(Orange) Black-Boxes - To Stream, or not to Stream?

It looks like the flight recorder from the tail of Air Asia QZ8501 has been located. Following MH370 and MH17, the loss of another airframe was of course met with a great deal of media interest, not least regarding the question of our ability to track aircraft. There have been calls to stream the data from flight recorders in real time in order to locate the aircraft more quickly and to remove the need to find them in order to gather the information stored from them. So why aren’t we doing this already?

Firstly, lets look at the positioning aspect. Up until recent years, aircraft were mainly tracked by radar and by information broadcast by the pilots themselves. Radar consists of large and powerful radio transmitters broadcasting energy into the sky, waiting to see if any bounces back. Transponders in aircraft detect when they are illuminated by a radar pulse, and immediately broadcast their identity on a radio channel. The radar tracking station on the ground detects a target at some bearing, altitude and range and correlates it with the transponder ID that arrives at the same time that the radar pulse returned. However, we now also have a system called Automatic Dependent Surveillance Broadcast, or ADS-B. This is a lower-cost system than primary radar, and the aim is to phase out primary radars in favour of ADS-B.

ADS-B transmitters contain a GPS recevier and a UHF radio transmitter. Aircraft broadcast their identity, GPS location, velocity, and other details every few seconds. ADS-B is a line of sight technology, so the range is typically a few hundred kilometres for aircraft at cruising altitude. However there are serious plans in motion to track ADS-B and AIS (the maritime equivalent – Automatic Identification System) using spare payload capacity on communication satellites. There are already satellites monitoring AIS and Iridium are aiming to deploy their ADS-B monitoring capability over the next two years. Following the loss of MH370, Inmarsat have also stated that they will offer a free live monitoring service, writing off a potential $15M revenue stream to improve the availability of aircraft tracking data. All in all, global live tracking of commercial aircraft is nearly here, and the black box data cannot provide a better estimate of the location of a downed aircraft than the final GPS fix squirted out of the ADS-B transmitter can.

So why are we not streaming the data so that we do not have to go and retrieve the black box? Well the cost associated with streaming this data is (believe it or not) huge. This seems like a mad statement here in 2015, but aviation-grade engineering and airworthiness certification procedures are not cheap. If we take ADS-B as our text book example of bolting something new onto an aircraft (something that basically just uses a GPS receiver and a radio transmitter), let’s consider the cost of that:

In the USA alone, the cost of ADS-B implementation has been $6.5 billion. That’s not a typo (billion). Bear in mind that it is basically doing the same job as the (free) Glympse app on a ($200) smartphone by lobbing a GPS fix out over a radio, but there is no hacking, modding or bodging in the air industry. There are no free apps written by a teenager and propped up by advertising running in the cockpit of an A380. Aviation-grade engineering and certification processes are rigorous, time consuming and incredibly expensive. The reason that ADS-B is being implemented as such great cost is because it will reduce costs in the long run, by reducing the number of primary radars that need to be installed and maintained. The data can also be used for a variety of other purposes too, there is a market in mining and selling the data. This is the reason satellite companies are keen to start monitoring ADS-B broadcasts, as I mentioned a moment ago. There would of course be no way that third parties would be allowed to monitor and monetise live streaming cockpit voice data or black box data. A Canadian company is offering live streaming of black box data for a mere $120k installation and $1000 per month (both figures are per aircraft). Those are the cost differences between consumer-grade communications systems, and aviation-grade communications systems. Sounds like one hell of a smartphone contract doesn’t it? There will be many further costs beyond those associated with training pilots and engineers to use the new system, changing checklists to accommodate the new system, taking aircraft out of service in order to fit the new system, etc. There is no way to monetise the data from it, and people will not pay a higher ticket price in order to morbidly fund an improved capability in the event of their tragic deaths. In short, we do not have streaming black box data because of the cost, but many (and I am in this camp) argue that we just don’t need it.

Flying is very safe. In the grand scheme of things, as long as you are flying on a premier airline, you are much more likely to die on roads than during a flight. People are scared of the concept of a plane crash because it might be a rather terrifying last few minutes of life. But still, it is very unlikely to happen. This also means, therefore, that the black box data is very rarely needed either (because crashes are so rare). When it is needed, the block box is almost always found. Of course, finding the wreckage will still be critical even if black box data is streamed live, so that bodies can be repatriated and the wreckage can be examined. Even with the black box data available, analysing the wreckage is still a very important part of understanding what happened and then engineering the fix to ensure the same failure is unlikely to happen again. So streaming black box data is very expensive and most of the time it will be unnecessary. The only benefit it might bring is in appeasing the general public in their desire to know as quickly as possible why a plane has crashed.