The Difference Engine: Off the radar

By N.V. | LOS ANGELES

IT IS annoying enough when you board a plane in New York bound for San Francisco only to end up in Los Angeles because the Bay Area is fogged in (which has happened to your correspondent twice). It would be downright unnerving if it happened because the air-traffic control system directed the plane there by mistake.

The “NextGen” air-traffic control system, which uses GPS satellites to pin-point every plane's precise position in the sky once a second, plus onboard radios that let each aircraft continually see (and be seen by) all others nearby, is to be rolled out in 2012 and fully implemented by 2022. Replacing today's patchwork of ground-based air-traffic control radars that sweep a small arc of airspace every 12 seconds—and lose aircraft between scans and as soon as they go out of range—with a blanketing mesh of GPS satellite signals should allow planes travelling busy routes to steer clear of one another, while flying in tighter formations though crowded airspace. The aim is to save fuel, time and lives, while handling an ever increasing amount of air traffic (see "Unfriendly skies", November 9th 2007).

And so it would except that, due to regulatory haste and shortsightedness, GPS coverage of America could soon go dark in places and become patchy elsewhere. Not only airlines would suffer. There are over 500m GPS receivers in use throughout the United States. Motorists, mobile-phone users, boat-owners, television broadcasters, the police, the armed forces, the emergency services and even farmers would be adversely affected. The Federal Aviation Administration (FAA) reckons it would cost airlines, in particular, more than $70 billion over the next ten years if they had to find fixes to cope with a GPS blackout. A leaked report suggests the airlines' loss of GPS services would also precipitate some 800 fatalities during that period.

The ultimate source of the trouble is a decision made in 2003 by the Federal Communications Commission (FCC) to grant special dispensation to a broadband satellite operator called SkyTerra, allowing it to fill gaps in its coverage by means of ground-based transmitters. SkyTerra's chunk of spectrum (1,525-1,559 megahertz) abutted a crucial frequency (1,575 megahertz) used by GPS satellites. However, SkyTerra's signals being mere whispers from space and its few proposed ground stations designed to operate at low power, any threat to GPS was dismissed as highly unlikely.

Everything changed when Harbinger Capital Partners, a New York-based investment firm founded by subprime-mortgage billionaire Philip Falcone, bought SkyTerra in 2010 and renamed it LightSquared. For Mr Falcone, the attraction was three-fold: SkyTerra's swathe of under-used frequencies; its licence to provide a nation-wide internet service; and, above all, the FCC's waiver allowing it to use ground-based transmitters where satellite reception was poor.

Mr Falcone quickly persuaded the FCC to rewrite the former SkyTerra licence. Instead of being conditional on offering an internet service primarily by satellite, with ground stations filling in only where satellite coverage was inadequate, the revised licence accepts that the network will rely almost exclusively on terrestrial transmitters.

And not just low-powered ones for serving inner cities. The company intends to build a broadband wireless network comprising 40,000 base-stations across the United States. These stations will put out 15,000 watts apiece. Typical mobile-phone transmitters in urban areas radiate between five and ten watts. Even the 100-foot towers used in open countryside transmit no more than 60 watts.

How this came about is a sorry tale of greed, haste and incompetence. Though politically savvy, the FCC is not noted for having the sharpest technical knives in the drawer. According to Aviation International News, last year it accidentally sold the total block of frequencies reserved for the B-2 stealth bomber. In the case of LightSquared, the FCC has no excuse for allowing a national network of high-powered transmitters to operate so close to GPS's frequency. It was clear from the time LightSquared asked for the waiver that there was going to be a huge interference problem.

But in the rush to reallocate underused parts of the spectrum—to fulfill the White House's promise to deliver high-speed internet connections to everyone in the country—the FCC has been guilty of riding rough-shod over objectors. The LightSquared proposal clearly hit all the right buttons at the FCC. And LightSquared had friends in high places, too. Mr Falcone was a classmate of President Barack Obama at Harvard and, though a registered Republican, has become a generous contributor to the Democratic Party of late. Likewise, the FCC's chairman, Julius Genachowski, was also a pal of the president at Harvard and a big supporter of his presidential campaign. Wireless broadband is a surprisingly small world.

LightSquared's plan is to build a fourth-generation wireless network based on the same Long-Term Evolution (LTE) technology that mobile carriers like Verizon and AT&T have started to introduce. But instead of becoming a carrier itself, the firm intends to sell broadband services wholesale to mobile-phone operators lacking fourth-generation networks of their own. It has already signed up Sprint and several smaller fry, such as Leap Wireless and Best Buy. MetroPCS is expected to follow. LightSquared intends to roll out its LTE network from 2012 onwards, with the goal of reaching 260m Americans by 2016.

Over the past year, LightSquared has insisted that its transmitters would not interfere with GPS. On June 30th, however, a study group organised by the FCC issued a 1,000-page report stating categorically that the firm's plan would cause massive amounts of harmful interference. The RTCA, the government advisory body (formerly known as the Radio Technical Commission for Aeronautics) that wrote the report, warned that GPS-based operations would be unavailable over the whole of the east coast of America if the LightSquared proposal went ahead. Apparently, tests carried out earlier in the year at Las Vegas, under the scrutiny of both LightSquared and a group of GPS users, ran into jamming problems, ranging from erratic behaviour to complete loss of signal.

Confronted with this evidence, LightSquared declared it would halve the power of its transmitters and delay operating them within the upper ten megahertz of its frequency band—to provide a buffer zone while the GPS industry introduced receivers able to reject adjacent signals properly. By moving further away from the GPS frequency into portions of the spectrum it shares with Inmarsat, a British satellite communications company, LightSquared reckoned it could solve the jamming problem for 99.5% of GPS receivers. The rest, it said, could be made immune by retrofitting them with proper filtering circuitry.

With some justification, LightSquared blames both the FCC and the GPS industry for creating the interference problem in the first place and allowing it to go unchecked for years. It is certainly true that firms making GPS radios for mobile phones, car-navigation gear and other consumer gadgets have been allowed to skimp on the circuitry inside their receivers—on the assumption that GPS's spectrum neighbours would always be other satellite services listening for equally faint signals, not thousands of powerful antennas broadcasting loudly. Besides, with GPS considered a sacred cow, it was felt that nothing would ever be allowed to knock it off the air.

But building anti-jamming circuitry into GPS receivers is easier said than done. It is certainly a good deal more expensive than the five cents per receiver that LightSquared quotes. What the company has failed to mention is that the transmissions it plans to use will be billions of times more powerful than the neighbouring GPS signals, even if it halves their transmission power. Those used during the Las Vegas trials were up to 800 billion times more powerful than the incoming GPS transmissions. No filtering circuit yet devised can block such interference without massively degrading the signal being protected. When GPS transmissions are so weak to start with, filtering is out of the question.

Where does that leave LightSquared? Difficult to say. Politicians of all stripes have rushed to GPS's defence. Even the White House and the FCC must now be having second thoughts about encouraging the company in its bid to become a wireless carrier. Your correspondent for one, though, would like to see the firm succeed—though not at the expense of a degraded global-positioning system. The fact is, America urgently needs another national broadband carrier—more so than ever now AT&T's $39 billion acquisition of T-Mobile looks set to reduce the competitive landscape effectively to a duopoly, with Verizon and AT&T controlling over 80% of the mobile market between them. Whatever the future holds, it is very much up to the FCC to get America—and the rest of the world—out of the mess it got everyone into.