BALTIMORE—Two months ago, Sinclair Broadcast Group, Nexstar Media Group, Univision and American Tower announced they would work together to setup an ATSC 3.0-based single frequency network in the Dallas market. Work is progressing on the SFN, which with a bit of regulatory help from the FCC should be partially lit up before next month’s NAB Show in Las Vegas.
The SFN effort is more than simply a trial of Next-Gen television, however. It is also a test of a proposed ATSC 1-to-ATSC 3.0 transition that relies on spectrum clearing, channel sharing and what may be an unprecedented level of cooperation among competing broadcasters.
Mark Aitken, vice president of Advanced Technology at Sinclair Broadcast Group, has been at the center of the rollout of this Next-Gen SFN trial. In this interview, he discusses how work on the single frequency network is progressing and how the trial will provide Sinclair with critical data that will help it achieve its goal of deploying SFNs nationwide to usher in a new era of television.
(The following is an edited transcript.)
TVTechnology: Sinclair, Nexstar and Univision, along with American Tower Corp., announced in January the joint construction and operation of an ATSC 3.0 single frequency network in Dallas. Can you bring me up to date on where the project stands?
Mark Aitken: There are three new SFN sites that will host the Next-Gen service of five local Dallas stations. We refer to this as a multi-channel, multi-tenant SFN—obviously Next-Gen-related.
We are in the process of providing a network infrastructure that ties to all of those facilities so that we have what we refer to as a Next-Gen Broadcast Operations Center. That will be a local presence that also is capable of being run from Baltimore [the home of Sinclair’s headquarters].
On a local level, we have the issue of getting the SFN up and running. First off is the main transmission site with Next-Gen, tying that into multiple SFN sites and all of the work that goes along with that, including timing and determining the proper power overlaps.
Once that is working, we must address how to operate that as a remote site. Eventually, that operation will go back to a regional center because it is envisioned that Dallas is just one of a number of DMAs that get serviced from a regional hub.
TVT: What stations will participate?
MA: The target stations—and I say target stations because all of this is subject to FCC approval—are KSTR, a UniMas station; KTSD, an independent owned by Cunningham Broadcasting; KUVN, the Univision affiliate; KDAF, the Tribune Broadcasting CW affiliate; and a fifth station, which I cannot divulge at the moment.
Two of those stations—KSTR and KTXD—will relieve themselves from their ATSC 1 activity and be hosted by the others in a channel-sharing arrangement. This is important: this channel sharing means there will be no channels lost to over-the-air viewers. None of the core channels nor the Diginets will be lost.
All five will also broadcast all of their channels—core and Diginet—as Next-Gen service on the SFN. I think that is a total of 20 channels.
TVT: That setup must put big demands on MPEG-2 encoders for the ATSC 1 side of the operation.
MA: This is a crowded field, and we need capacity for additional channels on the 1.0 stick. What that encompasses is bringing in new software-based encoders that are being optimized for a new and different level of service.
In that regard, we have a vendor with encoders that will do one HD and depending on program content five, six or seven SDs at the same time. Or, they are capable of supporting three HDs at the same time, or supporting a whole host of combinations in between.
TVT: I’ve often heard it said that getting a group of broadcasters to agree to anything is like herding cats. This Dallas arrangement seems to fly in the face of that.
MA: My emphasis in Dallas is on learning how to get along. None of this happens without participation and cooperation.
It’s about keeping that 1.0 piece whole, not impacting the consumer. Not impacting the revenue of stations and making way to fully impact the unleashing of new, future services. Those things equate to keeping the industry alive.
TVT: What is the target launch date?
MA: We are hoping to have part of the system up and running prior to NAB. I say trying to because once you figure out who your partners are for the sharing on the 1.0 side, you’ve got notification requirements and FCC filing requirements for post licensing through the commission.
If you look at what came out of the rulemaking, a secondary license gets issued for the purposes of being allowed to carry your content on another channel [channel sharing to continue 1.0 service]. And that’s further made difficult if you have a PBS non-com in that mix because of the nature of non-com rules. On top of that, the FCC form called out [in the ATSC 3.0 rulemaking that details the 1.0 hosted environment] doesn’t even exist at the moment. So, we’re working very closely with the FCC.
TVT: You’ve mentioned in the past that Sinclair wants to rollout a nationwide network of 3.0 SFNs. Will this SFN trial help you achieve that goal, and if so, how?
MA: What Dallas allows us to do is check off the list, “OK, here’s the equipment, here are the available vendors, here is the project timeline and here is the sequence of events.”
It’s a project in its own right, and it has its own timeline impacted by deliveries and acceptance and commissioning and all of those things. When you are talking about multi-channel, multi-tenant sites for SFNs, you’re building new transmission facilities.
It is really understanding what resources are available, what capitalization is required, what the interrelated services are and how they get sequenced. It is really a soup-to-nuts approach to a somewhat complicated puzzle.
By the way, what’s gone on in front of that is the actual engineering and planning of what is the nature of a properly designed single frequency network.
How do you get the coverages lined up, how do you coordinate coverage in an SFN environment versus interference? It’s all those engineering studies. It’s all those—in our case—the use of Progira software as the planning tool for understanding the nature of what the end result is going to be.
TVT: You’ve had an ongoing SFN trial in the Washington, D.C.-Baltimore corridor. What takeaways from that trial will be helpful in Dallas?
MA: What’s helpful is we understand how to synchronize and balance the propagation characteristics in that SFN, the nature of performance we need in the network. We know the nature of the components that go into that. However, we need more than a single market to understand propagation. We do propagation studies. But do they match in the real world what we have synthesized in the environment in a software program?
TVT: How will you be conducting field strength tests and other needed tests in Dallas?
MA: ONE Media will deploy its mobile lab, which is a completely outfitted van that will do every level of testing on the move, not just from fixed sites—so automated data collection on the move in a real mobile environment.
We have to take that data and compare that against the projected data coming out of Progira and ensure that what we think we’ll get is what we’re getting. We started that process in Baltimore, but that’s just one market.
TVT: Why Dallas? Was it the fairly flat terrain?
MA: Look, it would be foolish to start with the most difficult cases and there are plenty of those. Top of mind, you have Seattle, Pittsburgh and lots of other markets. But it wasn’t chosen specifically because it is flat. There were a whole number of considerations.
We’re looking at how we would do a connected, regional service. The plans that we have made in terms of the SFN aren’t just about the Dallas-Ft. Worth market. They extend down a whole corridor and the next in line is Waco.
It also was chosen because of the availability of partners. It was chosen because Dallas is a bit of a high–tech center. It was chosen, by the way, [because] Nexstar’s home office is in Dallas.
It was chosen for a whole number of reasons, but clearly because it is reasonably tame geographically, first things first. The mantra here is crawl, walk, run.
TVT: One theoretical SFN deployment scenario that is often described is ringing the perimeter of a market with several—maybe four—SFN sites with directional antennas pointed inwards toward the big stick. I know you’ve said there will be three SFN sites, but is this the type of approach that will be taken in Dallas?
MA: Actually, it is substantially of that nature. They are not directional antennas, but they have a degree of directionality. They’re cardioids.
If you looked at the plan, you would see there is a fourth site, but it actually overlaps into the Waco market, and that’s best suited to be served by the channel structure in Waco as opposed to Dallas.
So, we know we need that in the lower side of Dallas. We’ve focused on the high population densities—the northern sort of growth patterns of that market—to light that up first.
TVT: What transmitters will be used at the SFN site and what will the ERP be?
MA: We will be using 5kW Comark solid state transmitters. The ERP will be 100kW.
TVT: Is there a target average signal strength you are trying to achieve for the market?
MA: The target signal strength is 60µv/square meter at 1.5 meters off the ground, or the typical distance off the ground a person would be holding a portable device. That’s based on a 95 percent probability service factor.
TVT: What are your plans for synchronizing the transmitters on the network?
MA: It is a matter of going out in the field, looking at the impulse response of merged transmission systems. We’ll be flexing some of the tools like TxID [ATSC 3.0’s transmitter identification signal] that allow us to identify the nature of the multiple signals being received.
We know on a map where the mating point is, if you will, of converging emissions from towers. We will go out in the field and set those up. And we’ve got GPS lock and distribution of common timing across the network. It’s actually a pretty easy process.
TVT: If the GPS satellite signal were lost, is there a backup?
MA: The GPS assumes we have a running Rubidium-based clock synced. That way, you can lose satellite coverage and still have a stable clock.
It really is the case that GPS is bigger than a single SFN. You need a stable clock source that ties all of these entities together. But there are multiple ways of doing that and we will be testing other methods as well.
All the other industries that surround us do this on a day-in-day-out basis, whether it’s LTE Broadcast, DVB-T2, DTMB, go down the alphabet soup list. That’s not the difficult piece of the equation.
TVT: What about STLs? Microwave or fiber?
MA: While we have fiber to all sites, we are purposely dropping microwave into the middle of that to dissuade any concerns people may have about a mixed environment.
TVT: How important is sharing to the success of your SFN strategy?
MA: It is really important, because it is a cost share. It is also about efficiency of resource use.
The fact of the matter is I want to build an SFN and you want to build an SFN. Well, does it really make sense to deploy two different tower crews and do it sequentially? So, there are cost savings and resource savings and I would say that in the planning of that there is a level of redundancy and resiliency that gets built in by the nature of the quality of service desired.
TVT: You said ONE Media will be in the market with a specially built van taking measurements. Are there other plans for reception besides the engineering side of things—perhaps seeding the market with ATSC 3.0 receivers?
MA: Absolutely. We’ve got a number of parallel paths underway to bring to market the devices that are going to ultimately support the business. So, gateway type devices. We’ve got activities underway with a couple of proposals. A couple of the participants are actually building prototype receivers and prototype gateway devices.
We are in the process of lining up the type of launch services, but a lot of that is in the hands of Spectrum Co and John Hane [newly named president of Spectrum Co].
TVT: How much will this SFN deployment cost —or is that proprietary information?
MA: We will make public the cost when we tally it up. There won’t be any secrets here. What I will say is it is running substantially below the budget we projected. That budget was in line with projections many people have made over the last couple of years with respect to the cost of building out shared SFN sites.
TVT: Dallas is the model. But each market where you ultimately will deploy SFNs will be different. The big sticks in one case might be co-located on an antenna farm and in others they will be spread out throughout the market. So how much can you walk away with from this SFN trial that will be applicable to other markets as you deploy a nationwide network?
MA: I think the framework is 100 percent applicable. The variables inside that framework may be different. Is it three, four, five or a half dozen transmitters? That changes it. The nature of the work that has to be done on site is another variable. Is the existing tower fine, or does it need to be reinforced?
But we know these variables because all of these variables are in play in Dallas. While the specifics won’t be exactly the same, we are defining a framework of decisions that will have to be made. So, we will have a well-defined decision engine that we hope will help us deploy in virtually any market.
TVT: Is there anything else?
MA: I think the industry wishes us success and we wish great success to others in the industry that are doing their own Next-Gen implementations.
For a comprehensive list of TV Technology’s ATSC 3.0 coverage, see our ATSC3 silo.