A different view on MH370 tracking data

In the past week or so that MH370 has been missing I’ve read all sorts of stories about the disappearance of MH370. Some are entertaining conspiracy tales and others much more serious takes on what may have happened written by actual industry pros. And more than a few have called for wholesale changes in the way aircraft operate and are tracked around the world. The quip about “If I can find my iPhone anywhere in the world why can’t I find a huge airplane” has been used far too often. But what if the takeaway lesson here is not that we need major changes to the way aircraft are tracked?

Between MH370 and AF447 there have been calls to revolutionize the tracking of aircraft. Many want to live stream black box data to the ground rather than only have it on-board. And the technology is there to have that work. But it is not cheap. Bandwidth in the sky is still expensive today, particularly for global overwater coverage. And pushing the full set of black box observations in real time would use up a decent amount of that bandwidth. Early in 2015 the skies will see a constellation of Ka-band satellite global coverage which should bring the costs down to a more reasonable level, but that’s still a ways away and I’m also still not convinced it is necessary.


If we scale back to asking for only basic location/velocity data and scale back the sample rate then the costs become more reasonable. But what does that get us more than we already have today? Why not simply configure the radar transponder to not allow it to be permanently disabled? Doesn’t that accomplish nearly the same thing with virtually no overhead costs? Why are we talking about whole new systems and reinventing the tracking process when all the bits already exist pretty much as we need them?

Of course, that’s only tracking location, not all the vital signs from the aircraft. Do we need all that data on the ground all the time? I’m still inclined to say no. Certainly not without considering the costs to aggregate and manage it. Plus, as the frequency of the data polling or transmission is decreased (saving on bandwidth costs) the value of the data similarly decreases. If you’re only getting data every 5 minutes then it would be possible to miss the entirety of a major catastrophic event. The failures in AF447 occurred in roughly a 4 minute window. A transmission at 2:10:00am UTC would have shown all systems at “normal” settings and 5 minutes later the aircraft had already crashed.

I’m not some heartless asshat who doesn’t care what happened to the plane or who thinks that the families of the passengers and crew should suffer needlessly with a lack of information or, even worse, misinformation. But I also do not think that massive adjustments to the processes and systems are what get us to better data as a whole. In short, I’m mostly surprised that so many are calling for such drastic changes in the way aircraft tracking is handled today. It seems to me that a few rather simple adjustments would net very similar results at a fraction of the cost. And likely with equally sufficient utility.

Never miss another post: Sign up for email alerts and get only the content you want direct to your inbox.

Seth Miller

I'm Seth, also known as the Wandering Aramean. I was bit by the travel bug 30 years ago and there's no sign of a cure. I fly ~200,000 miles annually; these are my stories. You can connect with me on Twitter, Facebook, and LinkedIn.


  1. Valid points… Come down to the cost/benefit curve. Of course, government and regulators ignore that all the time – just look at Airport Security in many western countries.

  2. You’re right about missing the crucial 5 min for example. But also, if it’s only 5 min times before the plane pings the base with some info, you can pin point that 5 min window to figure out the location for example, greatly decreasing the time and cost of potential search and rescue mission.

    1. Sure, though the S&R teams knew more or less exactly where AF447 was when it crashed (they found wreckage within a day or two) and it still took 2 years to find the recorders. So spend a lot more money to know where the planes are and still not really expedite the recovery?? Doesn’t make much sense to me.

      Something more than what is in place today is probably called for. I just don’t think we need a massive redesign/rebuild of the system so much as a few subtle changes which aren’t nearly as costly.

  3. What are we talking about in terms of cost? Harvard Law School tells me this might cost half a million dollars per aircraft. http://blogs.law.harvard.edu/philg/2014/03/10/why-dont-we-have-streaming-data-from-the-missing-b777/ A bit of google tells me that this is about one tenth of one percent of the cost of a new B777. http://www.boeing.com/boeing/commercial/prices/ Amortized over the life of the aircraft, this seems like a pittance, especially when you consider that the aircraft is lost once it crashes. If we had real time data on AF447, someone would have told the cockpit, “push down on the stick,” and they’d all be here today.

    1. That’s quite a rosy view of the AF447 scenario, LTL. Just collecting all the data does not mean it is immediately processed and analyzed. You’re assuming that someone would be monitoring the data in real-time and be in a position to contact the crew in-flight. But we’re not talking about systems which provide a real-time two-way link including both voice and data. That’s going to be significantly more expensive.

      Also, that link you shared covers CapEx costs but not OpEx costs. There is brief mention of $2/minute for Iridium network fees for just position data. We already have ways to get that data with the existing hardware and comms systems on board without requiring a new kit to be installed.

  4. With so many onboard systems currently all that would be needed to transmit data in a more dire situation is just a little code. If x, y, or z happens then transmit more frequently. The plane can stay in charge of when that happens, but if health decreases or if anything outside the ‘normal’ function of flight happens then transmit the heck out of everything onboard. And all that polling of systems would happen onboard. I would think that some kind of forward thinking airline could start doing this on its own and earn INCREDIBLE loyalty from customers. Imagine if American/United/Delta just decided to do this and how it would change aviation across the globe. Yes it would be an investment but I have to think a lot of passengers would push their dollars that direction rather than somewhere else.

    And I agree, just remove the ability to turn off critical positioning systems like these.

    1. @ Sice — wouldn’t that kind’ve be like the discussion of a great warranty vs. a great product? “If your flight goes down, anywhere, anytime, your loved ones will know where you went down.” … just sayin’…

      1. That’s what this whole discussion is about, and how a plane has been missing this long when it shouldn’t have been. Simple changes could avoid that. And think if the plane went down in a controlled, gradual descent, it’s possible that people could have survived and been floating in the ocean; if they knew where the flight went those people could have been saved. So yes, I agree that better product is a more marketable angle but if I can choose an airline that takes safety seriously enough to invest in something like this you better believe I want it. And there’s no reason an airline shouldn’t offer that kind of service if I expect it from the new car I buy. Peace of mind can be a big seller.

  5. @ Seth – To those who simply think that there should be no way for the pilot to disable the transponder, what would you do in the event of an overloaded circuit? You have a breaker on every one of the electrical circuits in your home/ apt. in order to prevent a fire; wouldn’t you want the same thing on an airplane?

  6. I agree it’s surprising the transponder can still be disabled however my understanding from reports was that there are still radar blackspots where even having it switched on would make no difference. They said that MH370 was due to enter one of those so it was only when it didn’t come out the other end that the problem became apparent. Thus why there’s the suggestion of making more use of air to satellite rather than relying on ground to air which is patchy in places. I’d also argue a higher output beacon on the black box would be sensible even at an increase in size/weight (and this field costs) that this would bring.

Comments are closed.