitter.

The IEEE and Iowa Public Television co-produced a 51-minute TV program (see below) that is our best attempt at educating the public about the ABCs of digital reception. It can be viewed as streaming data at the URL above. I urge broadcasters to get a broadcast-quality copy and broadcast it as often as they can if they don't have their own versions available. This video can be broadcast for free by any television station.

Losing service or a signal in your home, that's one thing. Losing it at a cable headed is another. What should cable systems do to insure the best possible reception?

Cable operators can typically put up at least a 30-foot or sometimes a 50 or 75-foot or higher tower to mount their receive antennas. They also can install a high-gain directional antenna. If they need to, they can use a very robust high third-order intercept point amplifier that's very robust against overload. Another thing they can do is to insert a filter in front of the amplifier that further protects it, if necessary. Likewise, they can use low-loss cable coming down into the cable headend itself to go to their receiver, and, of course, they should be using the most recent type of DTV receivers that contain fifth- or sixth-generation circuitry, which is very good.

Broadcasters are counting on mobile DTV to generate some badly needed new revenue. Is the DTV signal up to the task of providing the blanket coverage that such a service demands?

That's what we're going to be proving later this year. MSW [Meintel, Sgrignoli & Wallace] did the OMVC's IDOV [Independent Demonstration of Viability] field testing last year that you may have heard about. We're slated to do this next round of field tests in eight cities starting soon. So it looks very promising.

At the moment, they still have a candidate standard that they are tweaking, and the final standard is expected later this year. You've got a whole bunch of broadcasters in the OMVC backing this, pushing this and willing to try it.As a scientist, I want to get all the data in before making a final statement, but it certainly looks very good, very encouraging, and very exciting at this point. [Editor's note: the OMVC is the Open Mobile Video Coalition, a consortium of broadcasters and system developers dedicated to getting mobile DTV off the ground as quickly as possible.]

What kind of improvements can we expect in the tuners and the receivers?

We're essentially at the sixth-generation [receiver] and we're still improving performance, but we're at the point where I feel comfortable. I was not comfortable until we went from the fourth to the fifth generation in terms of the improved equalizer that mitigated the severe multipath problems and produced more robust tuners that mitigated severe overload problems. Remember that after all of the full-power analog signals are turned off, many stations will be further optimizing their signals with increased radiated power, higher antenna locations, etc.

What about on the antenna side?

We'll probably see complex smart antennas, which will use diversity and will adapt to changing reception conditions. They'll have a processer in the front end of the receiver, perhaps with the equalizer controlling it, that selects a smart combination of antenna signals and then lets the equalizer proceed to correct this improved signal.

article originally published at TV Newsday.

The media's job is to interest the public in the public interest. -John Dewey