How Satellite Company Inmarsat Tracked Down MH370

mdsolar (1045926) writes "Malaysian Prime Minister Najib Razak has announced that, based on satellite data analysis from UK company Inmarsat, Malayian Airlines flight MH370 ended in the southern Indian Ocean, and no one on board survived. 'Effectually we looked at the doppler effect, which is the change in frequency, due to the movement of a satellite in its orbit. What that then gave us was a predicted path for the northerly route and a predicted path the southerly route,' explained Chris McLaughlin, senior vice president of external affairs at Inmarsat. 'What we discovered was a correlation with the southerly route and not with the northern route after the final turn that the aircraft made, so we could be as close to certain as anybody could be in that situation that it went south. Where we then went was to work out where the last ping was, knowing that the aircraft still had some fuel, but that it would have run out before the next automated ping. We don't know what speed the aircraft was flying at, but we assumed about 450 knots.' Inmarsat passed the relevant analysis to the UK Air Accidents Investigation Branch (AAIB) yesterday. The cause of the crash remains a mystery."

Comments

Executive summary...

[Last_Available_Usern]

We still have no idea exactly where the aircraft is, how it went down, or what to do now.

Little disturbing

[kid_wonder]

Did the Malaysian government just make a statement to the families based on a statistical probability?

Or did they make that statement based on debris found that was positively identified to the aircraft.

Re:Little disturbing

[Zocalo]

Inmarsat managed to eliminate the northen arc based on differences in expected doppler differences of the signal pings, when the last ping was received, and assuming a conservative fuel consumption to that point, there would have been insufficient fuel left for the plane to make land, hence it went down in the ocean. It's important to note that Inmarsat is unable to say where exactly, only that it is within a given range of the location where last known ping was now known to have been sent from, which is where the search for wreckage is now centred. I gather this is the result of a highly unorthodox set of data analysis that is well outside normal procedures for determining location, hence the reason it's taken so long - some of the techniques they used probably haven't ever been done before.

Re:Little disturbing

[bluefoxlucid]

I don't know why anyone would find that disturbing.

In Tres Roeder's "A Sixth Sense for Project Management," he shows a diagram of communications. This diagram shows information versus time.

In the beginning, information is unknown; then the information changes, back and forth. For example: a dollar estimate may be $3,000 for a project, then $85,000 when we realize we need to excavate cabling tunnels for a line, then $6,000 when we realize we can run this across our existing tunnel and have a new fiber optic pulled for $3,000, then $7,000 when we realize we're going to also need a new transceiver, then $4000 when we find out some of the other equipment is unnecessary, then $14,000 when we realize the scope of labor required is twice as big.

Finally, once we have enough information, that figure stays. Perhaps at $14,000. We also realize we've got the correct figure because we have a full analysis of scope and work required--or at least, the figure won't change until we've done a bunch of work and realized, deep into the project, that we missed something. In any case, it is now not likely to change simply because our information base is hot.

During the initial planning phase, communication should reflect this: the understanding of the situation--the lack of precision--and what is being done to pin that down is to be communicated; conclusive statements should not be communicated because the current understanding of the situation is inconclusive. Once the situation has reached a point of conclusion, then you communicate these conclusions.

What is disturbing about the Malaysian government here is they have been repeatedly saying, "We have no idea what's happening and there's a ton of information out there we're missing; but this is what happened." Then, five hours later, "Oh we found more debris, we think this happened instead." Then the next day, "Oh there was some satellite telemetry information we weren't done analyzing, but it's provided additional information, so we think the plane may have gone this way..."

In other words: They have piles of information they know they're missing, piles of information they have a plan for finding (i.e. "ongoing investigation"), and huge and visible gaps they know exist and expect to fill. They should not be communicating any conclusions at this time.

Re:Flight recorder

[ColdWetDog]

They have narrowed down the presumed crash site. TFA states that the Malaysian government takes this data as proof that the plane crashed near Australia. While important evidence, it's hardly proof - we will need actual debris.

The Malaysian government has been widely criticized about it's handling of this affair. They would like to wash their hands of it and go on to doing whatever it was they were doing out of the world's spotlight.

Re:Flight recorder

[Zocalo]

Still vastly better than what it was only a day ago, and there seems to be a lot more possible debris sightings in the search area which I take as a sign they might be in the right area and will hopefully pin it down some more. The race now is to find it before the black box transmitters go silent, a task for which the US is dispatching some specialist search gear apparently, because that's probably the only hope of giving the bereaved a chance at some closure left now.

Re:Flight recorder

[afidel]

They're probably worthless, the cockpit voice recorders are only required to have 30 minutes capacity with a recommendation for 2 hours, since we know it was at least 4 hours between the critical event (the plane turning south) and the crash the CVR's won't have any information about the events that matter (I'm assuming 777 uses digital recorders so they won't be able to pull phantom prior recordings like they sometimes were able to on analog recorders)

This story is so strange

[GodfatherofSoul]

It just seems like they have information that still doesn't make sense for what we're told are the available resources. The public info just seems so selective as if each government is trying to hold their surveillance cards as tightly as possible. And, intel from an old satellite seems like a cover story. This is all just so...off.

Re:This story is so strange

[SydShamino]

I thought that maybe a bunch of spy satellites picked up and stored the broadcasts, and that they can use timestamps from the various receptions to triangulate the position. That's sort of a reverse-active GPS.

Of course they'll never say "we got this from U.S. and British spy satellites" so they make up something about doppler shift data from a single satellite and hope they find the debris soon to corroborate the story.

Or maybe they did do it all from the doppler shift data they happened to store. It's at least plausible, and there's no need to create conspiracy theories when they aren't particularly shocking.

Mystery?

[alta]

The cause of the crash isn't a mystery. It most likely ran out of fuel.

The cause of why the whole damn plane went AWOL IS a mystery.

They haven't tracked it down

[BitZtream]

They have a theory, nothing more. Still no actual debris has been confirmed. They don't have the full picture so its REALLY easy for their theory to be wrong.

God you suck ass at actually posting facts slashdot.

Re:Flight recorder

[geogob]

It's only completely worthless if its silent.

On the contrary. A completely silent CVR tells you a lot; it tells you that the airplane kept on flying with every one on board either unconscious or dead for at least 2 hours before the crash. That's a critical information for the investigation.

Furthermore, through data/media forensic, you might be able to recover the previous data that was overrecorded, although I wouldn't count on it after 3 to 4 record cycles.

Re:Some questions

[Strider-]

How come the frequency information of the signal received by the satellite was saved? What is the purpose of saving all that data in normal operations?

The communications system in question is likely based on TDMA (Time Division Multiple Access). While I have not worked with Inmarsat systems, all the other satcom systems I have worked with log each connection, and various pieces of information regarding the connection. One of these parameters that is logged is the frequency offset (ie the difference between the expected and actual frequency). This is useful from a troubleshooting perspective as it allows you to spot transmitter and receiver components that are drifting out of specification. Some of the more advanced satellite systems (iDirect) will actually log the geographic coordinates of the uplink site, as this plays into the timing requirements for the network. Unfortunately, Inmarsat isn't this aggressive with their timing, so time of flight isn't an issue).

And why did it take three weeks to do that analysis?

This is pure speculation on my part, but I would wager they had to go back through significant amounts of logs in order to characterize the transmitter and receiver components on that particular aircraft. The doppler effect is going to be subtle compared to the thermal drift of the transmitter, so they need to factor that out before they can get at the thermal drift. Also every oscillator and transmitter is different, so they would need to characterize the transmitter that is on that specific aircraft (which is now of course missing).

Re:Flight recorder

[ScentCone]

as a practical matter actually finding the plane won't change much

Really? You don't think there's much of a difference between knowing it was a mechanical failure (or fire, etc) and knowing it was a deliberate criminal act? If the problem was related to payload or the aircraft's infrastructure or maintenance, you don't think it's vital for all of the other people flying on that same equipment to know what went wrong? If this was done by the pilot(s) at the behest of some organization or state, or otherwise in the service of some agenda, you don't think that's meaningful, in the context of trying to prevent it from happening again? Glad you're so relaxed about it. You probably don't do much business overseas, or ship expensive things that are central to your mission, or have relatives that fly on that equipment or in that part of the world, so that's probably why the death of hundreds and the loss of a huge, expensive aircraft is a yawner to you.

Re:Flight recorder

[BitZtream]

CVRs on those aircraft are 2 hours, not 30 minutes.

What I want to know, is why my phone (the smallest model made) can hold 1100 hours of compressed audio ... but these aircraft using NAND don't hold more than 2 hours of uncompressed audio (you don't want any quality sacrifices or artifacts from compression to screw up your analysis later) in a redundant array ...

Someones going to tell me that for the 30-40k those black boxes cost ... they can't put some actual storage space in the fucking things?

US Intel Said this on Day 1

[rockmuelle]

What's most interesting is that the anonymous reports from the US intelligence community the day after the plane disappeared said that the plane was on the bottom of the Indian Ocean. These claims seemed a little odd at the time since there was no supporting evidence at all and rescuers were still looking for debris on the original flight path. But, it's looking like they were spot on.

I'm going to go out on a limb and suggest that the only real conspiracy in this whole affair is the US govt's cover up of the initial leak. The plane itself likely just suffered a catastrophic failure and lumbered on until it ran out of fuel. But, the US govt also likely tracked it the entire time. That's why someone was able to make a confident pronouncement so quickly. They knew exactly where the plane was, if not exactly what happened. But, this intelligence capability (tracking all flying objects all the time) is probably highly classified. Rather than give it up for a civilian SAR effort, they decided to keep it under wraps, knowing that eventually the plane would be found and the capability is far more useful if no one knows it exists.

Re:ACARS

[icebike]

No it isn't, its not using data from a system that WAS TURNED OFF within MINUTES of the last radio contact.

How the fuck did this get marked as insightful? Its make a wrong statement that everyone has know has been wrong since the second day.

You should try to keep up with the actual events instead of lashing out on Slashdot.

The DATA transmissions ceased on the ACARS, but the radio system still pings the satellite.
The radio system keeps its link with the satellite as long as the actual transmitter has power.

Just because you stop tweeting on your phone doesn't mean the phone stops talking to towers.

Re:Flight recorder

[Anonymous Coward]

As a supervillian I prefer locations such as this, such as my base deep on the ocean floor. Hopefully they don't find it when they look for the damn plane.

Re:Flight recorder

[Lumpy]

They already know about your Skull Volcano island... they dont care because the US government hopes to contract out your services for the "protection" of the American People...

BTW: you have those genetically altered badgers ready yet?

Re:Flight recorder

[bemenaker]

You are making a PRESUMPTION that the transponders were turned off by hand. There is still the possibility of a fire or some other case. This is why recovery of the FDR's is so important. The pilots may not have been on the radio, but the FDR's record everything they say. The conversations between flight crew is crucial, along with all the airplane data.

Re:Flight recorder

[pixelpusher220]

So instead of this being something done by one or two people, you'd much rather it be a systemic problem with aircraft that tens of thousands of people fly in everyday?

Actually yes. The last time a 'few' people in cockpit tried nefarious stuff, the world changed and the US (my country) invaded Iraq and started headlong down the road to totalitarianism. A perfect storm really. An already corrupt 2 party system that doesn't do any serious check and balance, only raise money for re-election, now gets a perfect 'for the children' defense for *anything* proposed no matter how stupid or pointless it is.

Or we have a mechanical failure that can be fixed...

I'll take the latter every single time. The above sounds like a rabid libertarian point of view, but I assure you it isn't. I'm as left wing liberal as just about anybody. I believe in government and it having a purpose; but with the NSA and everything else we've seen in the last 15 years...it is truly broken to the core and it's up to us to fix it.

Re:Flight recorder

[Talderas]

Whether or not there are survivors was pretty much a moot point by March 15. If it was a violent crash into the water, no more than three dozen probably survived impact. Of those three dozen, probably 50% or more sustained fatal injuries that would kill them within 16-24hrs without immediate medical attention. The remainder would have perished within 3-7 days due to dehydration and lack of nuitrition. That's ignoring any exposure related issues, predatory marine life, or drowning. Dehydration, starvation, and miscellaneous death causes would have happened over 3-7 days if it was a soft landing.

Had they, on a slim chance, crashed on land rather than the sea there might still be some chance that survivors lived but that is going to be a hard landing and those that did survivor it may not have been mobile, conscious, or otherwise in a state where they might be able to find potable water or food wherever they landed.

The purpose of finding the plane is not to be able to declare the passengers and crew dead the fact that they waited this long was cruel because it unnecessarily kept hope alive. The purpose is to recovered the recorders so that it can be can be determined what might have happened and if there was any mechanical or electrical issues that would warrant attention for other 777s.

Why?

[GameMaster]

* We've had flight recorders on all major airliners for decades now.

* We've had satellite phone technology for decades now. (since 1979 for Inmarsat)

Remind me again why "black box" style cellular data transmitters aren't required to be transmitting cockpit voice data and full telemetry from every major airliner at all times yet? With a system like that, installed in a way that can't be tampered with by the people in t he plane and runs independently of the rest of the electronics in the aircraft, there's no reason we would know the exact location the plane went down and, most likely, why. Hell, even if they decided to be cheap and only have it transmit the telemetry in once-a-minute updates we'd still would have know where the plane was to withing a handful of miles from the first day it went missing...

Re:Flight recorder

[mean pun]

Which is why they waited literally days before asking the international community for help? Seriously, significant progress didn't begin until the other countries were allowed to start helping.

The first few days the obvious extrapolations from the normal flight path was searched, and that search was not only conducted by Malaysia, so other countries were involved from the beginning. When they realised things were not as simple as that they asked for more international help. I fail to see what they did wrong, even in hindsight.

Re:Flight recorder

[dj245]

Still vastly better than what it was only a day ago, and there seems to be a lot more possible debris sightings in the search area which I take as a sign they might be in the right area and will hopefully pin it down some more. The race now is to find it before the black box transmitters go silent, a task for which the US is dispatching some specialist search gear apparently, because that's probably the only hope of giving the bereaved a chance at some closure left now.

Forget the bereaved, how on earth will the media ever get closure if the plane isn't found?

Re:Flight recorder

[dgatwood]

The ones with the skunk glands?

Badgers? We don't need no steeeking badgers.

Re:Flight recorder

[flyingsquid]

The Indian ocean is very deep, it is a remote location and two weeks have passed already. This black box will be harder to find than that of the Air France flight which got lost over the Atlantic. Back then they said that the sender of the black box will run for a month. I don't believe that they will find it this time.

There's no doubt that they'll find it, the question is when. As we speak, the remains of MH 370 are sitting on the bottom of the ocean, under 5,000 meters of water, and they're not going anywhere. Nothing is disturbing the wreckage, so it will just sit there for months, years, or decades until someone comes along. The Titanic sat on the seafloor for 73 years until new technologies made it possible to locate the wreckage, and yet it was remarkably well-preserved given how long it had been underwater. I doubt it will take 73 years- technology has advanced a lot, and continues to advance- but even if it does, the plane will be waiting.

Whether anything useful comes out of the flight data recorders or not is another issue. After 2 years, the data recorders from the Air France flight still worked, I don't know if anyone really knows how long the data would still be good. Solid state memory is pretty indestructible, so if the chips can survive being immersed in saltwater, maybe a long time. The bigger issue is whether the pilot shut down the recorders as well. In the SilkAir crash, the pilot or copilot shut down the flight data recorder and cockpit voice recorder before deliberately putting the plane into a dive. Whoever hijacked this plane seems to have wanted its fate to be a mystery, so there is a real possibility that he shut off the recorders as well. If so, we may find the crashed plane, but if so, we'll never know anything more than what we know now.

Re:Flight recorder

[TheCarp]

> I fail to see what they did wrong, even in hindsight.

That is because you are not applying the right standard. You see, anytime any effort fails, they, by definition did the wrong thing. Let me explain. Lets say, you have Ace Queen and the flop comes out Ace Ace King, then the a queen and a duece drop.

Now you have a full house, so if you bet and win, you did the right thing, However, if someone else had the king, and wins, you did the wrong thing, because you lost. Even though you had the same information available either way, and the negative outcome was extremely unlikely, by applying this standard retroactively, you made the wrong choice.

I know this may not make much sense, but that is because you are clearly stuck in antiqueted pre-9/11 world thinking where we could take the chance of allowing nuanced arguments that require deeper understanding than could be intuitively understood by a 4 year old.

Re:Flight recorder

[DoctorFuji]

As for former P3 tacco, I give the following input: Only better and slower airborne asset is a helicopter. Given the remote location, the only way to get them to the search area would be by ship (destroyer (only a few helos, or aircraft carrier (many helos, but might not be available) So, what other choices do you have? Multi engine fixed wing search aircraft are the only way you could get coverage in this area. So the balance becomes how much area you can cover with radar and visual search patterns. You could search closer with surface ships, but you are talking alot of craft to cover a similar area, plus the lag time to gather them and get them to the location.

Re:Flight recorder

[aaarrrgggh]

The CVR only records for an hour or two of audio. In all probability, nobody in the cockpit was making noise the last two hours. The FDR would have the whole flight, and will likely show the cause of the crash being fuel exhaustion.

As best I can tell, there is nearly zero chance that there was a fire that turned off ACARS message transmission, then caused corruption in the flight management computer to add several waypoints off the programmed course, then slowly proceeded to short out the transponder 5 minutes later, then caused the VHF radio to stop working immediately after handoff from Malaysian ATC, all the while not impacting the ability for controlled flight of the plane.

Unfortunately, the bat-shit scary truth of the matter appears to be that the pilot decided to kill himself and everybody else on the plane, and there really isn't much that passengers or other flight crew can do to prevent the outcome.

Re:Flight recorder

[Collective 0-0009]

While this has captured much attention, quite honestly it is a yawner. Did you know that approximately 150,000 other people died that day? I'm willing to bet more than 2500 were preventable in some manner, making that 10x more important than MH370.

When you step back and actually review numbers, many things seem insignificant, if you have no personal feelings or emotions tying you to the event. Like this post, discussion, and website.

Re:Flight recorder

[styrotech]

I appreciated the joke, but on a more serious note...

The media don't get closure, they get amnesia.