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

http://www.tvnewsday.com/articles/2007/09/13/daily.4/

TECH ONE OF ONE WITH MERRILL WEISS
SFN: ARE MANY TRANSMITTERS BETTER THAN ONE?
TVNEWSDAY, SEP. 13, 9:04 AM ET
There is no doubt in the mind of the technical consultant who helped develop the technology. He believes that a single frequency network of transmitters operating on the same channel can blanket a market with a stronger, more uniform signal than the conventional big-stick approach, improving indoor and planned mobile reception.
To get the greatest coverage, you put an antenna at the highest possible point on the tallest tower you can find and you plug it into the most powerful transmitter the FCC will let you have. 
But, despite all those superlatives, simple might not always be best.
In the new world of digital broadcasting, it might be smarter in some markets to deploy a network of complementary transmitters operating with less power and relatively low antennas.
The so-called single frequency networks (SFNs), according to their proponents, blanket markets with a more uniform and pervasive signals than the conventional big-stick approach, improving indoor and planned mobile reception.
SFNs also may prove a handy alternative when, for whatever reasons, the big stick just isn't practical.
Indeed, that's the case in New York City. There, two separate groups of broadcasters are experimenting with SFNs, having concluded that the their common "tower"-the Empire State Building-just won't accommodate all their digital signals come the analog cut-off date of Feb. 17, 2009.
Representing the 10 largest stations in New York, the Metropolitan Television Alliance is looking at SFN technology as an interim solution, a way to bridge the gap between the analog cut off and the day in 2011 when the all the stations are supposed to migrate to a new mast atop the 1,776-foot Freedom Tower, the tallest of five skyscrapers being built where the Twin Towers fell on 9/11.
MTVA President Paul Bissonette told TVNEWSDAY in an interview posted on Tuesday that testing will soon begin of a prototype SFN system, including four low-power transmitters in Brooklyn.
Meanwhile, Ion Media and Richland Towers have reported positive results from their preliminary testing of a separate SFN system comprising one transmitter in West Orange, N.J., and another in Times Square.
The technology is still in its infancy. Only a handful of systems have been deployed in the U.S. with its 8-VSB digital broadcasting standard. Because of that, Bissonette offers no guarantee in New York. "Nothing like this has ever been attempted before in an urban environment," he says.
For a better understanding of SFN technology-its variations and potential, TVNEWSDAY turned to Merrill Weiss, a technical consultant based in Metuchen, N.J., who may be the foremost authority on the technology. He holds a key patent and has been involved in most of the applications, including those in New York.
The recipient of the NAB's 2006 Television Engineering Achievement Award, Weiss explains the technology and how he expects a full-blown SFN system that he is working on in Philadelphia to soon settle any question about SFN feasibility once and for all. 

An edited transcript follows:

TVN: Give me the short course on single frequency networks.
Weiss: A single frequency network is a system in which multiple transmitters are used to deliver signals to an area that a station is trying to cover, and the signals from those multiple transmitters overlap one another at least in some areas.
We talk about small-cell systems and large-cell systems. In the large-cell systems, you're basically spreading the transmitters far apart and you try to minimize the amount of overlap. In a small-cell system, you have transmitters close together and you actually intentionally overlap the signals from adjacent transmitters, but you try to prevent overlap two cells away. Now, when you talk about single frequency networks, there are three types of transmitters-digital on-channel repeaters, distributed transmitters and distributed translators.
An on-channel repeater basically receives the main signal over the air and simply retransmits it. The issue there is that there is some time delay that causes an echo. So the trick is to minimize the time delay.
Distributed transmitters all share the same channels, but they are simultaneously fed with a signal and synchronized in such a way that they do not interfere with one another.
The third approach is distributed translators. You actually need two channels. You put the main signal on one and then you can have multiple translators on another channel that are synchronized just as the distributed transmitters are.
TVN: Basically, you're talking about multiple transmitters operating at lower powers rather than one big stick.
Weiss: Maybe. You could put up a large facility just as you would have built in the past, but find that it doesn't reach all of the service area because of some terrain blockage. You then could put in an additional transmitter to fill in the part of the service area that you could not reach from your single transmitter.
TVN: In the distributed transmitter and distributed translator approaches, how do you prevent the transmitters operating on the same frequency from interfering with each other?
Weiss: In reality, they do interfere with each other. The trick is that we take advantage of some technology that's built into each one of the digital receivers that is designed to deal with echoes that occur in a natural environment. That technology is the adaptive equalizer. There's one in every digital receiver and they've been getting better over time.
The big reason that the early digital receivers didn't work too well and people were concerned was because the adaptive equalizers were not really quite good enough.
But over time they've gotten a lot better and as a consequence, the VSB transmission system is working a lot better. And so we can take advantage of that in the systems that use multiple transmitters by making the signals from those transmitters look like echoes that occur in the natural environment due to signals bouncing off of buildings, mountains and things of that sort. 

TVN: When you say echoes, you mean ghosts.
Weiss: Yes, except that ghosts are an analog phenomenon. They create ghost images in the picture. In the digital world, you get exactly the same phenomenon, but instead of showing up as ghosts in the picture, what it can do is destroy the receiver's ability to extract the data. 

TVN: What was your particular contribution to this technology?
Weiss: I developed a way to synchronize the transmitters so that the signals that they emit would be seen as echoes by receivers. 

TVN: How does that work?
Weiss: When you do distributed transmission, you need a studio-to-transmitter link to each of the sites and what you get for that is the ability to adjust the relative timing of all the transmitters. You have complete control over them.
We insert a little extra data in the signal to each of the transmitters. That little extra data tells the transmitters to produce outputs that are appropriately timed with one another. 

TVN: How many of SFN installations are you aware in the U.S. right now?
Weiss: Well there are five distributed transmitter systems either on the air or in under construction that I'm aware of. There was a sixth that was put up for test and experimentation purposes. 

TVN: Where are they?
Weiss: One is in State College, Pa. Penn State put that on the air in 2003. There's one down in Virginia, operated by Shenandoah Valley Public Television. There's one going on the air in Philadelphia-an eight transmitter network that soon will be on the air. They are testing two systems in New York City. And Tribune had one in Indianapolis that was for test purposes.
TVN: I know that you were involved in the State College project. What does that look like?
Weiss: That is designed for four transmitters, one of which is a megawatt-class transmitter. It's 810 kilowatts, but it's high enough that it's equivalent to a megawatt, and then there are three gap fillers, one for each of the three major population centers-State College, Altoona and Johnstown. 

TVN: So they have one big stick and three little ones.
Weiss: Yes, but only one of those smaller ones is on the air at the moment. They initially only had the money to do those two transmitters. Also, they were doing it on an experimental license basis and so they have to complete the field testing and get data to the FCC. Once they do that, they'll go ahead with the other two transmitters. That field testing is almost complete. 

TVN: What's the verdict there?
Weiss: Oh, it's great. It works. I mean it basically shows that, in a preponderance of locations in the State College area where people could not have received digital television from the main transmitter, they now can get it, and, in many cases, they can get signals that are strong enough for reception indoors on a set-top antenna.
Now, just to be complete, there are some locations where there's some amount of interference between the two signals and it actually makes it a little worse.
You have to be aware of that when you build one of these networks, what you're doing is balancing the gains versus the losses. You try to gain a lot and lose only a few. And you try and do it in a way that the few can still receive a signal, if they use a directional antenna.
TVN: You've got a big project right now-WTVE Reading in the Philadelphia market. Tell me about it. [Editor's note: WTVE is just coming out of Chapter 11 bankruptcy. Trustee George Miller is hoping improved coverage and $2.6 million in annual revenue will attract a buyer at his asking price of $12 million. The station, which operates on analog channel 51, is now filled with infomercials, paid religion and paid sports.]
Weiss: Reading is a large-cell, distributed transmission system with eight transmitters operating on digital channel 25. Some of the transmitters are 80 to 100 miles apart.
They had a construction permit that allowed them to have a megawatt class transmitter that was about 50 kilometers or about 30 miles outside Philadelphia. What we did instead was put a relatively high power transmitter in at the Philadelphia antenna farm and then fill out the service area with the other distributed transmitters.
It's got one additional issue that we had to deal with and that is that there is an adjacent channel station [channel 26] in the market.
Part of the management of interference between adjacent channel stations is to co-locate the transmitters whenever possible. Because of the fact that there was an adjacent channel in the market, what we had to do was put a higher power transmitter in the neighborhood of that adjacent channel station, but at a lower power and close enough to it and close enough in power that [the adjacent channel] signal didn't overwhelm the signal from the distributed transmission system that we're trying to put in.
So, to make all this work, we were pretty much forced to put a higher power transmitter in at the Philadelphia antenna farm. So that in the regions where that other station, which is a megawatt class station, is very strong, we had a strong enough signal that we could overcome the interference from it.
As we move away from the adjacent channel station's transmitter to where its signal gets much weaker, we use lower power transmitters. We are making the signal from the distributed transmission system stronger than the adjacent channel, but not so strong as to cause any interference with it. We also designed a special antenna pattern that prevents what I call hot spots around the base of each of the transmitters.
TVN: So you put a big transmitter at the antenna farm with everybody else and seven other transmitters scattered throughout the market. What type of towers are you using outside the antenna farm?
Weiss: Six of them are cell towers and one of them is a broadcast tower. Some of them have the antenna only a hundred feet off the ground even though the towers are sometimes much taller than that. And some of them are higher than a hundred feet depending on what we're trying to do. 
TVN: When will the system be up and running?
Weiss: Well, we're turning it up now. Part of it's on the air already and we're turning on additional transmitters day by day.
TVN: And at the end of the day you're going to have much better coverage within your permissible coverage areas than if you had put one big installation in at the antenna farm?
Weiss: That's correct. The market includes outlying communities like Reading, Lancaster and Allentown. They're modest sized cities. The signal will be about 20 dB stronger in those places than we could get from the antenna farm.
And using the antenna farm alone was not really an option because we couldn't cover Reading, the city of license, from there. There was an obstruction in the way. 

TVN: Is this going to more expensive than the conventional approach?
Weiss: It turns out that it will be in this case because of some things we had to do to meet some FCC limitations, but it didn't inherently have to be.
TVN: I'm told that the distributed transmission approach is much better for mobile broadcasting. Would you agree?
Weiss: Yes. When you are using a mobile receiver, you can imagine going around a corner or just going down the street and going past a building that blocks the signal from a single transmitter. But, if there are multiple transmitters, then you have signals coming from multiple directions and it's much more probable that if you lose the signal from one, you'll still have the signal from another.
Of course, this depends on how the network is designed, but, if the network is designed for mobile applications, then you get what's called transmitter diversity that allows you to continue to receive a signal even though major obstructions may appear and disappear in the path to a single transmitter. That's just one factor. Another is that having transmitters closer to receivers provides a more reliable service. You can see that in the cell-phone system.
TVN: ATSC is working on an in-band mobile broadcasting standard. I suppose it will have the same kind of adaptive equalizing technology that the DTV receivers do.
Weiss: Right. One of the big challenges is to make the equalizer work better. Or, you could eliminate the need for the equalizer in certain ways by making the signal less susceptible to echoes.
TVN: What kind of interest are you getting from broadcasters in single frequency networks?
Weiss:Well, let's put it this way: I know that a couple of years ago one group broadcaster that took a look at its stations figured that it could use distributed transmission in half of its markets. 


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