Alternative to "Black Boxes" Takes Flight
Air France crash renews interest in real-time flight-data reporting systems
16 September 2009—Deep in the waters of the Atlantic lies the only hope of knowing exactly what brought down Air France flight 447, which killed all 228 people on board. It has been more than three months since the crash, and the investigators' holy grail, the aircraft's 'black boxes,' have long since ceased emitting their distinctive ultrahigh-frequency 'ping,' so the search for them has been called off. The recorders, and any insight they could have provided, are lost.
In this age of constant connectivity, many people wonder why the aviation industry still relies on onboard recording devices. Why not a system that transmits information in real time?
It's a question that's come up during the past 20 years in the few instances when black boxes—which are actually bright orange—have been destroyed or lost after crashes. But the latest accident has prompted Airbus, the maker of the downed Air France aircraft, to study alternatives or supplements to the black boxes, like transmitting data in real time or extending the life of the pinger longer than 30 days to give investigators more time to find the recorders.
'I don't think anybody believes that the black box should be replaced, but it should be augmented with technologies that provide more information in real time,' says Bill Tempany, CEO of AeroMechanical Services (AMS), maker of Afirs UpTime, a real-time flight-data reporting system.
Thomas Schmutz, vice president of engineering with the aviation recorders division of L-3 Communications, says real-time data broadcast would be an outstanding, if pricey, complement to the onboard data and voice recorders. 'It is too costly for all the airplanes flying all around the world to broadcast all the data all day in real time,' he says. Besides, airplane black boxes are recovered in the vast majority of instances.
An industry-wide adoption of real-time broadcast of data would also require a mandate from government agencies, such as the U.S. Federal Aviation Administration. Schmutz says the FAA is cautious about introducing new rules that could cause a lot of expense for the air travel industry. 'In a time when airlines are not especially profitable, it becomes difficult to require more,' he says. He added that in the early part of 2010, the FAA will require that black boxes increase their recording time from the last 30 minutes of a flight to the last 2 hours.
There are an estimated 10 000 planes in the air around the world at any given time and only so much spectrum available. And no one single band of spectrum would be ideal. While planes might make do with a VHF band over land, those flying over oceans rely on more expensive satellite communications.
AMS's Afirs 'blue box' reporting system, already in use in a couple of hundred aircraft owned by 31 small airlines around the world, relies solely on satellite communications to transmit flight data in real time. And the system does so cost-effectively, Tempany says.
The system uses an onboard processor to compress data received from a plane's flight-data acquisition unit. The compressed data is then transmitted via Iridium satellites to AMS's servers in Toronto and Calgary, in Canada. Once there, the data is decompressed and routed via the Internet and delivered, sometimes even as text messages, to airline officials or ground control. What helps keep the system's costs down, Tempany says, is that the Afirs transmits data in real time only when certain predetermined conditions occur, such as the loss of cabin pressure or an engine fire, for example. (Pilots in the cockpit and ground control personnel can also trigger the Afirs emergency mode manually in the event of a problem.)
'The black box [system] records the data and waits for somebody to find it. [But] when our system goes into emergency mode, it transmits the location of that aircraft—the latitude, longitude, speed, altitude, acceleration—every two seconds,' Tempany says.
Schmutz is not impressed. He says a system that transmits data in real time only when an emergency is under way could fail to provide investigators with crucial information. 'The most important data when you have a problem isn't necessarily the data you broadcast after you detect that you have a problem,' he says. 'You want to back up 10 minutes and broadcast all of that. If you're in trouble, you're not going to have the time to do that. And you're counting on systems that are in trouble to deliver data.'
Krishna Kavi, an engineering and computer science professor at the University of North Texas, who conducted a study of the EgyptAir flight 990 crash in 1999, says that transmitting continuous real-time data is not as daunting as it may seem.
A plane would not have to rely solely on satellites to transmit its data, he says, because software-defined radios can automatically detect what kind of bandwidth is available at any point during a flight—whether it's VHF, HF, or satellite.
Ten years ago, he determined that each plane needed 2 to 4 kilobytes per second to transmit the data required by the FAA. At the time, there were more than 3000 planes in the sky at any given time over the United States alone.
'Most of us today can get 3 to 6 megabits per second of bandwidth at home. And what we're talking about is maybe one-thousandth of it,' Kavi says. 'The [airlines] are willing to provide Internet on the planes. So if they can provide Internet, they can easily transfer kilobits of data from each plane."