Thanks to Jeff Hoel for providing the transcript to the episode 154 of the Community Broadband Bits podcast with Michael Calabrese on the wireless future. Listen to this episode here.
Michael Calabrese: These new trends toward opening far larger bands for unlicensed shared public access, as well as these dynamically-shared bands, such as the Citizens Band Radio Service, taken together, we will have, I think, transformed what's perceived to be spectrum scarcity into spectrum abundance.
Lisa Gonzalez: Hello. This is the Community Broadband Bits Podcast, from the Institute for Local Self-Reliance. I'm Lisa Gonzalez.
Using our electronic devices without being tethered with a wire has become more mainstream. Expanding that wireless connectivity depends on the availability of spectrum and careful spectrum policy. Michael Calabrese, Director of the Wireless Future Project at the New America Foundation, joins Chris today. They talk about the difference between licensed and unlicensed spectrum, uses for both, and how new approaches can optimize this valuable resource. Lean more about Michael's work at newamerica.org/oti .
Now, here are Chris and Michael, digging into spectrum and the wireless future.
Chris Mitchell: Welcome to another edition of the Community Broadband Bits Podcast. I'm Chris Mitchell. And today I'm speaking with Michael Calabrese, the Director of the Wireless Future Project at the New America Foundation. Welcome to the show.
Michael Calabrese: Yeah, thanks, Chris. This is great, to be part of your podcast.
Chris: Well, I'm excited to be talking to you. You've been in this space far longer than I've been working on community broadband. I think the expertise you bring is quite high. But let's start with you telling us a little bit about, what is the wireless future, and what kind of work do you do?
Michael: My Wireless Future Project is part of New America's Open Technology Institute, which is really about, you know, all things open. Open networks, such as net neutrality. Open -- you know, open technology, open airwaves. And, you know, I focus on the wireless side. And for over a decade, we've been really promoting policies that will result in ubiquitous high-speed and affordable wireless connectivity for, you know, all Americans, and, hopefully, globally as well. And, kind of at the forefront of that, is -- has been to open more spectrum for unlicensed access, that is, you know, shared public airwaves, and also dynamic spectrum-sharing, as an alternative to sort of the exclusive licensing that is what allows, you know, big mobile carriers like Verizon and AT&T to dominate their industry segments.
Chris: There's a lot of things that I want to jump in with, I think. One of the ones that comes to mind is a quick question, which is, you know, there's -- all these devices that people want to buy now are wireless, increasingly.
Michael: Um hum.
Chris: And a lot of people, myself included, are saying, look, the future's going to be wireless. But to get there, we need a lot of fiber as well. Fiber optic connections. And I'm just curious if you can explain to the audience why that might be. If you agree with us.
Michael: Oh, yes. I mean, that's -- In fact, I wrote a report, maybe two years ago, explaining why the wireless future that we envision, you know, must be built on a high-fiber diet. And the reason is that if you really want, you know, those three outcomes I mentioned -- you know, wireless connectivity for devices, you know, like smart phones, tablets, laptops -- if you want all that to be available everywhere -- high-capacity -- and especially if you want it affordable, you can't do that over the expensive licensed spectrum, and through carrier infrastructure. Because you have to pay through your nose. There's bandwidth caps, there's high pricing. And so, what people are almost naturally turning to is Wi-Fi, you know, as an alternative. And right now, you know, I think most people don't even realize that, at this point, roughly 60 percent of all the mobile data traffic -- all the broadband traffic that goes over mobile devices like smart phones -- is not actually going over licensed spectrum through carrier towers. It's going a very short distance -- you know, a hundred feet or so -- over unlicensed spectrum into a wireline network that's already been provisioned for other purposes. You know, for -- you know, at home, at work, at Starbucks, maybe a community network. And so that trend will increase. Because the more deeply we penetrate fiber everywhere, the closer you'll always be to a wire. And the more unlicensed and shared spectrum the public has access to, the more, you know, ways that it can get into those wires over the air without relying on the carriers.
Chris: One of the things that just came up, in terms of licensed versus unlicensed, is this -- the really horrible train accident -- the Amtrak accident just north of Philadelphia. And you wrote an article, which is what reminded me that I really wanted to bring you on this show. Along with Patrick Lucey, who couldn't join us today, but is with the Open Technology Institute as well. And in it, you talk about how Amtrak had been delaying implementing some of this safety technology, because they really wanted this licensed technology. And, in fact, they could have been moving ahead with an unlicensed approach. So, maybe you can just sort of explain the difference between licensed and unlicensed, in those terms.
Michael: Well, what happened with Amtrak is that, seven years ago, Congress mandated a safety technology called "positive train control." This was part of the, you know, Railway Safety Act. It would automatically slow or stop a train once it exceeds a speed limit, such a ahead of the curve where the Amtrak train jack-knifed off the rails, north of Philadelphia. And positive train control is based on, you know, a series of beacons and sensors that -- where the train, as it's going, communicates, you know, with the railway infrastructure and safety infrastructure, all along, you know, the railroad right-of-way, you know, through wireless connections. So, the question was, in implementing this, what spectrum -- you know, in other words, what airwaves would the railroads use? And so, they decided to ask for this extremely expensive television-band spectrum. A fairly small amount, but spectrum that is so prized that at -- you know, that at the auctions that are scheduled for next year, the Federal Communications Commission expects that the wireless industry will pay roughly one and a half to two billion dollars for that much spectrum. And they wanted to have that just, you know, to connect along the rails. So, Congress wasn't giving it to them. The FCC was prohibited by law to just give it away. Because the law requires that it be auctioned.
Chris: And, to be clear, the -- this would be spectrum that only Amtrak would have, and nobody else would be able to use under any circumstances, right?
Michael: Right. This would be exclusively-licensed spectrum, meaning that only one company could use it, you know, at a given -- in a given geographic area. So this licensed spectrum -- you know, they really needed it only in a very specific place. And not even at high power, just along the rail lines. And yet they were looking for this very expensive spectrum, you know, which is licensed over huge areas, and which is -- as I said, is probably -- costs -- is valued at one to two billion dollars nationwide. So, since the FCC was prohibited by law to just give it away to these private rail companies, the FCC's advice was, well, go to the secondary markets and negotiate, you know, leases, or to purchase the licenses from other parties that hold this spectrum. You know, utilities and others. So the freight railways formed a consortium and more or less did that. But Amtrak didn't have the resources, and there's just a lot of problems with secondary markets. So, eight years later, they had just barely, you know, gotten hold of the spectrum and had not deployed the positive train control system that would have avoided the crash.
And alternative -- you know, what we write about in the article, is that, increasingly, industries like that -- like utilities, health care, industry, and others -- are finding that shared spectrum is much more plentiful and economic, for -- even for critical uses. So, for example, the majority of wireless monitoring of patients -- healthcare patients -- is done over unlicensed spectrum. And unlicensed spectrum is spectrum that simply -- bands of frequencies that are left open, for shared public use, at lower power. The U.S. leads the world, by far, in the deployment of smart grid and automated smart meters, because they're using unlicensed spectrum, rather than paying for that connectivity from mobile carriers. And in industry after industry, for inventory control and so on, there's an increasing move toward using this unlicensed shared spectrum. But the mentality remains, you know, among some industries -- and the railroads were on that list -- the mentality that they need to have exclusive licenses so that they have absolute certainty. But right now, you know, that spectrum is really not available, except at extraordinarily high prices, and often after long delays. As we saw in Amtrak's case.
Chris: I think there's a good public policy question, which is, of the best spectrum we have available, should we be devoting slices of it to just one technology, when you could be sharing it? I mean, I come back to, actually, the thing you started off with, in noting that some 60 percent of our smart phone traffic goes over Wi-Fi. You know, I think if you went back 20 years ago, and you told people, in 20 years, everyone's going to have these hand-held devices, they're going to be able to communicate, they're going to be able to share thoughts with anyone in the world, and all of this is going to take place on just these narrow bands that we consider to be "junk," people would probably think you were crazy.
Michael: That's right. I mean, the -- sort of the old approach was command and control, where, you know, each industry kind of came, in turn, and got their exclusive assigning of frequencies. And that's turned out to be very inefficient. Because all of the "good" spectrum -- you know, the stuff that will travel fairly long distances, or will, you know, penetrate through obstacles, is assigned to someone. And most of it's not being used very intensively. Technology's changed. Industry's evolved. And yet, all the old assignments are still there. So as a result, policy-makers have been moving away from exclusive licenses toward a new framework based on flexible use and shared access to spectrum bands.
Chris: This is where we actually have just had action from the Federal Communications Commission, regarding something, I think, you know, you and I are both pretty excited about: this citizens band. Why don't you tell us what's going on there?
Michael: Oh, sure. Yeah. So, this is, like, a whole other path toward broadband abundance. One is, you know, as we mentioned earlier, Wi-Fi. So, you know, Wi-Fi grew up on these "junk" bands -- on this unlicensed spectrum that was originally set aside for things like cordless phones and microwave ovens. Wi-Fi has become so prevalent that even the Wi-Fi bands are getting, you know, crowded. But there are other, much larger bands of spectrum, particularly those being used by the federal government -- the military, in particular -- where most of the capacity is unused. And so if you have, instead of exclusive use, if you have dynamic sharing, you can unlock that capacity, and potentially create bandwidth abundance. And that's what's happened now.
A few years ago, I was actually part of this process to write a report for the President's Council of Advisors on Science and Technology -- the PCAST. And in 2012, they recommended to the President, in a report, that there could be a spectrum -- a shared spectrum superhighway across more than 1,000 MHz of just federal bands of spectrum, IF it was all opened up for dynamic sharing. And they recommended that this could be done, like, working around the incumbents -- like military radar and other federal uses -- by using a geolocation database system, and, in some cases, sensing. This was called a Spectrum Access System. And that there could be different tiers of access, where the incumbent users -- the federal agencies -- they would be protected from interference, but you'd have opportunistic access by anybody else, essentially unlicensed access, as well as the option to have some priority access licenses that could be sold or auctioned for short-term use, for particular industries, maybe by Amtrak.
Chris: At the recent Freedom To Connect event, in New York City, that David Isenberg put on, Milo Medin, from Google, gave a presentation in which he was talking about this sharing. And he noted that there's bands that are reserved just for the landing system on aircraft carriers. And that you don't have a whole lot of aircraft carriers in Kansas City. So we would be able to use some of that spectrum without a threat. And if I understand it correctly, you have -- you'd have some people that would be paying for priority access. And that would be on a dynamic basis. Basically on a sort of -- is it by, like, five-minute blocks, or -- I'm just -- I don't understand how it really works, at that level of detail.
Michael: In effect -- it's funny, because Milo and I drafted the chapter on, you know, for the PCAST report on how this could work. And Milo's exactly right, because this Navy band was the one that the PCAST recommended be the first to be opened up for this, you know, dynamically shared model, which the FCC adopted last month. And it's called the Citizens Broadband Radio Service. And what it does is, there's 150 MHz of spectrum, that's mostly dedicated to the Navy. But they only use it off-shore.
Chris: Right. And 150 is an incredible amount, right?
Michael: Oh, yeah. That's about as much or more than any, you know, of the national mobile carriers have, you know, for their networks. And this would be, now, available for shared use. There will be a passive sensing network set up -- some sensors set up along the coastline, to detect if a ship comes close to shore, using its -- with its radar on. If it does, the devices will be told to get off, you know, those particular channels that the ship is using. But, other than those situations, and across the entire inland part of the country, that 150 MHz will be available all the time. And it will be available -- most -- a majority of it -- 80 MHz of it all the time for unlicensed use. They're calling it General Authorized Access -- was the FCC's term for it. But then there also will be as much as 70 MHz available for priority access licenses, or PALs. And what these are 10-MHz licenses that companies could acquire for -- they're essentially three-year, non-renewable licenses. Which is much shorter-term. Today, licenses are wide-area. They're typically for 8 or 10 years. And they're automatically renewable. These would be non-renewable, would be shorter -- much shorter term. And are for much smaller land areas. The PCAST actually recommended much more of a Google AdWords market approach, where -- essentially, it would be like a spot market.
Chris: That's what I was thinking. Right. That's what I was ** thinking of.
Michael: Yeah, yeah. Exactly. Like, the idea was that -- like, if a hospital campus needed to know that they had quality of service for some particular use, they could go in and just buy -- essentially, you'd pay for interference protection, in just a very small area, where you really need it. But the FCC thought that, initially, that was too difficult to administer. So, what they've -- they sort of jury-rigged the traditional system, to make the license areas very small, the license terms very short, and non-renewable. Eventually, though, once -- I think once the Spectrum Access System proves itself, that we will end up with a Google AdWords type of, you know, auction system, where you just get, you know, the rights for the short time and place you need it.
Chris: So, there's so much more detail that I'd like to go into. But I want to keep this at a reasonable length. And I'm curious, a lot of us envision a future -- or we want to envision a future -- where neighbors can get together, or entire neighborhoods, and build a wireless network with off-the-shelf parts. That would not be very difficult to build. And we've been hoping that some of these spectrum -- these new spectrum approaches -- would enable that. Is that the kind of thing you think we can hope to move toward?
Michael: Well, I think these new trends toward opening far larger bands for unlicensed shared public access, as well as these dynamically-shared bands, such as the Citizens Band Radio Service, taken together, we will have, I think, transformed what's perceived to be spectrum scarcity into spectrum abundance. Meaning that, you know, for communities, there will be plenty of wireless capacity there that will be free for the taking. Then it just becomes a question, really, of the equipment and, in a sense, what you might call the business model. It's really interesting. I think many people assume the main obstacle is spectrum. Because they see -- for example, there was an auction in January of this year that raised a record of $41 billion, for about 50 MHz of spectrum -- for about a third of what the Citizens Band Radio Service is opening. So, the assumption is that it's spectrum that's scarce. But spectrum isn't scarce at all. Less than 20 percent of the best spectrum is even in use. So once we open it up for this dynamic sharing, and have more unlicensed, more opportunistic access, spectrum won't be the obstacle. But there will still be challenges as far as getting equipment on a scale, you know, that makes these sort of networks affordable. And then getting the political will, you know, to put them in place.
Chris: Right. I think that's something that sometimes people may not have realized, which is, one of the reasons Wi-Fi is so successful is because there's such a large market for it. And if device makers aren't convinced there will be such a large market, then they're not going to be developing the machines, the technology in the right way to enable that outcome.
Michael: Right. Right. Yeah. Wi-Fi is premised on the economics of the Internet, in effect, where the investment is at the edge, you know, by individuals. And it's all very incremental and affordable. So, just as, you know, you expand the Internet every time you attach a new server in a new location, you know, Wi-Fi is similar. Because you attach a router to a wireline connection. And then broadcast connectivity. So what's really going to be needed in the future is just to try to extend the logic and the affordability of Wi-Fi to these community networks, so that instead of -- you know, instead of separate hot spots, you have true, you know, in a sense, hot zones of mesh networking.
And there are examples of that. I mean, we -- you know, our Open Technology Institute has been working with communities to build out these networks in some low-income areas in Detroit, and Philadelphia, and Brooklyn. And Mayor de Blasio, in New York, is working with us on this new RISE program, which -- it's also based on neighborhood resiliency -- where there were going to be working with them to build out community Wi-Fi networks that can withstand a hurricane. Because that was the initial effort in Red Hook, part of Brooklyn.
So these are all really exciting developments, and, hopefully, as communities realize that the spectrum is there for the taking, that there will be more efforts like that.
Chris: And I think we end up back where we started, with -- you know, there's a lot of promise for future spectrum, as long as we continue to see a federal agency -- the Federal Communications Commission -- manage it correctly. But communities should be preparing, by getting fiber deep into their neighborhoods ...
Michael: Um hum.
Chris: ... so that they can make sure they have a lot of areas where they can connect these nodes. And we'll see what happens.
Michael: Yeah. And also by opening their public fiber. Because, you know, many communities have fiber that's been put down for various public purposes -- I-Nets, school-nets, ...
Chris: Right. Traffic signals. All kinds of things.
Michael: Right. And if -- you know, and often there's excess capacity there. And if that was -- if that was, you know, open, it would spur not only more community networks -- Because, you know, as we said earlier, you know, Wi-Fi on a high-fiber diet is much more potent, right? You know. As long as you're close to the wireline backhaul, and that's fairly inexpensive, you can put a cloud of connectivity over a neighborhood. So, more fiber, more fiber, more fiber access to it. We'll have more spectrum. So, the pieces are coming together. We just need to find, you know, in a sense, the right recipe to navigate around the economics and the politics.
Chris: Right. Well, thank you so much for taking the time to come on the show and talk with us about these issues.
Michael: Yeah. Thanks, Chris. This was great.
Lisa: Send us your ideas for the show. E-mail us at firstname.lastname@example.org . You can follow us on Twitter. Our handle is @communitynets . If you use Facebook, be sure to like our Community Broadband Networks page. Thank you again to Persson for the song, "Blues walk," licensed through Creative Commons.