FCC 700 MHz Decision: Just Not Good Enough

23 Comments

So much has been written about the 700 MHz wireless auctions, and there has been a lot rejoicing on part of Google and other consumer groups who oppose the incumbents. FCC Chairman Kevin Martin got his way, but I am not buying all the doublespeak coming out of the FCC, Google, or the incumbents.

They all say they want to spur the broadband competition, and build a third pipe. If they are actually serious about this, then how about giving the spectrum back to the people – i.e. make it unlicensed (like the spectrum used for Wi-Fi.) In doing so, the FCC would actually be ensuring that no single party owns this third pipe.

The innovation and competition that could emerge as a result would start an economic chain reaction with real long-term economic ramifications. Such an explosion could far exceed the short-term gains that come from auctioning off the spectrum.

The FCC wants to use the proceeds of the auction to help pay for the DTV set-top box upgrades, which is a worthy cause – but isn’t there another way to pay for those upgrades and reap more gains over a longer term? Another argument against unlicensed spectrum could be that there will be enough white space between bands that is going to be unlicensed. Will that be enough? I don’t think so.

Related Posts

* 700 MHz explained in 10 steps
* Inside the 700 MHz spectrum land grab.
* White Space, the new wireless frontier.

23 Comments

Richard Bennett

I don’t see much point in building networks that can only be used by people sitting outside away from trees, houses, and buildings. If you have a business case for one, I’d like to see it.

In the real world, WiFi networks have to either be contained inside buildings or deployed outdoors with external antennas and wires piercing the walls of buildings.

WiFi signals don’t propagate over the same distances and at the same speeds as 3G signals because the transmit power is so much less.

Who do you think you’re fooling when you claim WiFi is a replacement for 3G? It’s the most foolish thing I’ve ever heard.

Jesse Kopelman

Once outside of buildings or urban canyons, the range of 2.4 GHz WiFi is more like 500 – 1,000 ft, not 100 ft. Can you get 54 Mbps at this range? Of course not, but you can can get speeds comparable to 3G deployments. What most people describe as WiFi interference issues are not interference (i.e. C/N is too low) but too many contention events being triggered by other uses that are weak enough to be irrelevant to C/N but still strong enough to trigger a CSMA event.

Anyway, where does this fear of interference come from? There is no hidden scheduler problem. Other networks are just noise and noise is irrelevant as long as you can maintain sufficient C/N. That’s the my whole point of scheduling vs. contention. So you end up with 5mi radius cells instead of 20mi radius cells. So what? In the high population markets where carriers make their bread, you end up with smaller cells sizes anyway just to deal with capacity issues. In areas where you don’t have capacity issues there is plenty of spectrum to go around and you can stop worrying about noise from other carriers and grow back your cells more to the limit of path balance.

If 700 MHz were made unlicensed how many people would actually deploy high-power base-stations? This is especially the case for sparsely populated areas without the customer base to support multiple commercial offerings. It’s not like equipment and tower space are free. The truth is that if 700MHz were unlicensed the only change would be in low population areas, where a business case could finally be made for offering service. In high population areas user density would necessitate small enough cells to diminish interference concerns. Couple that with the reality that very few parties have the wherewithal to build a ubiquitous network even in potentially lucrative markets and where is the problem?

Richard Bennett

That’s a bizarre comment, Mr. Kopelman, far off the track. We’re discussing wireless interference, which is a more serious problem at higher signal strengths than at low ones. WiFi networks (and other unlicensed spectrum uses) are low power and low propagation, hence they achieve spatial reuse rather easily and require signal regeneration after very short distances. I don’t know if you’ve ever had occasion to deploy a WiFi network in a large enterprise or campus setting – I work for a company that builds such equipment – but if you had, you’d realize that WiFi signals roll off so quickly that you can’t get the highest throughput after 100 feet or so (even without any obstructions) in the 5 Ghz band, and after a few more feet in the 2.4 Ghz band. But that’s fine, WiFi is a local area network technology, it’s not a WAN or a MAN.

We employ different technologies for MANs and WANs, and yes, they may employ scheduled access to resolve contention, but that’s only part of the problem. There is also the problem of how schedulers interact with each other, and how to defeat hidden scheduler problems. These are problems that WiFi doesn’t address because they’re extremely difficult.

So to simply wave them off and hide behind your faith in the ability of engineers to defeat fundamental laws of physics is to invite the regulators to behave in such a way that nobody gets to use the 700 Mhz bands for any use more sophisticating than opening a garage door.

But that may be what you want.

Jesse Kopelman

Richard, as a Real Wireless Geek I call shenanigans. What exactly are these unresolvable issues? Thousands of existing installations based on unlicensed spectrum would like to know. The issue with WiFi is the contention based MAC, not unlicensed spectrum. I’ve seen real examples of people getting macrocellular style coverage using unlicensed spectrum and scheduled MAC equipment. Yes, cell sizes will be smaller than they could have been with the lower noise floor that comes with licensed, but so what? All the money saved on spectrum and overpriced carrier grade equipment can pay for some extra base-stations. Anyway, look how the FCC is channelizing this spectrum — mostly 12 MHz and smaller chunks. You’re not going to be able to have giant cells with the tight reuse only 12 MHz of spectrum entails. Maybe you need to step out of your research and standards ivory tower and talk to some people who actually deploy and operate wireless networks. Om, you are 100% correct here.

eas

Tom Coseven, wouldn’t rural areas have less interference, and therefore couldn’t transmission occur over longer hops?

Richard Bennett

Om says: “Guys, the spectrum-slice is big enough, and the interference issues can be resolved.”

Until we know how to resolve them, it’s irresponsible to make this claim. I’ve been inventing protocols for wireless networks since 1990 and I don’t know how to do it, and neither do any of my colleagues in the research and standards community. You may as well be saying we don’t need to worry about Global Warming because some day the geeks will give us Cold Fusion.

Listen to the real wireless geeks, Om, not the uninformed, faith-based speculators.

Martin

Preach it on the mountain, brother Om! It’ll be interesting to see how stringently the FCC enforces the new “Carterfone” rules they’ve applied to the spectrum, or if it’s just more Kevin Martin-style Kabuki.

And ignore the cynics here. We can make the technology workable. It won’t be perfect, but nothing ever is at first. What matters is the political will and the influence to make it happen. We’re getting more of the former–now all we need is the latter.

Google’s gambit was a start–now we need to build on that.

Tom Coseven

The problem is Om and his friends have urban myopia.

Many of the problems of unlicensed spectrum can indeed be solved economically for dense urban areas where the metro backhaul infrastructure has excess capacity (or where the population is willing to fund additional backhaul with tax dollars).

South Dakota and Colorado have two Senators also, and the FCC is required to represent the interests of all regions of the country. A licensed wholesale model may have worked (if the wholeseller could be forced to build out rural areas in a reasonable time frame, and you could figure out how to regulate the wholesale pricing), but recommending unlicensed shows an ignorance of rural and suburban economics.

dwilbanks

Wow, I love the idea.

4.6 Billion dollars is a huge, huge, huge amount until you think about it in terms of $16 for every Man, Woman and Child. The FCC is giving it away for ULTRA-Cheap.

Nima Negahban

I think they’re are two perspectives here and Om is talking about one , and the comments are talking about the other.

Om is speaking about the dangers of increasing corporate governance over our lives by giving them even more spectrum.

The posters are talking about the technical ramifications of having it unlicensed.

I agree with Om on this , technical problems can always be engineered around. I would much rather have a public technical problem on my hands then being at the mercy of one private body.

Om Malik

Guys, the spectrum-slice is big enough, and the interference issues can be resolved.

What cannot be resolved is that once the spectrum is licensed to one of the big three – AT&T, VZ or Google, we are almost always are going to be at their mercy.

It is the choice between lesser of two evils, though I have faith in technologists’ ability to come up with viable solutions.

Sure, I am a little idealist about this, but it this whole saga is not very confidence inspiring.

ValleyGrunt

Om:
For a gent with broadband reporting background, you are a bit disappointing here. Unlicensed usage would lead to it being a RF junkyard, especially at that frequency; a zillion different signals colliding with each other making it quite unusable.
It has to be licensed frequency, for power and QoS reasons

Rick

They should have mandated the 3 sections of 700 spectrum open. That would have created 3 competitive pipes; enough for a market.

Richard Bennett

With the exception of Barlow Keener, your readers are making sense, Om, and you should think about what they’re telling you. WiFi was made good use of unlicensed spectrum at 2.4 and 5 GHz for Local Area Networks. The regulations that govern this spectrum effectively limit transmissions to a hundred feet for the highest data rates. You can’t build a competitor to Cable Internet and DSL out of hundred foot hops. In order to increase the range, you have to increase the power and that limits the number of networks that can be independently deployed and managed in any given area. That’s why Muni WiFi is proving impractical.

If you want an alternative to cable and DSL, the 700MHz bands must be higher power and that means they must be licensed. We don’t have the technology today to do it any other way.

Sorry, but that’s reality and you ignore it at your peril.

David Mackey

I have to agree with several of the posters – wouldn’t having the 700 Mhz unlicensed cause a lot of interference?

Tom

Yes, and turn the 700mhz into a chaotic frenzy as the public goes hog-wild with the spectrum? What’s going to promise that the people will have a usable network in a years’ time?

Barlow Keener

In 1981, the FCC gave away for free to the monopoly telephone providers the spectrum now being used by the by the big two wireless companies (wireline). Soon after the FCC gave away for free the spectrum to the non-wireline companies – these companies had to prove they were in the wireless business which meant the paging companies (non-wireline). Although Congress decreed that the 700MHz spectrum is to be auctioned by the FCC by January 28, 2008, it would make common sense to turn 700MHz into unlicensed spectrum. Look at how 2.4GHz has been sliced, diced, used and reused. Almost every home and business makes use of this unlicensed spectrum. That is a lot of equipment built and sold and who knows how many tax dollars have been generated by 2.4GHz (Just look at the WiFi chipset in your laptop). Giving away 700MHz could generate lots of tax dollars in the near term and give Americans another choice for broadband.

DEC

WiMAX has both: unlicensed and licensed. You won’t see much commercial development in the unlicensed bands due to interference. Your cordless home phone works in a 2.4GHz (unlicensed) band, and the microwave (and other phones) can wreak havoc.

Aswath

I am most decidedly not a radio person, but is it feasible to operate a “long-range” frequency in an unlicensed form?Shouldn’t the access be controlled, lest even people who are geographically far away step on my frequency? It is a sincere question and not a rhetorical one.

Comments are closed.