Credit: CBS

There was a bizarre column in the New York Times this week about net neutrality that managed to mangle the issue as severely as anyone has (Net Neutrality and Economic Equality Are Intertwined.) The author, Eduardo Porter, is the paper’s economics writer, but most of the column deals with technology issues rather than economics. It illustrates what happens when people opine on tech policy without understanding the “tech” part.

Let’s start with his metaphor:

Imagine a network of private highways that reserved a special lane for Fords to zip through, unencumbered by all the other brands of cars trundling along the clogged, shared lanes. Think of the prices Ford could charge. Think of what would happen to innovation when building the best car mattered less than cutting a deal with the highway’s owners.

In the world that I live in, California, there actually are special lanes on the (public) highway for carpoolers, low-emission vehicles, impatient commuters willing to pay a toll for faster service, police, and emergency vehicles. Originally, these special lanes were only for carpoolers, but there is so little carpooling that the public policy arguments for additional services were accepted by the state legislature as compelling enough to warrant inclusion. If I’m running late to a meeting in Santa Clara, I’ll pay the toll to use the special lanes and be happy to do it. If I’m in less of a hurry, I’ll happily save two or three dollars to enjoy the music in my car a little longer.

This is not simply a quibble with the writer’s analogy. The cellular network effectively reserves special lanes for voice and text messaging apart from the slower lanes it has for data. It does this for good reasons. Voice needs lower latency than web data, and less variation in the deliver time between one packet and the next. The ear is challenged by the variable nature of mobile communication under the best conditions, as signal quality inherently drops as we move farther from the tower that handles our calls and handoffs between towers are often accompanied by the loss of a packet.

Voice is low bandwidth, so it’s a reasonable tradeoff to boost its quality at the expense of high bandwidth data. We also pay more per packet for voice than we do for data, again a reasonable trade because we appreciate clear calls. Text messaging costs more than e-mail on a per-byte basis, but it has the advantage of being inherently spam filtered so again there’s a rationale for the fee.

So the analogy is neither apt nor persuasive because it confuses the quality requirements of services of different kinds with discriminatory pricing schemes for a services of the same kind. Economists are supposed to know that all commodities are not equally valuable.

He then jumps from the notion of rational price discrimination between disparate services to the criticism of irrational price discrimination based on deal-making capacity in connection with some of the complaints Netflix founder Reed Hastings has been making recently about the treatment of its Internet-based video service by ISPs:

“Net neutrality has broad consumer and voter support,” Reed Hastings, the chief of Netflix, said in an interview. “It is important for the sake of public access that the rules apply equally.”

This is quite a leap. There may be many policies and entitlements with broad consumer appeal that aren’t sound policy. Consumers may favor free ice cream, especially young ones, but it’s neither sound economic policy nor a particularly good use of the personal calorie budget. There’s also the annoying fact that most voters have no idea what net neutrality means and why they should care about it. So we’re left with another unsound argument in support of a unsound premise.

Hastings asks an intriguing question:

“If I watch last night’s ‘S.N.L.’ episode on my Xbox through the Hulu app, it eats up about one gigabyte of my cap, but if I watch that same episode through the Xfinity Xbox app, it doesn’t use up my cap at all,” Mr. Hastings wrote on his Facebook page. “In what way is this neutral?”

This echoes Susan Crawford’s complaint that some over-educated people were trying to drag the Xbox issues “down into the weeds of network architecture technicalities,” but I have an answer to share anyway. When Tim Wu first wrote about the notion of net neutrality, he was very clear that he wanted ISPs who also offer non-Internet services such as TV programming and voice to be neutral at their Internet gateway with respect to Internet-based services, and he was also clear that there needn’t be an assumption of equality between services provided within their private networks and those provided over the Internet. This is a simple distinction that should be easy to grasp: If the service is provided solely within a cable network, it’s regulated by the Communications Act under the appropriate title, but if the service is provided over the Internet it falls under the FCC’s Open Internet rules. There is no assumption in law or policy that services that depend in the interconnection of multiple private networks must be competitive in all times and places with those provided solely over one private network.

There’s a sound rationale for making a distinction here. The cable network was designed to deliver television programming to the mass audience, it’s by far the most efficient means in the universe of doing this. It sends one copy of SNL to every subscriber, not a personalized copy to each interested viewer, so there’s no popularity penalty built into it as there is on the Internet. If you’re the kind of viewer who sometimes watches SNL, you can record it on your DVR and watch it anytime you want without any sort of cap or limitation, and indeed without even paying for Internet service.

If you don’t have your own DVR or you forget to record SNL, you can use an “On-Demand” service provided by the cable company that effectively uses their DVR to watch it at your convenience, once again without paying for an Internet service. If you’re a Comcast customer with an Xbox, you can use your Xbox to access the Comcast DVR as well, once again without paying for Internet service. Comcast and the other cable companies can do this because they know what the most popular shows are, they have licenses to these shows, and they have the ability to populate their cable network with stored content with a large audience.

This isn’t comparable to the Netflix service that provides individual consumers with highly idiosyncratic content that generally does not have mass appeal. Netflix is working the long tail, while Comcast is serving the mass audience. Porter confuse the very different nature of these services with the market conditions around their delivery systems:

The emerging dispute between Netflix and Comcast underscores the core weakness of the Internet economy. To reach the multitude of online services competing for your attention, you must first get through a bottleneck that is not competitive at all: high-speed broadband access…

Costs are higher when there is little competition. If only 43 percent of American households with income under $25,000 a year have wired access into the home, it’s because most of the rest cannot afford it.

The issue of bottlenecks is less significant technically than the fact that not all services are equal from the standpoint of audience appeal, marketing, and licensing. As much as many of us love the Internet, the mass audience of actual consumers holds it in lower esteem than their cell phones and cable TV services. There are more cell phone subscriptions in the U. S. now than there are people, and substantially more cable TV subscriptions than Internet service accounts at the home:  85% of American homes pay for cable or satellite TV, but only 67% pay for Internet services. This isn’t a question of cost, because Internet service is cheaper than a nice cable TV plan with HBO. As Matthew Yglesias wrote in Slate’s Moneybox economics blog recently, “People Just Aren’t That Into the Internet:”

The problem for people who do want to watch all their TV over the internet is that to provide enough video content to everyone for that to be the standard way of doing things, you’d need much more broadband capacity. And we could build much more broadband capacity, but people would have to want to buy it. And at the moment, it seems like people don’t really want to. Of course they would want to if cable television stopped existing, but all the infrastructure is already there. Now maybe aggregate population preferences will change over time. There’s certainly some evidence that they’re shifting a bit. But hard as it is for web junkies to remember, lots of people seem perfectly happy checking Facebook on their phone.

So the issue is that cable TV is not only more efficient that Internet TV, it’s also more popular, so the economics that may some day make Netflix competitive on a technical basis with with ABC’s Tuesday night lineup delivered to your DVR just aren’t there.

Porter also makes the usual defective comparisons between U. S. triple play offerings the low price packages offered by Free.fr in France over the DSL infrastructure that was put in place by the formerly government-owned phone company. Free is a price leader that only offers services that Americans would consider competitive in a limited group of large cities, and only has fiber in selected neighborhoods in Paris. Free’s efforts to buy market share in mobile service started strong, but have now hit a wall as users discover the service quality to be quite poor:

In addition, each week around a thousand customers who had switched to Free rejoin Orange, SFR or Bouygues Telecom. So it’s become a numbers game — how can Free balance growth and stop the sort of churn that could undermine its strategy right from the start?

So what we have in the spat between Internet-based and cable-based providers is a conflict between technologies that can only be resolved in the Internet’s favor by changing its network architecture to something more friendly to high-value content, mass distribution, and consumer choice. The alternative proposed by Porter, Wu, and Crawford is to force TV consumers off cable and onto the Internet by imposing irrational price discrimination on cable TV.

There are many reasons to like the Internet and many ways to promote it, but using brute regulatory force to mask the technical differences between Internet and cable TV strikes me as very destructive and not at all consumer-friendly. It’s also massively ignorant of those dreaded “network architecture technicalities,” but nobody ever said technology has to be easy to understand.

The pervasive misunderstanding of how we use — and could use — the electromagnetic spectrum was eloquently addressed in layman’s terms by The Myth of Interference, a 2003 Salon article by David Weinberger that I wish every reporter, legislator and wireless policy “expert” would read. The Myth charts a sensible path towards a physically veridical, technology based approach to the use of the electromagnetic spectrum. Central to The Myth’s argument is the fact that current policy, based on the licensing of frequency bands, was shaped by the technology of the 1920s. Today, as The Myth explains, the technologies of wireless communications are vastly expanded, but regulations are still stuck in the 1920s. And the incumbents want to keep them there.

The bad news is that the New York Times article doesn’t cleanly separate myth and analogy from physics and technology. The article mixes literal, physical reality with analogies that foster certain monopoly-preserving conclusions. The article fails to distinguish which is which. It talks about “slices of radio waves,” it confuses licensing with ownership and it treats Wi-Fi offload as if Wi-Fi were an alternative to the use of the electromagnetic spectrum.

Even the headline — “Carriers Warn of Crisis in Mobile Spectrum.” — isn’t what the story is about. A more accurate headline would have said, “Experts Question Carrier Spectrum Claims.” The lede is buried in the fourth paragraph: “But is there really a crisis?” The article misses key specifics, such as the fact that in the (now failed) 36 billion dollar T-Mobile merger with AT&T, the spectrum was valued at $6 billion, and the fact that only about 10% of cell towers have fiber optic backhaul, which means that cellular data capacity can be limited by old DS-1, DS-3 and microwave backhaul technology rather than by wireless electromagnetic receiver overload.

The result is journalism that’s (perhaps unconsciously) shaped by the industry the journalism purports to question. It is as if the cellcos claim 1+1=3, while a few dissident experts — Martin Cooper, David Reed and me — protest that it is 2, then the true answer must lie somewhere in the range between 2 and 3.

Don’t get me wrong. I’m glad that the Times is beginning to explore the idea that the spectrum-as-real-estate analogy is a regulatory fiction.

As I have problems with the notion of infinite spectrum and with Reed’s and Isenberg’s claims to “spectrum expertise” I left the following comment.

I’m not aware of any peer-reviewed academic paper that argues that the information-carrying capacity of RF spectrum is infinite. If there is one, can you please provide a citation?

I’d like to see how to design a system of that allows an infinite number of transmitters to communicate with an infinite number of receivers on a particular radio frequency in a given location without interference (the impaired ability of receivers to extract bits from a radio signal,) or simply a large number of transmitter/receiver pairs. In particular, I’m looking for a coding system that suspends Shannon’s Law.

I’d also like to see information about the credentials of the so-called experts on spectrum cited by this article, as I’m not aware of any academic papers, patent filings, or products that exploit the infinite spectrum concept that can be attributed to the three experts (or to anyone else.)

We know that Cooper was one of the eight inventors listed on Motorola’s cellular patents in 1975, and also that Cooper was a co-inventor of ArrayCom’s SDMA patent from the early ’90s, but ArrayCom has not been successful in creating any infinite spectrum systems. SDMA has extensibility limits, as does CDMA. They both achieve simultaneity by subdividing the code space, and this approach does not go to infinity.

As far as I can tell, neither Reed nor Isenberg has worked with actual spectrum systems in either a research or a product development setting, but I’d love to see evidence to the contrary in the nature of academic papers, patents, or products.

Finally, I don’t understand why the claim is made that the carriers invented the spectrum crisis. The National Broadband Plan forecast a spectrum crunch two years ago, and not only was it not a carrier product, Isenberg was a member of the team that produced it.

It seems a bit more prudent to base spectrum policy on science than on the speculation of non-experts writing in the popular press. Call it an elitist bias if you wish, but I think science is worthwhile.

So far Isenberg hasn’t responded.

Last week the House Technology and Innovation Subcommittee (of the Committee on Science, Space, & Technology) held a hearing on the spectrum crunch in an effort to get to the facts on a subject that’s become strangely controversial. As a witness, I filed written comments and delivered an oral summary on the key points. This being a geek committee hearing, I had PowerPoints, which you can see on the webcast.

The day of the hearing, there was a rather bizarre little story in the New York Times on the spectrum crunch on which the writer, former Wired staffer Brian X. Chen,  turned to some colorful sources who aren’t up-to-date on the issues. For a very long time, various people have hoped that radio technology would permit multiple users to use common frequencies at the same time. We can do this partially (the CDMA technology in all 3G and 4G phones permits this kind of sharing,) we can’t do so without a loss of performance. (For a complete breakdown of what’s wrong with the article, see my post on Innovation Files; for the technical issues, see my testimony.)

It turns out there’s still not much understanding in policy circles about the multiple ways in which users share resources on networks. In fact, the fundamental distinctions between networks of different types such as the traditional telephone network, broadband networks, the Internet, and the mobile network relate to the manner in which users share network resources.

It’s fashionable in wireless policy these days to tout the benefits of “Dynamic Spectrum Access,” a form of network sharing in which the network and the connected devices hunt for unused spectrum before doing what all wireless networks do. The alternative is for the network to chose a frequency according to its license. It’s important to understand that DSA is an alternate form of licensing, not an alternate form of networking. You want DSA when there’s been a massive failure in spectrum licensing that you can’t correct any other way than to grab what looks like unused spectrum and temporarily erect a network on it.

This is something like searching your neighborhood for an unused car every time you want to take a trip. You may find one that hasn’t been driven for a week and take it just as the owner is heading for the door. All spectrum below 3 GHz has an “owner” already (except for explicitly unlicensed spectrum) so there’s always this potential issue with grabbing idle spectrum. There are additional issues related to services such as GPS in which the transmitter is very faint and far away and all the local users are passive receivers.  LightSquared wasn’t allowed to use spectrum in the general vicinity of GPS because of finicky receivers that can’t observe their licensed boundaries.

Commercial networks are good at sharing because they’re able to assume control over swathes of spectrum in which they operate and schedule transmissions in a scenario where the network knows who wants to do what and when. The typical cell tower is shared by 1000 users and millions of applications. The typical range of spectrum is used by less than 1,000 licensees across the entire nation, and most of them run a single application. So the efficient way to share spectrum is the way the commercial networks already do it.

For DSA to become generally useful, two things have to happen:

• We need technology that improves on the simultaneous spectrum sharing technologies we currently have (such as CDMA, SDMA, and MU-MIMO)
• We need to replace legacy receivers (such as the low quality GPS receivers that don’t tolerate Light Squared) with smart receivers that do the things that Brian X. Chen’s contrarian visionaries would like to see.

As I told the panel, “we’re not there yet” but we probably will be in 10 -  20 years if we start working on the problem seriously today.

Will we?

I don’t want to give away my testimony in advance, so let’s just say I’m looking forward to a lively hearing with diverse viewpoints in play. Federal support for spectrum research is obviously an important issue going forward, as are more robust receivers and a more dynamic system for reassigning spectrum to new uses.

A Solution in Search of a Problem

The manufactured controversy over TV-on-demand through the Xbox 360 is more entertaining than enlightening.

DNS Integrity in the Real World

Those who followed the SOPA and Protect IP copyright law debate in December and January will recall an argument raised by certain members of the tech sector to the effect that enlisting the Internet’s Domain Name System (DNS) in the fight against pirated goods would undermine Internet security.

Freeing Federal Spectrum

The NTIA released its report on clearing the 1755 – 1850 MHz spectrum of government allocations today, and it’s good news and bad news.

Spectrum Policy Collides with Competition Policy

Today’s Senate Subcommittee on Antitrust, Competition Policy and Consumer Rights hearing on the proposed transfer of fallow spectrum from cable companies to Verizon sheds light on how poorly traditional competition policy fits the networking business.

The background is somewhat complex. The FCC’s last big spectrum auction took place in 2008 when the “digital dividend” freed up some airwaves that had formerly been used by television. Digital TV channels can be placed closer together than analog channels were, so a more efficient packing scheme made this spectrum available for sale. The spectrum was arranged in five blocks, called A-E, in two ranges, low and high. Most of the blocks consisted of pairs, separated to allow transmission on one half of the pair while the other half was doing reception, but the E block was unpaired. The D block was not successfully auctioned as the FCC wished to sell a single nationwide license for it and the reserve price wasn’t met but it has since been given to public safety.

Here’s a handy map that shows how the pairing works.

Note that the A block consists of 6 MHz next to Channel 51 at 698 – 704 MHz and another 6 MHz from 728 – 734 MHz and that the E block is a single slice without a pair, and that there is C block spectrum in both the lower band and the upper band, with the upper band slices twice as wide as the lower band slices.

The A-C blocks sold for wildly different prices because the A and C blocks had significant restrictions: the A block was directly adjacent to active TV transmitters on Channel 51 is most urban markets, and the FCC imposed artificial net neutrality restrictions on the C block in accord with the fashion of the time. The average prices by MHz per million population (“megahertz pop” in the lingo) were:

The biggest winner of B block spectrum was AT&T, the biggest winner of C block spectrum was Verizon, and the A block was mainly won by regional networks such as MetroPCS, US Cellular, and Cellular South. AT&T paid a significant premium to be free of the net neutrality rules and the interference caused by the high power TV transmitters on Channel 51 in the urban markets, and the regional carriers who could live with channel 51 got a discount. Verizon did best of all by accepting the net neutrality rules. So the assumption of flexibility played a big role in determining the auction price.

Here’s a map of the Channel 51 transmission contour.

Spectrum is harmonized around the world according to “Band Classes” devised by the 3GPP, the standards body that defines such things as LTE, the new 4G standard that’s hitting the U. S. market now in a big way and starting to appear in the rest of the world in a much smaller way. There are three band classes of interest for 700 MHz, identified in the first diagram as BC 12, BC 17, and BC 13. Note that Band Class 17 is a subset of Band Class 12 that excludes the discount A Block, and BC 13 is distinct and non-overlapping with classes 12 and 17.

At this stage, AT&T plans to resell devices conforming to Band Class 17 and Verizon to resell devices conforming to Band Class 13 (in the upper C Block ). These devices will be able to use their native, licensed networks only, which means they won’t be capable of roaming onto other networks (except insofar as these devices may support other frequencies as well.) Hence the notion of “interoperability:” 700 MHz devices will not roam or “interoperate” with other band classes and networks but the ones they’re built for.

This irritates the small carriers who bought A Block spectrum at a discount because they would like to use the same devices that AT&T and Verizon resell rather than more specialized devices tuned to their A Block frequency and also capable of roaming onto the B and C blocks. Cellular South (now known as “C Spire”) is the only regional network to offer the iPhone to its customers so the entire group of A Block winners is somewhat disadvantaged in terms of the very best devices, but there are a few Android devices adapted to their networks. MetroPCS offers LTE today with such devices. Chips are available to support Band Class 12 so there is not an insurmountable technical hurdle to building Band Class 12 devices. Making them work well is a different matter, however.

Leaving aside the question of the propriety of the FCC essentially requiring AT&T and Verizon to subsidize handsets for the A Block carriers and focusing in the technical details raises some interesting issues.

We learned from the LightSquared issue that it’s never good to be dependent on a low power signal when you have a neighbor who uses a high powered one. Even though the signals are distinguishable from each other in terms of their patterns of digital bits, the radio energy of a high power transmitter confuses receivers designed for low power signals. As radio waves decay with distance, they give off interference energy above the frequency of the original signal, and this can be significant when the power difference is great between the lower and higher powered transmissions.

When multiple transmissions interact, receivers can experience “Inter-modulation” (IM) distortion, defined by Wikipedia as:

…the amplitude modulation of signals containing two or more different frequencies in a system with nonlinearities. The intermodulation between each frequency component will form additional signals at frequencies that are not just at harmonic frequencies (integer multiples) of either, but also at the sum and difference frequencies of the original frequencies and at multiples of those sum and difference frequencies.

If that’s not clear, here’s a diagram:

The diagram shows IM distortion as the two smaller spikes the left and right of the two big spikes that represent the signals. The IM spikes in this example are significantly stronger than the background signals represented by the more solid lines.

There are a few ways to work around IM distortion. The easiest is to raise signal power, which is accomplished in cellular systems by siting towers in a ring around the IM distortion source and by increasing the battery draw in mobile devices. There are limits to this approach because towers are expensive and cellular systems are very low power compared to those TV transmitters on Channel 51.

Engineers hired by the regional carriers seeking the device subsidy claim that three towers close to each TV tower will do the job, but AT&T’s engineers put the number closer to 12. Another way is to add filters to the devices, which raise the cost and increase the battery drain, and yet another is to use more sophisticated signal processing, which once again reduces battery life. All of these methods require extensive field testing, so there is a significant overhead in terms of the time to market for new devices.

This brings us to this question: Is it reasonable for the FCC to add expense to the smartphones that AT&T and Verizon sell, to reduce their battery life, and to delay the introduction of new devices built by Apple, the Android crowd, and the Nokia/Microsoft partnership in order to enable roaming between regional networks and national ones? “Reasonable” is in the eye of the beholder, of course.

If you’re a regional network, the proposed interoperability rule costs you nothing and impairs the users of the national networks, so you’re happy. If you’re a national network, the rule increases your costs and irritates your customers so you’re not happy. And if you’re a device manufacturer it impairs your ability to get new devices to market so you’re not happy either. The “interoperability” mandate will also stress the analog engineering skills of the device makers, and that’s an area where they don’t need any more problems. Analog engineers are in short supply, which is evident every time a smartphone shows poor antenna performance.

The cheapest and easiest way around this problem would be for the FCC to adopt the same solution they went for in the LightSquared case: They can take Channel 51 off the air. With the interference source gone, Apple can simply build all of their 700 MHz devices to function on the A, B, and upper C blocks without special testing and engineering for the A Block. If the FCC doesn’t want to do this, we’ll have to evaluate whether their reasons for keeping Channel 51 alive are more compelling than the reasons the device manufacturers have for not wanting to filter the interference it spills into the A Block.

As it stands, the A Block licensees have the power to buy as much interoperability as they want from the companies that build their smartphones. They’re going to pay higher prices for these Swiss Army knife phones than the more narrowly tailored phones used by the national carriers, but they got a deal on their spectrum.

The FCC is looking for precise estimates of the costs of an interoperability mandate, but they’re only part of the story given the rationale. The underlying assumptions seems to be that the consumer buys a smartphone and keeps it for a decade, roaming at will and changing carriers every time a great deal is available. This is clearly not the way the smartphone market works today, or you wouldn’t see people camping out at the Apple store to get the new iPhone.

I fear this is simply “Wireless Carterfone,” an attempt to re-live the glory days of 1969 when the courts and the FCC required interoperability for the telephone network. While that decision lead to cheaper and more plentiful fax machines, modems, and answering machines, it’s not really parallel to the situation we have in the rapidly-changing world of cellular technology. We have to think about how this mandate will affect the roll-out of 5G and 6G services as well as faster and better smartphones even if we can convince ourselves that it makes sense to have the national carriers subsidize the regionals.

First, I made a mistake in that post.  I said the hearings would take place on February 23^rd.  I misposted – they’re going to take place after February 23^rd, a period of time that includes time immemorial, but may prove to be soon.

Aside from that correction, I stand by last week’s remarks, in particular the idea that these hearings will be important because they’ll tell us whether the Senate can spot competition when it sees it.  Specifically:

 “…many observers see broadband space as a series of ”stovepipe” or staccato markets – firms specialize and compete in providing signal, or in manufacturing devices or their operating systems, or in social media or other applications, and their competition is narrowly limited to that segment.”

The problem, in part, is that advocates look at the broadband sector and they see telephones, particularly since they both involve communicating and, more importantly, are subject to the dictates of the FCC.  And the old Ma Bell system and the broadband Internet are easy to confuse, as one delivered a monochromatic, static dial tone and the other is a platform for a competition among a burgeoning number of devices and services that are changing every day life, so if you’re not careful – or observant – you might find yourself  thinking about broadband and the Internet in “telephone” terms – counting the wires to the home, or presuming that whoever provides signal provides everything that is somehow related to that signal,which is like arguing that the electric company is going to monopolize hair driers and refrigerators.

In an effort to demonstrate these fallacies, John Bergmayer, a senior staff attorney at the advocacy group Public Knowledge, last week editorialized that the FCC and whoever else is listening should nix the Verizon-cable deal, because it would be anti-competitive.    I’d make three points about what Bergmayer has to say.  The first is that his argument is garbled – it switches course almost in mid-sentence – particularly on whether an Ethernet connection and an LTE connection compete.  The second is that he doesn’t get that the broadband sector, in practice, is more than a series of “stovepipe” or staccato markets…to reuse the quote I excerpted a few paragraphs ago.  And third, and perhaps most  Important, is that what he contemplates, or seems to contemplate, as a solution is to turn broadband into — you guessed it – the old Ma Bell system; after all, when he looked at broadband, he was seeing Ma Bell and the FCC, right?

Let’s start at the top.  Bergmayer laments the death of “facilities based competition,” meaning there are too few options for getting signal from where it starts to your device (or as they say in agricultural economics, from “moo to you”).  Well, there’s cable, and in some places fiber, and in others DSL conveyed by the old phone lines, although everyone understands it’s not as good as the first two (as Bergmayer says, “the physics doesn’t allow it.”  And then, of course, there’s wireless, which is flooding your home with evermore powerful signal, as do “4G’ technologies such as LTE, which are not as fast as a cable or fiber hook-up, but are now as fast or faster than the DSL that Bergmayer regards as inadequate.  (And can I count the prospect of being able to move from WiFi island to WiFi island as these appear with growing frequency?  Probably not in Bergmayer’s book.)

Not so fast.  Because wireless should be disregarded when we assess how you get signal, says Bergmayer.  Why?  For one, it’s not fast enough – he says, although it’s obviously gaining rapidly on wireline options and for large parts of the population, works perfectly well.  In fact, advocates always disparage mobile technologies until they have to admit they got it wrong – check out this piece from 2004 when the Consumer Federation said that mobile hasn’t eroded “Bell market shares” and VOIP is “nascent at best” before the first paragraph ends.

But the more damning evidence is that, as Bergmayer says:

If mobile wireless was a substitute for wired we’d see large numbers of people dropping one for the other. But people who can afford it tend to have both. It’s fair to note that wireless has substituted for wired telephones for millions of people, but this is a fairly low-bandwidth application–there are no hopes in the near future for a mobile broadband wireless service that affordably matches all of the performance characteristics of cable or fiber and can sustain the same sort of use.

OK, let’s parse that.  Sure, we now all agree that mobile and landline telephony compete –  but as wireless via 4G becomes more powerful, should we expect this competition to spread?  No, and to prove it, “people who can afford it tend to have both.”  Yes, and people who can afford it have two cars, even two houses, hell, I have a friend with two tuxedos – I mean, isn’t that the height of something, two tuxedos? – but the cars and the houses and the tuxedos compete.   Having both tells us that 1) they’re relatively cheap, and 2) they allow the user to mix and match the services she’s consuming.  So are wireline and wireless the “never the twain shall meet” affair that Bergmayer posits, or are they evolving competitors whose competition is continually moving up the ladder of more highly-valued tasks?

In fact, isn’t Bergmayer’s entire essay self-contradictory in this regard?  If wireline and wireless don’t compete, as he asserts, then who gives a rodent’s behind whether Verizon wireless and a bunch of cable companies are cross-selling services?  In Bergmayer’s world, that’s like going to the tailor and being able to get your hair cut or your  oil changed.  I saluted Senator Kohl – a great Senator and a great Brewers fan –   he owns a piece of the Club as well as the NBA Bucks – last week for being able to recognize what Bergmayer doesn’t – that the only reason to think about the Verizon-cable deal in the first place is because they compete.

But more fundamental is Bergmayer’s refusal to leave the Ma bell world and see the broadband market as it is – a “cage match” in which connectivity companies (both wired and wireless), device manufacturers, operating system developers, and application and service providers continually compete and partner with each other, forming and reforming relationships to capture the bulk of the value created by the integrated broadband experience – the one that brings you cloud-based applications over wireless signals to devices that did not exist in the lifetime of a child not yet in school.  The world left Bergmayer’s view behind the day the iPhone was introduced and the world has never looked back.  Electricity was the platform that gave rise to hair driers and refrigerators.  But signal is not only the platform on which devices and applications sit, but the signal itself competes with those devices and applications for the consumer’s allegiance.  Do you have an iPhone so that you can use AT&T or Frontier or Comcast – or do you have Cox or Verizon or Time Warner in order to have marvelous devices and their applications?  When it comes to wagging, which is the tail and which is the dog?  Let me try it this way – who’s more likely to tell the other to piss off?  Any one signal provider to any one smart phone maker?  Or the other way around?  That is “cage match competition” – a struggle in which Verizon and AT&T, Comcast and Cox, Dish and DirecTV, Apple and Microsoft, Google and Facebook, Amazon and Twitter all compete to be the platform on which the others rest.  And the result is incredible innovation, new products and services, mutual price discipline, and a burgeoning adoption rate.  If this sector wasn’t competing, what would be better if it was?

And that’s the last point – so what?  What if Bergmayer’s right?  What do we do then?  I have great sympathy for this question.  My wife and I are building a retirement place in West Virginia and we have hillbilly broadband as provided by Frontier Communications, which is the official signal provider in zombie movies and coma wards.  Junkies get the needle to move faster than these guys.  But at least they charge what other providers of comparable service charge because 1) there are anti-trust laws that prohibit predatory behavior on poor but deserving homeowners in remote locations and, 2) there’s always mobile.  Hell, I’d love a grown-up, big-boy, long-pants, high-speed connection, but I’m in a very un-dense location that doesn’t justify much more than what I get, and it’s hard to see why policy ought to subsidize me beyond that.  But Bergmayer lets us know where he’d like to go:

Broadband unbundling or open access rules like those common today overseas and the system that allowed thousands of competitive dial-up ISPs the flourish in the 1990s are examples of these sorts of policies. Rules about roaming would be another example.

Wow.  “Unbundling and open access rules” mean that a company builds connection infrastructure, and then has to share that infrastructure with its competitors at prices determined in a regulatory hearing.  It’s The Little Red Hen turned on its head – in which the Fox and Turkey Lurkey and the other lieabouts not only eat the bread the Hen bakes, but probably the Hen herself.  And roaming rules are a variant of that – as I’ve discussed earlier, smaller and rural wireless phone companies want the FCC to let them roam and send text over the networks of their larger competitors.  For example, here’s a quote from a Sprint executive explaining why the FCC should force Verizon and AT&T to let Sprint use their networks so Sprint’s customers could roam them:

“The expectation of consumers is their smart phone is going to work wherever they go. Data is not just an abstraction for consumers now. They use it every day and rely on it more and more instead of voice.”

Meaning, in English, Sprint didn’t mill the bread or bake the bread, but now that these other companies have, they’d like a piece, thank you, and want to compete against these other companies using the other companies’ investments.  If Bergmayer knew the history of the industry, he’d  now that it was the presence of these regulatory restrictions that held the U.S. back from developing cutting-edge infrastructure.  In the 1990s, cable was not subject to these conditions and telcos were, so you got a lot of cable investment but little telco investment.  In 2003, the courts threw this business out, and the “telcos” started investing tens of billions in fiber systems.   Now Bergmayer wants to bring that regime back – what do you think will happen?  Who’s going to invest when your investment can be used as a hostage against you?  Back in the Ma Bell era, companies’ investments were guaranteed through regulation, so you could argue that this was what they signed up for.  But today the companies are betting tens of billions of their own dollars without guarantees – should we treat them as “public utilities” nonetheless?

Moreover, Bergmayer’s nostalgia for “thousands of competitive dial-up ISPs” suggests amnesia.  Raise your hand if yuou want to get rid of \the connection you have now and get Covad and Earthlink back?  Who cares that there were thousands of them?  They all did the same thing – nothing.  They applied a technology they didn’t develop or improve, they made little in the way of investments, and the only reason they existed was because the FCC adopted the policies Bergmayer favors – they all had the right to force the phone company to sell them access to the phone companies’ infrastructure at a regulated price, and they lived like the barnacles who attach themselves to whales for their sustenance.  And once innovation took place elsewhere, Bergmayer’s paragons disappeared.  And that’s where Bergmayer’s ideas lead.

So the Kohl hearings really are important.  Are they going to embrace this view of the world?  Or are they going to lead the government towards a new understanding  of broadband competition and an agenda that works – broadband infrastructure expansion and adoption (particularly in unserved neighborhoods and such important sectors as education and health), eliminating obstacles to the system (like the mis-allocation of spectrum in the hands of broadcasters), guaranteeing users’ privacy and security, and allowing the market structure of this chain of industries to reveal itself through competition, as opposed to merging fully formed from the brow of advocates.

Lower frequencies do offer advantages, but my view is that these advantages are often overstated. Most usage of mobile broadband networks will occur within higher population densities in which networks will have to be designed for capacity rather than coverage. In these scenarios, low and high frequencies offer almost equivalent performance.

In contrast, mobile communications systems used globally have frequencies much higher than this, ranging from 450 MHz to 2500 GHz, with most systems in the U.S. operating at either 850 MHz (cellular band) or 1900 MHz (Personal Communications Systems – PCS) band. In trying to cover an area with the minimum number of sites, using 1900 MHz takes somewhere between 2 to 4 times as many sites as 850 MHz. The exact ratio depends on multiple factors such as path loss, the link budget, cell tower height, and the geometry of the area being covered. Lower frequencies, such the 700 MHz band in which LTE is being deployed in the U.S. right now, requires even fewer sites than 850 MHz, though only slightly.

For rolling out a new “greenfield” mobile wireless network, fewer cell sites equates to a lower-cost deployment. For instance, it’s less expensive to roll out a greenfield LTE nationwide network at 700 MHz than to roll out WiMAX at 2.5 GHz. Both use almost identical radio methods, namely orthogonal frequency division multiple access (OFDA) and 2X2 multiple-input-multiple-output (MIMO) smart antenna systems. The biggest difference is the frequency and the larger number of cell sites required for the higher frequency.

For that reason, a 700 MHz LTE deployment makes a lot of sense as an “underlay” network, a network built for coverage. But as pointed out in many of my papers, including “The Spectrum Imperative: Mobile Broadband Spectrum and its Impacts for U.S. Consumers and the Economy – An Engineering Analysis” (http://www.rysavy.com/Articles/2011_03_Spectrum_Effects.pdf) – even wireless networks that use the most advanced wireless technologies available, such as LTE, have extremely limited capacity. It only takes a handful of users simultaneously streaming video in a coverage area to consume sector capacity. For that reason, operators that have an underlay network will need to add capacity once they start loading their networks with subscribers.

A number of operators are planning to use their AWS licenses at 1.7 GHz (and eventually other frequencies) for exactly this overlay purpose. (Ultimately, cellular and PCS bands will also be refarmed for LTE.) It takes more sites to build out at 1.7 GHz, but the increased number of sites simultaneously translates to much greater capacity. So even if it were to take three times as many sites for 1.7 GHz as for 700 MHz, MHz for MHz, the 1.7 GHz network will have three times as much capacity as the 700 MHz network, and the overall LTE network now has four times the total capacity as it did with just the 700 MHz band. Of course, if an operator built out on higher frequencies, such as 1.7 GHz, 1.9 GHz, or 2.5 GHz in the first place, the operator would have a high capacity network from the beginning. As such, a network build at just higher frequencies would not ultimately cost any more to achieve comparable capacity.

Given that lower frequencies have such good propagation, one might wonder whether it might actually be a liability to use them in denser deployments with smaller cells. In other words, the signal in one cell might keep propagating into neighboring cells and cause excessive interference. Actually, an operator can prevent such excessive propagation by using down-tilt on the base station antennas. Thus the operator could ultimately build a high capacity network using lower frequencies and smaller cells.

The bottom line is that lower frequencies do offer advantages, but my view is that these advantages are often overstated. Most usage of mobile broadband networks will occur within higher population densities in which networks will have to be designed for capacity rather than coverage. In these scenarios, low and high frequencies offer almost equivalent performance.

• Amateur Radio operator Brian Justin filed an application with exhibit (shown below) for special temporary authority to “be able to operate antique Heising modulation on 470.0 kHz on or about x-mas evening and several other days” to commemorate Reginald Fessenden’s “original claimed voice transmissions over 100 yrs ago.” The transmissions were to take place on 470-475 kHz from Forest, Virginia.

• Chesapeake Operating, Inc. filed an application with exhibit for special temporary authority to “provide music and announcements throughout Chesapeake’s corporate campus” and for “determining propagation and coverage while simultaneously considering a waiver to operate permanently under 15.221(b)” of the FCC’s Rules. Operation is to be on 1300 kHz and 1610 kHz in Oklahoma City, Oklahoma. The applicant says it’s parent company, Chesapeake Energy, “is the Nation’s second-largest producer of natural gas, a top 15 producer of oil and natural gas liquids and the most active driller of new wells in the U.S.” “Chesapeake is considering the use of low power AM broadcasts at its corporate campus that could be used for a variety of purposes. For example, the system could be used for disseminating severe weather information (e.g., tornado watches, tornado warnings, ice storms, etc.,) street closings, traffic re-routes due to construction, as well as during outdoor events such as the farmers market that Chesapeake sponsors during the summer months and outdoor activities associated with United Way campaigns, concerts, and family events.”
• Phillip J. Williams filed an application with exhibits for special temporary authority to operate using spread spectrum on HF and VHF frequencies in the Amateur Radio Service. Current rules don’t permit spread spectrum operation below 220 MHz. In the tests, comparisons will be made with other digital modes such as JT65A, Olivia, MT63 and PSK31, including with regard to weak-signal capabilities. Experiments will focus on minimum required transmitter power and developing operating procedures for the Amateur Radio community. Operation will take place in Euless, Texas in various Amateur bands between 1.8 and 148 MHz.
• The Center for Remote Sensing of Ice Sheets at the University of Kansas filed an application with exhibits for experimental license to conduct testing of a 72 MHz link used to control the “Meridian Uninhabited Aircraft System,” an aircraft that carries a variety of scientific payloads, including ice-penetrating radar, for research on the flow-ice sheets in Greenland and Antarctica. Operation will be at several locations in Kansas and Utah on 72.01-72.99 MHz.

• National Public Radio filed an application with exhibits for experimental license to evaluate the feasibility of using a Cognitive Modulator. This is envisioned as an alternative to consumer FM modulators long used to allow audio from a personal electronic device to be played through a vehicle’s FM radio. These modulators have their drawbacks: they can cause interference to other FM listeners, FCC rules limit their power such that it can be difficult for them to overcome interference, and they may need to be retuned as the vehicle travels into range of new, interfering FM stations. Preliminary testing led by NPR suggests a Cognitive Modulator operating at 87.7 MHz may present a solution to the above service problems. Such a device would sense the amount of interference and noise (I+N) at or around 87.7 MHz and adjust its transmitter carrier power to provide a desired C/(I+N) in a vehicle’s FM radio. Experimental operation will be in New Haven, Connecticut on 87.7 MHz
• Lockheed Martin filed an application with exhibits for experimental license to operate at Syracuse, New York on various frequencies in the bands 109.375-137.000, and 960-1400 MHz. This is to test electronic-support-measures receiver systems for the U.S. Navy on vessels being deployed overseas.
• Cosmogia Inc. filed an application with exhibits for special temporary authority to operate communications inks in support of the Dove 1 satellite mission. This is a “technology demonstration to: a) test the basic capabilities of the low-cost bus built from non-space, Commercial Off-the-Shelf (COTS) components; b) show that a bus constrained to the 3U cubesat form factor can host a small payload; and c) demonstrate the ability to design, produce and operate satellites on short schedules and low cost. Dove 1 will do this by transmitting health and payload data to the ground.” The satellite is due to be launched as a secondary payload on the maiden flight of the Taurus II from NASA’s Wallops Flight Facility. It will be placed in a nearly circular orbit of 280 km, which will decay with the satellite burning up in the Earth’s atmosphere approximately 2 weeks after launch. Amateur beacon transmissions on 145.825 MHz will commence upon deployment of the satellite. A half-duplex, spread-spectrum radio on 2.4016-2.4776 GHz will be used for main payload downlink and for telecommand uplink. The satellite has dimensions of 10 cm x 10 cm x 30 cm. Its mass is about 5 kg.
• The Wisconsin Wireless and NetworkinG Systems (WiNGS) Laboratory at the University of Wisconsin, Madison, filed an application with exhibit for experimental license to test fixed point-to-point backhaul and vehicular networking on TV white=space frequencies. Operation will be in the vicinity of Madison, Wisconsin on 174-216, 470-608, and 614-698 MHz. The experimental platform is called Wide Band Digital Radio. Its major function is to perform frequency translation from Wi-Fi frequencies in the 2.4 GHz range to UHF-TV frequencies.

• Lockheed Martin filed an application with exhibits for experimental license to conduct radiosonde factory acceptance testing as part of a government contract. During testing, the radiosondes are attached to a weather balloon and deployed from a Lockheed Martin facility in Marion, Massachusetts. The weather balloon can travel a ground distance of 250 km and reach a height of 30 km. The average duration of the deployment is 135 minutes. The expected number of deployments is about five per month. The radiosonde transmitter uses a monopole antenna that directs transmitted power towards the ground. Testing will take place on various frequencies between 400.25 and 405.5 MHz.
• Carlson Wireless filed an application with exhibit for special temporary authority to test white-space radio technology in rural locations of Hawaii prior to database and device certification. This is to compare performance of white-space radio propagation to that of WiMAX and 900 MHz radios in very dense tropical cover and in heavy rain conditions. Operation will be in Pahoa, Hawaii and in Keaau, Hawaii on 470-608 and 614-698 MHz.
• America’s Cup Event Authority, LLC filed an application with exhibit for special temporary authority to permit video production, and to coordinate operations and security for the Americas Cup World Series Sailboat Race in the vicinity of San Diego. Several frequency bands are requested including 470-476, 476-482, 482-488, and 506-512 MHz (i.e., television broadcast channels 14, 15, 16 and 20), television broadcast auxiliary frequencies 2025-2110 MHz, and amateur frequencies at 2390-2400 MHz and 3300-3500 MHz.
• Robert Miller Consulting filed an application with exhibits for special temporary authority to operate on TV channel 44, 650-656 MHz, near Green Bay, Wisconsin to conduct research on the effects of wind turbines on over-the-air TV reception. The applicant says the “proliferation of wind turbine deployment and the associated history of television interference problems have prompted an urgent need for development of tools to assist in the placement of the turbines so as to minimize interference.” This testing is funded by the U.S. Department of Agriculture, and there is the prospect of more funding for more exhaustive tests depending on these initial test results.
• ShawnTech Communications filed an application with exhibits for special temporary authority to operate in Ridgeville, South Carolina on 851-869, 869.2-893.8, 869.70-893.31, 1930.2-1989.8, and 1931.25-1988.75 MHz. Details are not available due to a request for confidentiality. This appears to be a test of a managed-access cellular system for intercepting unauthorized phone calls from a prison. It further appears that a cellular operator gave its consent for the test.
• Boeing filed an application with exhibits for special temporary authority to test RFID tags that Boeing and commercial airlines use on various items aboard commercial aircraft. The device being used is certified for unlicensed use in Europe but not in the U.S. Testing will be in Goodyear, Arizona on 865-867 MHz.
• The South Coast Air Quality Management District filed an application with exhibits for experimental license to operate a wind-profiling radar, which depends on the scattering of transmitted signals by irregularities in the index of refraction of the atmosphere. The irregularities are caused by turbulence in the wind. By determining the Doppler frequency shift, the speed of the wind can be determined. Temperature data can be obtained by measuring the propagation velocity of an acoustic signal. The hardware includes a receiver/modulator, a final amplifier/preamplifier, a digital control and data processor, and an antenna system. These items were developed by NOAA and are fabricated by Vaisala, and will be owned and operated by the applicant, a government agency that manages air pollution control in the southern California counties of Los Angeles, Orange, Riverside and San Bernardino. The data collected will include hourly profiles of low-level winds between 100 and 5000 meters above ground level (m AGL) and “virtual temperatures” between 100 and 2500 m AGL. This data will be collected to improve meteorological analyses, as well as air quality forecasting and modeling in the South Coast Air Basin. Operation will be on 915 MHz at Irvine, California.
• Harris filed an application with exhibits for experimental license to test transmission and reception of voice and data from 1.35 GHz to 1.39 GHz at various distances and locations at its facility in Rochester, New York. Stationary and mobile tests will be performed to transmit voice and data in both urban and rural settings. Tests will replicate in-theater tactical communications. This testing is partly in support of U.S. government contracts. The tests will use the Harris AN/PRC 117G wideband tactical radio.

• BAE Systems filed an application with exhibit for special temporary authority to test next-generation “communication intelligence” for unmanned aerial vehicles (UAVs). Operation will be in Hudson, New Hampshire on 1626-1660 MHz.
• Orbital Sciences filed an application with exhibits for experimental license to operate from Persimmon Point, Virginia on 2222-2228, 2239-2243, 2258-2260, 2267-2271, 2286-2290, and 5764-5772 MHz. Orbital is under contract to NASA/Johnson Space Center to develop a commercial cargo transportation system for delivery of cargo to the International Space Station. The contract includes two demonstration flights of this system, and eight operational flights to the Station. The experimental operation is in support of various communications needs for these flights from NASA’s Wallop’s Flight Facility, including flight termination system uplink, multiple S-band telemetry data downlinks, a C-band radar system with transmit and receive, and a GPS uplink.

• RF Film, Inc. filed an application for special temporary authority to provide wireless video transmission from film cameras during the production of “Spiderman 4” in Los Angeles. Operation will be on 2363-2371 and 2380-2388 MHz. Those frequencies are in a band normally used for aeronautical telemetry. The applicant has consulted with the frequency coordinator for that band, (AFTRCC), which approved their use on a non-interfering and temporary basis.
• Google filed an application for special temporary authority to test an “entertainment device.” It will test the functionality of “of all subsystems, including WiFi and Bluetooth radio. Users will connect their device to home WiFi networks. This line of testing will reveal real world engineering issues and reliability of networks. The device utilizes a standard WiFi module, and the planned testing is not directed at evaluating the radio frequency characteristics of the module (which are known), but rather at the throughput and stability of the home WiFi networks that will support the device, as well as the basic functionality of the device. From this testing we hope to modify the design in order to maximize product robustness and user experience. Utilizing the requested number of units will allow testing of real world network performance and its impact on applications running on the device, so that any problems can be discovered and addressed promptly. All devices will be used by and registered to specific individuals (all Google employees), and Google will maintain a record of each device, so that they can be easily recalled at any time during testing and when testing is complete. The devices will be tested at Google facilities and in and around the employees residences.” There will be 252 devices in the test, which will take place in Mountain View and Los Angeles, California; Cambridge, Massachusetts; and New York, New York on 2400-2483 MHz.
• AirScan filed an application with exhibits for experimental license to test “state‐of‐the‐art airborne surveillance and security operations for government and private service customers.” Transmissions will be from aircraft in the Titusville, Florida area on 2475.5 and 2458.5 MHz.
• Panoscan filed an application with exhibits for experimental license to test video transmission from a robot it’s developing for law enforcement inspection purposes. Operation is to be in Sylmar, California on 5725-5858 MHz. The transmitter is an Iftron Mondo Stinger 5.8 video transmitter. Apparently, prior work in development of the radio portion of the robot fell under Part 15 of the FCC’s Rules, and now it does not, necessitating the experimental license. Panoscan says it has a request pending before the Commission for waiver of Section 15.247 of its Rules to allow the use of digital modulation.

• GE Aviation filed an application and exhibits for special temporary authority to conduct outdoor testing of its HEET radar system, a “proprietary three-dimensional radar scanner for radar cross section measurements. This one of a kind scanner is currently in checkout phase. Eventually the system will be used on military bases.” Operation will be in Evendale, Ohio and in Peebles, Ohio on 6.5-18 GHz.
• Telephonics Corporation filed an application with exhibits for experimental license to operate in Huntington, New York on 8850 MHz. This to support testing of the ARSS-1 portable radar system. The radar operates on a single channel at a pulse repetition frequency of 5 kpps. The pulse width is 17.0 μS and the receive interval is 183 μS for a total repetition interval of 200 μS.

• Telephonics Corporation filed an application with exhibit for experimental license to conduct tests of its model RDR-1700B maritime surveillance and imaging radar, which the company describes as a multimode airborne search radar that uses pulse compression techniques to provide various search and imaging capabilities, using a programmable waveform generator that can generate different pulse widths, pulse repetitions, and modulation. The radar operates over the frequency band of 9.2 to 9.5 GHz. The radar is continuously changing frequency thereby minimizing the number of undesired pulses being received by fixed-frequency marine and aviation weather radars. This testing is to improve the radar’s signal processing techniques for the purposes of improving the radars ability to search, detect and track multiple targets during over-water surveillance as well as search and rescue and weather detection/avoidance capabilities. Development of imaging techniques that provide the ability to identify the size and shape details of objects detected beyond visual ranges or bad weather conditions will also be part of the testing. Operation will be in the vicinity of Farmingdale, New York.

• The University of Nebraska – Omaha, filed an application and exhibits for special temporary authority to test repurposing of Furuno marine radar to count aircraft at a non-controlled airport. Operation will be at the Council Bluffs, Iowa airport on 9410 MHz. The applicant says it wants to investigate marine radar in this application as a step toward creating a system to prevent aircraft collisions. The radar system in this experiment will include a stationary radar antenna linked to a 10 inch radar display that will transmit data to a computer, which will be programmed to count aircraft. The data collected includes the distance from the radar, the heading from the radar, and the heading of the aircraft.
• Tachyon Networks filed an application with exhibits for special temporary authority to test an 18” terminal mounted to a C-12 military aircraft. Communications will be with one of three Intelsat-owned, U.S. licensed satellite hubs. This is in support of a U.S. Army contract for communications in Afghanistan related to airborne intelligence, surveillance and reconnaissance. Operation will be centered on Middletown, Delaware on 14.0-14.5 GHz.

• Mokulele Research Corp. filed an application with exhibits for special temporary authority to test airborne mechanical tracking antenna performance. Mokulele will use millimeter-wave spectrum from a directional antenna on the ground pointed straight up. The airborne receiver antenna, installed inside the cabin of a small aircraft, will intercept the narrow beam, and immediately activate its reflector to the optimum angle in order to sustain strongest signal level, while the aircraft’s pitch and bank angles change. The aircraft will fly over the ground station between 8,000 and 18,000 feet AGL in tight circles of approximately 0.5 nautical mile diameter. The signal strength, optimized by the tracking antenna, will be recorded for later analysis. An airborne-antenna signal re-acquisition algorithm will also be evaluated. Operation will be on 46.75-46.95 GHz at Haleiwa, Hawaii.
• Honeywell filed an application with exhibits for special temporary authority to conduct flight testing using a developmental sensor to collect data on potential helicopter obstacles such as power lines and towers. The data collected will be used to learn about the detection criteria of such targets. Operation will be in Torrance, California; Phoenix, Arizona; and Everett, Washington on 92-94 GHz. The sensor antenna connects to a PC‐based data processing system used to operate the antenna, display, and capture results. The antenna radiates a 0.7 degree horizontal by 4.0 degree vertical beam. The modulation is a linear frequency modulation that uses up to a total of 1.0 GHz about a center frequency of 93.0 GHz (i.e., 92.5 GHz – 93.5 GHz). The bandwidth is swept repeatedly at a rate of 500 us per sweep.

• Raytheon Missile Systems filed an application with exhibit for special temporary authority to conduct tests on 94-96 GHz at Tucson, Arizona. “This application is being filed for the experimental development of a directed energy device to be exported that will use radio waves to achieve the mission.” (“Directed energy device” appears to be a euphemism for directed energy weapon.) “Because this technology is very new, there is a great deal to be learned still about how to effectively direct the radio energy while ensuring that there is no lasting harm.” “[A]ny personnel present will have volunteered to work on this technology.” The device to be tested will have an input power of 800 watts and an effective radiated power of 50 megawatts.

This is the case because radio networks share common frequencies or channels. A TV broadcaster transmits a signal from a single antenna, and millions of TV sets receive it (most ignore it, but that’s another issue.) There is only one transmitter in this system, so there is no problem with interference caused by the multiple elements of the system. Transmitters generate interference with each other on a shared spectrum system, but when there’s only one transmitter, there aren’t any collisions. TV broadcast is one transmitter, multiple receivers, so it works best when the one transmitter can blanket a large area with a common signal.

Data networks operate in a very different way because they feature two way communication and multiple transmitters. Multiple transmitters thrive on frequencies that travel limited distances, for two reasons:

1. Interference from collisions – multiple stations transmitting at the same time – decreases as the number of potential transmitters visible to a given receiver declines. More transmitters means more collisions, in other words, and each collision is a waste of capacity. This problem is mitigated in several different ways, and each of them is most effective when the pool of potential transmitters is small. Limiting the travel of each transmission limits collisions.
2. The overhead of detecting and correcting collisions is a function of the distance a packet travels before a collision can be detected. In the old Ethernet system devised at Xerox PARC, collisions were detected in the first 64 bytes of each packet. Transmitters were capable of seeing that their packets had collided, so they truncated packets that weren’t going to make it. This was a very low overhead system in which collisions didn’t impose a significant toll on the pool of available bandwidth. Wi-Fi is very different because transmitters aren’t aware of collisions as they happen, only after an entire packet has transmitted and an interval has elapsed for the receiver to acknowledge the packet. No acknowledgement means a collision probably happened. The waiting interval for the acknowledgement is sized to the latency of the network over its largest extent, so it’s a function of the speed (a thousand feet per microsecond) of light over the distance that the network is expected to cover. Wi-Fi networks can’t be larger than 5000 feet without bending the rules and reducing capacity. They require an acknowledgement within 10 microseconds from end of packet, during which time both their own (transmit) packet and the acknowledgement have to cover the whole distance, and the packet has to be checked for correctness. The greater the separation of the most distant stations, the more likely they are to collide as well, because they have to wait longer to see each others transmissions, an additional form of system overhead.

So the two reasons are the probability of a collision happening and the overhead of the collision avoidance and detection systems. Cellular systems don’t have big collision problems because they rely on scheduling systems to allocate air time in a way that prevents collisions, so they can break Wi-Fi’s 5000 foot limit without losing efficiency. This makes them better able to use the propagation benefits that come from the 700 MHz band as opposed to the more limited 2.4 GHz band that most Wi-Fi systems use.

But these things are not completely black and white, and we’ll explain why in the next installment.