Crash of Flight 111 homepage

With data from flight and cockpit recorders sometimes lost or irretrievable following a crash, should airlines begin transmitting such data in real time to ground stations?

After I watched a rough cut of "Crash of Flight 111" here at the NOVA offices, a colleague who watched it with me asked, "Why don't they just beam black-box data in real time to ground stations via satellite?" She was responding to another colleague's remark about how unfortunate it was that, as mentioned in the program, the flight data and cockpit voice recorders on Swissair Flight 111 had ceased recording roughly six minutes before the accident.

The New York-to-Geneva flight had mysteriously plunged into the ocean off Nova Scotia on the night of September 2, 1998, killing all 229 people aboard. The black boxes could have provided valuable clues as to the cause of the crash. But without their data on the final minutes of the flight, the Transportation Safety Board of Canada undertook what became a four-and-a-half-year, $39-million investigation to try to determine the cause by other means.

Neither I nor anyone else in the room that day had an answer to my colleague's question. So I decided to look into it. After talking to experts at the Federal Aviation Administration (FAA) and National Transportation Safety Board (NTSB) as well as in the academic and private sectors, I got my answer: They're working on it, with intriguing possibilities not just for black-box data transmission but also for better troubleshooting during emergencies, improved airline efficiency and maintenance, even telemedicine. But it's easier said than done.

Why do it?

The advantages to having continuous, real-time transmission of flight and cockpit information are many. First, in the case of a crash, black-box data would be available to investigators immediately. They wouldn't have to wait for the recorders to be dug out of the ground or fished from the seafloor. This has important implications in our age of terrorism. Authorities need to know quickly whether a crash was terrorist-related; suspect trails grow faint with each passing hour, much less the days or weeks it can take to recover black boxes and analyze their recordings.

And when black-box recordings are lost, as was the case with the two planes that struck the World Trade Center, downloaded data may be the only data. (I should note here that if Swissair Flight 111 had had a real-time data link, it would have quit operating at the same moment as its flight and cockpit recorders. Like black boxes, data-link systems would rely on a plane's electrical power, which the Flight 111 pilots shut off during their attempt to isolate and fight the fire that eventually brought the plane down.)

A robust data link would also mean a lot more information than just black-box data could flow, and it could flow both ways. As it is now, aircraft, particularly when flying over oceans, are often incommunicado for long periods. "Next to a remote desert island, it's about the only place you can hide and be out of touch," one aviation expert told me. With a broadband communications link, people in the air and on the ground could be in constant, detailed contact.

This would have clear benefits, most significantly the ability to cope with emergencies. If air-traffic controllers clearing aircraft for takeoff could glean pilot intent from incoming data, for example, they could reduce runway incursions, a leading cause of aircraft mishaps. "If you're polling this data every second, and you see that a pilot who hasn't been cleared is powering up and taking his foot off the brake, you know he's going to move," says Jay Brown, a computer scientist at the FAA Tech Center in Atlantic City, New Jersey. "With airplanes, it takes two or three seconds to get them to move, so you could potentially stop a runway incursion."

During a technical emergency, pilots could get critical advice from engineers and other experts on the ground, who would have real-time knowledge of that plane's engine conditions, flight-control positions, and other essential data at their fingertips. During a hijacking, air controllers, government authorities, and other key personnel could make more effective decisions, even warn possible targets. Had controllers been able to detect instantly the flight-path changes and other burgeoning anomalies on the 9/11 planes, perhaps something could have been done, at the least, to lessen the loss of life on that tragic day. Flight attendants could even better tend to mid-flight medical crises. Sensors attached to a patient, for instance, could give ground-based doctors vital signs and other measurements upon which to base potentially life-saving decisions.

Whether intentionally triggered or not, emergencies on aircraft are thankfully rare, and it is during routine operations that such a high-bandwidth link would be most continuously useful. If, for example, computers monitoring a data link could alert mechanics that a certain part of a flying aircraft was in imminent need of replacement, an airline could have that part waiting at the plane's destination. If airline staff had a real-time sense of exactly where their aircraft were at all times, they could better schedule personnel to greet planes, collect baggage, and the like. If the FAA saw that an airline was constantly monitoring an aircraft's engines—a clear safety advantage—it might extend a required overhaul from every 15,000 hours to every 20,000 hours, a significant cost saving to the airline.

Making it work

Several ways to develop such a system are under consideration. The FAA Tech Center, which researches, develops, and tests new aviation equipment for the FAA, has been working with the San Diego-based Titan Corporation on a prototype system architecture. It relies on existing communications links such as air phones (when an aircraft is within line of sight) and satellites (when it's not) to transmit black-box or other aircraft or cabin data to ground stations. Titan's Richard Goelz says the project grew in part out of the 1996 crash of TWA Flight 800, whose black boxes took over a week to recover, at enormous cost. "The project started with the idea that there needs to be a better way of making data like that available without an extended search," he says.

Another company, AeroSat Corporation of Temple, New Hampshire, has a Department of Transportation grant to demonstrate a different way of doing the same thing. Instead of air phones or satellites, AeroSat proposes using aircraft themselves as transmitter/receivers. When a plane is beyond line-of-sight, it would send its data to another flying aircraft in the fleet. That second aircraft would pass the data to a third plane, and so on down the line until one of that airline's planes is within line of sight of a ground station and can dump the data down. "It's like rain coming off a roof," says AeroSat's Bill McNary: the data passes along horizontally in a high-altitude network of daisy-chained aircraft until it can find a way down.

Regardless of approach, experts agree that the technology exists and that the amount of data to stream is quite manageable.

So what's the delay? Cost, for starters. "Purely from an accident investigation standpoint, airplanes almost never, ever crash," says Jim Cash, the NTSB's senior technical advisor for black boxes. "So to be continuously sending data off an airplane that could go its entire 40 or 50 years of flight service and never have an incident or accident—nobody's got that kind of money, especially the airline industry. It's still much cheaper to put a box on an airplane and just have it run its merry life recording data that nobody will ever look at."

Costs include designing, building, operating, and maintaining the system, which may require a series of new ground stations and/or expensive satellite time. There are also privacy and competition issues to consider. How, for instance, do you ensure, even with encryption, that only Singapore Airlines receives Singapore Airlines data? And there's no guarantee that a plane under stress won't lose its data link just when it's needed most, as would have been the case in those final six minutes of Swissair Flight 111. For this and other reasons, most experts agree that such systems would supplement rather than replace black boxes.

Despite the hurdles to wireless transmission, it may not be too far off. Aviation specialist Paul Czysz, a professor emeritus at St. Louis University, believes that all it would take to spur an official drive for a telemetry system would be the crash of a major jetliner over mid-ocean in which the black boxes were unrecoverable—"a Titanic event," as he calls it. "You're going to have to have something like this," Czysz says of a real-time data link, "just to make sure you know what happened."

And if there's one thing that everyone—from crash investigators to victims' families to just about anyone who ever flies—wants to know in the wake of disastrous crashes like Swissair Flight 111, it is what happened.