We Win Again On 3650-3700 MHz. So What Does It Mean?

Back in 2004-05, a bunch of us fought to open up the 3650-3700 MHz band for unlicensed use (Sometimes refered to as 3.65 GHz rather than 3650 MHz). While we did not get “pure” unlicensed, the FCC’s “hybrid unlicensed” regime gave us pretty much everything we wanted.

In August 2005, a group of tech firms led by Intel filed a Petition for Reconsideration. This group, which I dubbed the “WiMax Posse,” wanted the Commission to reverse itself and optimize the band for WiMax operations. Notably, this meant adopting a licensing regime instead of the open spectrum rules we won in March 2005.

By this time, Powell had left and been replaced with Kevin Martin. Martin had earned the eternal scorn of Netheads by deregulating DSL (actually a process begun by Powell). And, unlike Powell, Martin had no record of support for open spectrum. So even though the WiMax Posse and the various licensed wireless providers who came in to support them raised no new arguments, no one knew whether Martin would reaffirm the 2005 rules or side with the licensed spectrum/WiMax posse.

So I let out a huge sigh of relief and felt a modest sense of accomplishment when the FCC issued an Order denying the WiMax Posse Recon Petition and basically reaffirming our March 2005 win. Commissioner Adelstein had a very nice concurring statement highlighting the important roll played by WISPs and Community Wireless Networks (CWNs) in getting wireless connectivity to rural and underserved urban communities.

So what does this mean for wireless deployment for WISPs, CWNs, and muni systems? How do I read the FCC tea leaves in light of last month’s FCC decision terminating two important open spectrum proceedings? See below . . . .

So What Rules Did the FCC Adopt?

The FCC adopted fairly straightforward rules for the band. The base stations operate at a bit over 20 watts, which is about 20 times the power allowed to a standard “Part 15” transmitter in 2.4 GHz or about 3 times as much power as allowed for 2.4 GHz devices with “smart” antennas. Anyone deploying base stations must register in a central, accessible database. All parties that deploy have a responsibility to work together to resolve interference issues, and no party deploying has a superior right to any other party deploying — everyone must work together as equals. (The usual rule in licensed services that allow multiple users is “first in time, first in right.” We strenuously resisted this because it gives incentive for operators to try to “capture” territory by deploying all over the place as fast as possible and then pumping out noise for the sole purpose of causing rivals to work around this “first in time, first in right” user). But the FCC will not resolve issues or arbitrate between parties. (After all, industry and technoLibertarians have for years maintained that we don’t need an FCC because private parties have a natural incentive to work things out. This provides an excellent chance to see if this is true.)

To protect the incumbent earth receiver station licensees, the Commission created substantial exclusion zones around the existing receiver stations. Because these receiver stations are clustered along the coasts and the great lakes area, the exclusion zones prevent deployment of unlicensed devices in most major U.S. markets. On the other hand, the rules require the existing licensees to listen politely if someone wants to come along and negotiate a specific unlicensed system deployment. Again, the Commission will not arbitrate between parties. The licensees do not have to do a deal, they just have to take the call.

Much lower power mobile devices will only function if they receive a “permission to operate” signal from a base station. This will keep mobile devices from moving into the exclusion zones. At the same time, as 3.65 GHz base stations get deployed, manufacturers will have incentive to include 3.65 GHz capacity in devices, ultimately allowing point-to-point, mesh and cellular architectures to develop.

Finally, all devices will need to use “contention based protocols” that give all transmitters an opportunity to send signal. This condition got somewhat modified in the FCC’s most recent Order. Wifi uses a “listen before talk” protocol that checks to see if the frequency is available. If yes, it transmits. If no, it waits a random amount of time and tries again. Wimax uses a scheduling protocol to avoid interference, it doesn’t check if someone else is using the frequency before transmitting (remember, WiMax is designed to work in a regime governed by exclusive licensing or with no rules governing traffic management). Devices coordinate by knowing each other’s schedules. The FCC refers to this later type of system as a “restricted” contention based protocol, since it only avoids interference with devices using the same protocol.

The WiMax posse and friends really wanted to get rid of the contention based protocol requirement, or find a way to bless the WiMax-type restricted contention based protocol so that they could use WiMax in the band. A few of the more enterprising ones (surprise!) asked the FCC to explicitly bless their own proprietary solution as the anointed contention based protocol for 3.65 GHz devices. The FCC wisely took a pass on this. It allowed restricted protocols in the lower half of the band (from 3650-3675), while permitting “open” contention based protocols in the entire band. So if a group in a set geographic area voluntarily coordinates to use WiMax, they can use the lower half of the band without squeezing out the listen before talk people. At the same time, the listen before talk people can use the entire band and maximize use of spectrum available in real time.

Exactly Why Is This A “Licensed” Service?

It’s a stupid technical law thing. The FCC has always treated unlicensed spectrum access authorized under Part 15 of its rules (which is why devices using unlicensed spectrum are “Part 15 devices”) in a particular way. That includes the idea that once you get a certification, you are done and we don’t keep track of you. To continue to manage and exercise jurisdiction over the use of the band through a database and to require certain conduct by people using the band (and to avoid requiring the incumbent licensees to negotiate with mere “unlicensed” operators), the FCC classified this as a “licensed” service instead of unlicensed.

What Makes These Rules So Great?

All these rules address complaints by operators of wireless networks using unlicensed spectrum. First, they give unlicensed operators some high power options — particularly in rural areas for backhaul. The band does not have a resident population of cordless phones and other wireless devices that have made the existing open bands so noisy (hence the phrase “junk bands”), and the exclusion zones and the need to have base stations deployed make the band unattractive for consumer devices. (Would you want to make a cordless phone that won’t work in the top 50 markets and you can’t even guarantee it will work outside the exclusion zones unless it sense a base station?) While 3.65 GHz band does not rest below the 1 GHz “sweet spot” coveted by wireless providers (like the broadcast white spaces), it’s not a bad place to be. It has similar propagation and penetration characteristics as the existing 2.4 GHz band, and potentially provides operators with much needed “elbow room” — at least where the spectrum is available and not subject to exclusion.

Which brings us to the requirement to negotiate in good faith, but without an enforcement or arbitration mechanism. What good does this do, I’ve heard a lot of people ask. Well, it’s a good question and we’ll have to see. A lot of operators I have spoken to say that the biggest problem with coordination is (a) they never know where anybody else is, so they may accidently interfere with some else or suddenly find their own network subject to interference that hadn’t existed before; and (b) actually getting another network operator to return calls. Sadly, a number of people do take the attitude of “I don’t have an obligation to talk to you, so I won’t.” Will making them talk actually produce real results? Given that the alternative is failure to get a reliable network up, either from failure to reach agreement (you don’t coordinate with me, I won’t coordinate with you, and neither of our networks will work) or self-help (you cut accrues my network, fine, I’ll point my transmitters and blast at you until you agree to coordinate) I’m hopeful that a requirement to actually come together and talk will be enough to get operators to reach agreements.

The same is true of the requirement of existing licensees to at least talk to would-be operators in the exclusion zones. While licensees as a class may fight over the rules, experience teaches that individual licensees in non-competing services (and these licensees are not competitors) can be reasonable if you can show them that your system will not interfere with their system. Especially where the would be operator is a known and respected provider or a local government, I expect licensees will cut deals to allow 3.65 GHz devices to operate.

Who Wins and Who Loses?

Nobody really loses. Even the WiMax posse comes out winners, in that they can use WiMax in the band (in addition, there is an IEEE standards committee to come up with an “open” WiMax contention-based protocol that would allow use of WiMax in the entire band). True, they don’t get to screw their competitors. But they will actually get to manufacture and deploy equipment now, rather than continue to fight over how to divide this band up into little licensed bits.

The big winners short term will be folks primed to sell the equipment (Cisco, Tropos, and others) and those outside the exclusion zones able to deploy these systems. WISPs and muni systems in rural areas will have access to more (and cleaner) spectrum at higher power than ever before. The ability to do high power point-to-point links will prove particularly useful for folks connecting isolated farms or sparsely populated areas who need to get a backhaul signal to a fiber line a hundred miles away. Because the spectrum is open, and therefore much more affordable than licensed spectrum, we can expect to see small providers springing up to self-provision. With 50 MHZ to play with, you can expect to get some decent broadband speeds.

I predict the biggest winners among operators will be municipal broadband systems. The rules of this spectrum band line up very neatly with the network architecture needs of muni systems. High power where you need it, minimal interference from others, and mobile devices for public safety or even folks with lap tops once you’ve got the overall network deployed. Furthermore, because municipal governments are the kind of large, responsible entities that satellite receiver licensees can trust to stick to their network designs and respect the interference concerns of licensees, I also predict that local governments within the exclusion zones will be able to negotiate for use of this spectrum.

You said Commissioner Adelstein had some nice things to say about us?

As I have mentioned before, both Commissioner Adelstein and Commissioner Copps have proven themselves friends of community wireless networking, WISPs, and open spectrum generally. Over the last two years, I have had numerous meetings with Commissioner Adelstein (and his advisor Barry Ohlson) on this band, and he has been steadfast in resisting the blandishments of the WiMax posse because he believes in the power of these rules to help CWNs and WISPS.

Commissioner Adelstein took yesterday to emphasize the point about how this band can support the important work WISPs and CWNs do to bring broadband to America. As Commissioner Adelstein said:

During my time at the Commission, I have pushed for flexible licensing approaches that make it easier for community-based providers to get access to wireless broadband opportunities, and the rules we affirm for the 3650 MHz band should help make wireless broadband services available to a large number of new users. Today, we uphold our earlier decision to put in place a regime that doesn’t rely on first in time and provides equal access to all. I have heard from representatives of the Community Wireless Network movement about our 3650 MHz licensing rules, and they are thrilled with the hybrid approach and the positive impact it will have on their efforts to deploy broadband networks in underserved communities around the country.

That Was Nice. But What About The Future? Any More Open Spectrum On The Horizon?

A year and a half after I worried that no one would champion the cause of open spectrum, I think we have a better picture of what we will see. Clearly, Democratic Commissioners Copps and Adelstein recognize the importance of open spectrum and care about the contributions WISPs and CWNs make to our national broadband policy (such as it is). For example, they both had very nice things to say about open spectrum when the FCC made it’s Massport/OTARD decision last November.

But they are not in charge of the FCC. Kevin Martin is in charge of the FCC. As Chair, he sets the agenda. So what does the 3650 MHz decision tell us about Kevin Martin and Open Spectrum?

Martin has proven himself no friend to exclusive licensing at the expense of open spectrum, as some of us feared. On the other hand, he does not find open spectrum particularly exciting or compelling. Yes, it’s nice if it works, and if some heavy hitters like Microsoft or Philips show willing, he’ll push ahead even if it means stepping on a few incumbent toes. But Martin is no techno-ideologue like Powell, as demonstrated by Martin’s recent decision to terminate the Interference Temperature and receiver standards proceedings. Those proceedings had the potential to radically transform the wireless landscape by making spectrum broadly available to devices in real time. But with the entire industry opposed and only the public interest guys steping up to the plate, Martin decided to shut the proceedings down rather than let them linger forever in limbo.
(Which, given that he wasn’t going to move forward on them, was the right thing to do.) Importantly, perhaps, Martin shut those two proceedings down in a way that would let them come back again in the future. The Order terminating the proceedings simply found that the record at this time did not support the proposed rule changes and therefore the FCC terminated the proceedings. (Copps filed concurrences, joined on receiver standards by Adelstein, that noted this and expressed the hope that the FCC would continue to work on these someday.)

So I don’t think we will see anything really new or exciting on open spectrum for the foreseeable future, other than the pending White Spaces proceeding (which, I should add, has produced the only Fox News Editorialvwith which I have ever agreed). It is not that Martin is anti-open spectrum. If folks have improvements or tweaks on the rules I expect he will be happy to entertain them. But Martin is not an open spectrum believer like Powell. He won’t convene a spectrum task force to explore new technologies or new spectrum paradigms, and he won’t burn political points on untested technologies with no industry support.

Any Last Thoughts?

I will toss in a cautionary note that if operators can’t figure out how to play nicely together, the open spectrum movement (and the “eliminate the FCC because we don’t need it” movement), will take a major hit. If operators keep running back to the FCC all the time, refusing to cooperate, asking the FCC to arbitrate or resolve conflicts, then the FCC is going to decide that the pro-exclusive license camp is right – that someone has to play “traffic cop” and that, to paraphrase the Supreme Court, only by limiting access to spectrum for everybody can anybody use it productively.

On the other hand, if we make the 3650-3700 MHz band rules work, we will have taken an important step to reclaim the public airwaves for the public without the need to go through a handful of licensed intermediaries. It won’t prove the case conclusively, of course. Nor do I expect the existing licensees to ever say “Yup, you’re right, we don’t need exclusive licensing anymore, we’re turning in our licenses and giving up or cushy little government monopolies.” But it will provide a real proof of concept that a few simple rules and interference avoiding technology can exploit the potential of wireless far more efficiently and far more equitably than turning licenses into property.

So it’s put up or shut up time for those of us that believe the arguments of Benkler and Lessig and Reed that open spectrum can eliminate the need for licensing, generate wealth and innovation, and foster civic engagement and true First Amendment freedom of the airwaves.

Stay tuned . . . . .