IT WAS just another flight for the nine passengers flying from Palmerston North to Christchurch, New Zealand, on the night of Friday 6 February 2003. As the aircraft approached its destination, the darkness and bad weather meant the pilot of the small aircraft, a Piper Navajo Chieftain, was flying the landing approach on instruments. Without warning, the aircraft crashed into a tree 2 kilometres ahead of the runway. The pilot and seven passengers were killed; only two survived.
The accident investigation revealed that, during its entire approach, the aircraft had been below the glide path for the runway. It could even have crashed sooner; analysis of the instruments found that the aircraft’s glide slope indicators were telling the pilot to descend further. So what went wrong?
Investigators discovered that a call had been made from the pilot’s cellphone just moments before the plane intercepted the signal from Christchurch airport’s glide path beacons. The call remained connected for 3 minutes – right up to the moment the plane smashed into the tree. The investigators’ report concluded that the “possibility must remain that…interference from the pilot’s own cellphone might have caused erroneous indications.”
Advertisement
It is now 45 years since the US Federal Aviation Authority introduced a ban on using non-essential electronic devices in flight. But the FAA has a new fight on its hands.
American Airlines and telecommunications company Qualcomm are claiming successful tests of safe in-flight cellphone systems. Other aircraft and communications companies are busy testing their own systems, all with their eyes on a big cash prize. OnAir, a Geneva-based communications technology company, for example, says the in-flight cellphone market could be as large as $2 billion per annum by 2009. In December last year the US Federal Communications Commission, the body in charge of regulating the cellphone networks, said the new technology could overcome their objections to the use of cellphones in flight. The FCC is now reconsidering its ban (see “To ban or not to ban?”). While the fight over one of the last sanctuaries from cellphone chatter is hotting up, a burning question remains. Will safe flight be the first casualty of this airborne gold rush?
It has never been proved that a phone was directly responsible for an aircraft accident, but pilots have reported numerous incidents where flight navigation and control systems were affected by people using cellphones in flight.
One of the few studies to analyse pilots’ accounts of systems failure, published in 2001, examined reports filed between 1986 and 1999 to the NASA Aviation Safety Reporting System in the US. It revealed there had been 86 malfunction incidents of some sort traced to passengers using electronic devices inside the cabin; over a quarter of these were attributed to cellphones (see Graphic). In a similar study, published in 2003, of reports to the UK’s Civil Aviation Authority between 1996 and 2002, 35 incidents were put down to cellphones. The CAA has told Â鶹´«Ă˝ of at least a further 10 incidents between 2003 and 2005.
The incidents included problems with navigation systems such as GPS and glide slope indicators, flight control systems such as the autopilot, false warnings of unsafe conditions such as baggage compartment smoke alarms going off, and communications being interrupted due to noise in the pilot’s headphones.
There is no way of knowing whether these incidents constitute the sum total or are merely the tip of an iceberg, because no one knows for certain how many cellphones are left on in flight, and no-one has analysed the total number of incidents globally. But based on theoretical emissions and laboratory measurements, the magnitude of the danger is becoming clearer.
Danger signals
“Cellphones have the potential to interfere with aircraft control and navigation systems,” says Steven Rines, an experienced communications engineer who co-chairs the FAA committee that is developing standards airlines would have to meet before cellphones could be used in flight. “The maximum levels they are allowed to transmit exceed the levels that aircraft equipment is tested against. We are concerned about these extreme situations.”
Jay Ely, a scientist at NASA’s Langley Research Center in Virginia, has found further, hidden, dangers in cellphone signals. The phones emit some radiation unintentionally – from the oscillating circuits inside the device rather than from its transmitter. Virtually all electronic devices produce these spurious emissions, even when not directly transmitting.
“A call had been made from the pilot’s cellphone that remained connected right up to the moment the plane smashed into the tree”
Ely found that when he placed a standard GSM handset within a foot of a CDMA (code division multiple access) handset – the type more widely used in America – signals from the handsets interfered with each other to produce spurious emissions at the frequencies used by aircraft Global Positioning System receivers and Distance Measuring Equipment. DME is a navigation aid used by aircraft to determine their distance from a particular radio transponder en route. So if two passengers were sitting next to each other, one making a call on a CDMA handset and the other on a GSM handset, that could cause interference signals on GPS and DME wavebands.
Truong Nguyen, one of Ely’s colleagues at Langley, has found even more conclusive evidence of cellphone-induced problems. In July 2003 the FAA received a report from a pilot who did not wish to be identified that a Samsung SPH-N300 cellphone, which is equipped with a built-in GPS receiver, caused an aircraft’s GPS to lose its signal completely (NASA technical report, March 2004, Evaluation of a Mobile Phone for Aircraft GPS Interference).
The pilot’s company managed to repeat the effect in several subsequent flights, in different geographic locations, on different days, and with three different GPS receivers, each using separate antennas. Every time the phone was turned off the interference vanished. Nguyen managed to reproduce the effects in a lab and measured significant interference in the GPS wavebands. It was the first confirmed, independently repeated example of interference between cellphones and flight systems. And the phone wasn’t even carrying a call.
The emissions in the GPS waveband are particularly worrying because these signals are weak and susceptible to interference, and yet modern aircraft are relying more and more on GPS signals for positioning during landing. However, Paul Guckian, an engineer at Qualcomm, who studies the effects of cellphone use in flight, says the significance of the result should not be exaggerated. “It’s not representative of the typical environment on a commercial airline,” he says.
Typical, however, is hard to define. To test for the effects of the radiation from cellphones in a commercial flight in a controlled way you would have to vary the locations of the phones throughout the cabin, change the power output levels, vary the number of phones and types of phones broadcasting simultaneously, and test all these configurations for each different aircraft design. “It’s just too expensive and too complicated a task,” says Ely.
Mindful of this problem, Bill Strauss and Granger Morgan at the department of engineering and public policy at Carnegie Mellon University in Pittsburgh, Pennsylvania, decided to test cellphone emissions in aircraft another way. Strauss built an electronic box of tricks that could sample different parts of the radio spectrum over time to measure levels of radiation in both cellphone frequencies and in frequencies used by aircraft navigation systems, such as GPS.
They stowed the box in a bag and placed the bag in the overhead lockers on a variety of commercial aircraft run by two different airlines. They measured emissions in 38 flights, and the results were startling. Strauss measured emissions that showed calls were being placed from the aircraft throughout the flight, including during take-off and landing. There was an equally high level of illicit activity on the GPS frequency. Strauss calls the results “disturbing stuff”.
“One cellphone equipped with built-in GPS receiver caused an aircraft’s GPS navigation system to lose its signal completely”
Picosolutions
So how are airlines and phone operators claiming they can solve all these problems? Their answer is picocells, small base stations about the size of a laptop that can be installed in an aircraft to handle passengers’ cellphone calls and relay them via a satellite to the ground. When you switch a phone on it “listens” for a signal from a base station. When it hears a signal it responds, adjusting its output power to the minimum strength required to communicate with the base station. The idea is that cellphones will communicate with the picocell rather than ground base stations, and this should alleviate all the problems. Because the picocell is so close to the handsets, the phones will automatically drop their output power to a very low level, hopefully preventing any interference with the avionics.
There are several pitfalls with the picocell solution, however. For starters, what if the picocell failed? Without a back-up, the crew would have to tell passengers to switch off their phones, to avoid the risk of interference. And at low altitude or on the runway, signals from ground base stations might overpower those from the picocell, so handsets might choose to connect to one of these instead. One possible solution being looked at by many manufacturers is to let picocells block other base stations from connecting to the handsets. “You’re essentially jamming the other base stations so that the devices on board only connect to the picocell,” says Ely.
But this raises technical and legal issues. For example, if the picocell is switched on at the airport, “the picocell may interfere with the terrestrial network,” says Ely, “or with other handsets of people outside the airplane standing in the airport.” And crucially, the picocell only controls the power of the cellphones’ broadcast signal. “That doesn’t mean that the cellphone’s spurious emissions will be reduced,” says Ely.
Various solutions have been proposed, but these all have potential pitfalls – no one can agree on the best way forward. Morgan, for one, is unhappy with all the proposed developments. “I don’t like people using mobile phones in aircraft, and I don’t like that several airlines including Lufthansa and United are moving towards installing picocells. I think that’s a mistake,” he says. Donald Willis, radio spectrum planning manager for the FAA, agrees. “There have been no scientific tests yet to prove picocells are safe,” he says. “As far as we’re concerned there’s no scientific data.”
The FAA are about to start their own tests of interference effects, but these will take time with all the variables to be considered. An aircraft’s susceptibility to cellphone interference will depend on many different factors. Age, shape, and even the individual custom refitting for each airline all influence how radio signals propagate out of the cabin windows and door frames and into the navigation antennas, or internally into the wires that lead to the crucial avionics devices themselves.
And while some newer aircraft, such as the Airbus A380 superjumbo, or Boeing’s 787, have been designed to be more resistant to electromagnetic interference, the older aircraft that were designed before the early 1990s could have bigger problems. These represent the lion’s share of the roughly 15,000 passenger aircraft in service around the world today.
New wireless technologies will bring more problems. The next generation of high-speed wireless devices will use a technology called ultrawideband (UWB), which transmits in bursts across huge swathes of the radio spectrum. Ely has studied the effects of UWB devices on avionics and found they can badly mess up an aircraft’s navigation systems. Barely a single flight system was unaffected by the UWB emissions. “We will simply not allow UWB devices on board,” says Rines.
While that remains a problem for technologies of the future, another problem has already surfaced with current cellphone provisions. A spokesman for Lufthansa, which now operates some flights that enable passengers to make in-flight internet phone calls using a Wi-Fi network, says people have already complained about how irritating it is to sit next to another passenger making calls during the flight.
No matter how much money there is to be made, and no matter how convenient it might seem to phone ahead for a pizza delivery on landing, the people concerned with safety remain unconvinced; for now it looks like aircraft will remain one of the last cellphone-free havens. In a congressional hearing in July, Nicholas Sabatini, the associate director of aviation safety at the FAA, said that any airline that wants the FAA to allow cellphones on its flights would have to prove each individual handset model was not capable of interfering with the flight and navigation systems of the specific type of aircraft on which the phone would be used.
Because the rest of the world’s aviation agencies usually follow the FAA’s regulations, it is unlikely the ban will be lifted anywhere else in the world until the FAA is satisfied. Jonathan Nicholson, a spokesman for the UK’s Civil Aviation Authority, for example, says it will not be revoking its ban “unless you can prove it is 100 per cent safe.”
Additional reporting by Gerry Byrne
To ban, or not to ban?
THE US Federal Communications Commission (FCC) bans cellphones from aircraft because the airborne signals can clog up base stations over a wide area.
The problem occurs because of the way cellphones work. Cellphones on land can normally hear around three or four base stations, and remain registered simultaneously with all of them. This takes up a slot on each base station, and there are only so many slots to go round. On land this is normally not a problem as buildings and terrain usually mean the number of phones within sight of a particular base station is manageable.
However, phones in an aircraft are likely to have line-of-sight contact with scores of base stations, so they will use up slots in many more base stations than the designers of the network intended. Fly a plane full of handsets over the same area and you will soon clog up the base station registers, jamming the network. Hence the FCC’s ban.
To solve this problem, several companies have suggested installing a small on-board base station, called a picocell, which would handle all calls made from within the plane. The picocell forces phones trying to place calls to drop their power output to a minimum. It then routes all calls through to a satellite, avoiding any interference with ground networks.
In December 2004 the FCC announced it would rethink its ban, saying it believed the picocell systems might alleviate the problems its ban was put in place to stop.
However, some problems with the picocell solution have already been identified. Airlines are likely to install picocells that only accept calls from certain cellphone networks. So how will other cellphones be stopped from by-passing the picocell and making calls directly through ground base stations?
George Cooper, chief executive officer of picocell-maker OnAir, says his company is examining the possibility of making their picocell jam all other network signals and only allow those phones that have been registered with the airline’s network to make calls.
But Paul Guckian, an engineer at communications company Qualcomm, thinks jamming is a mistake. “We’re very sceptical about the jammer proposal,” he says. “If you generate noise on the aircraft that will mask the ground networks then that noise will also have a negative impact on the ground networks.”
