Skip to main content

How Inmarsat Hacked Their Data to Find Flight MH370

The mystery behind the missing Malaysian Airlines flight MH370 is no more. Engineers working out of the British company, Inmarsat, have used a “groundbreaking but traditional mathematics-based process” to conclude that the plane landed in a remote region of the southern Indian Ocean. Plane wreckage has yet to be found to support this notion.

Yesterday afternoon, the Malaysian Prime Minister, Najib Razak, announced in a press statement that the investigation was completed with a never-before-used analysis. However, few news articles are getting their hands dirty with the details of this mystery-solving method.

From most of the articles you’ll get a sense that it has something to do with the Doppler effect, trigonometry, a satellite and a magical mathematical equation that ties everything together. But how do these elements coalesce into a single, coherent story?

If you look at a general equation of the Doppler effect, you’ll notice that nowhere does it offer information about an object’s position in space. But you can still get an idea of an object’s location and direction of travel from the Doppler effect.

If you have an object that emits a signal, like a train blowing its horn, the frequency of that signal depends on the speed and direction the object is traveling and also on the speed and direction of the instrument measuring the signal.

Malaysian Airlines lost touch with Flight 370 on March 8. After that, the plane’s only form of communication was brief “pings” indicating that the plane was still operational. An Inmarsat satellite detected a handful of these electromagnetic signals, which turned out to be key to solving the mystery.

The frequency of each signal would change as the plane moved with respect to the satellite. If the engineer’s knew the original frequency at which the plane emitted the pings, then they could determine the change in frequency and thus calculate the velocity of the plane with respect to the satellite.

 The frequency of the ripples that the car emits change as the car moves. If you were on the left, you would measure a higher frequency than the original frequency, and if you were on the right you would measure a lower frequency. Credit Charley Whisky.

Velocity, alone, cannot tell you anything about position. But if you have a reference point in space with a known position, you can determine your target object’s position with respect to your reference point.

Say the satellite and the plane were both at the same height above the Earth’s surface and the plane was flying in a straight line away from the satellite. This simplified version breaks the problem down into one dimension. If you knew the position and velocity of your satellite and you knew the velocity of the plane, you could easily map the plane’s position over time.

Now consider the more realistic scenario that the satellite is many thousands of kilometers higher than the plane. You can still solve the position problem, but you are now dealing with a solution that requires some trigonometry in a three-dimensional space. And since the satellite is orbiting Earth and the plane is also moving, albeit at a different velocity, this becomes a very complex problem very quickly.

 Imagine the satellite sitting at the top of the sphere. Now, picture the plane flying across the surface. Notice the multiple angles and positions you would need to account for while attempting to model the plane's flight path. Credit: Peter Mercator.

I imagine that most of data-crunching marathon the Inmarsat engineers took to determine the plane’s crash site to within 100 miles, was spent identifying how best to determine MH370’s final destination given the few data points available and testing that series of equations using known flight paths of other Malaysian flights.

The single Inmarsat satellite that detected the plane’s pings measured the frequency of only eight pings in total. From that, engineers plotted the plane’s likely course across the Indian Ocean using a combination of the Doppler effect to calculate the plane’s velocity with respect to the satellite and trigonometry to then map the plane’s flight path and ultimately determine where it likely crashed after emptying its fuel tanks.

The ten Inmarsat satellites orbiting Earth were built as part of the Global Maritime Distress and Safety System and have never before been used for this type of mission, said Inmarsat Senior Vice President, Chris McLaughlin.

Comments

1. At last a detailed scientific take on this story. Thanks!

But how would the satellite have measured the change in frequency? Wouldn't its receiver only be tuned to one frequency? Doppler shift of a siren is easily noted by us as our ears are listening for many frequencies at once. Doppler shifts of stars are also measured by spectral analysis. In both cases, there is a built-in capability to hear or see the frequency shift. But not in a radio tuned to one station at the time. While it might be able to tune into other frequencies, they are likely much more widely spaced frequencies, certainly more than of the order of ppm changes.

Any aerospace experts out there able to comment on this?

1. I am also intrigued by this question.

A comment on slashdot by mdsolar (http://news-beta.slashdot.org/story/14/03/24/1611208/how-satellite-company-inmarsat-tracked-down-mh370) suggests that for troubleshooting purposes the satellites log the difference between the expected and observed carrier frequencies for each established connection. This sound reasonable, provided the delta is recorded with sufficient accuracy.

2. The comment was from strider, not mdsolar.

3. The orbit of the INMARSAT satellite (inclined at 1.66° to the equator) which leads to a latitude-dependent Doppler shift with a 24 hour period, the velocity and orientation of the plane (the moving platform Doppler shift is a function of position, speed and heading), and the short-term stability of the carrier frequency from the plane's transmitter, including considering possible variations in the power supply voltage. There will be a cluster of velocity-dependent positional solutions for each ping, which can be evaluated for a unique consistent solution for a plane's route, if one exists. If the transmitted frequency varies at source and/or if the plane's route and speed varied erratically then there may be no unique solution.
- Nalliah Thayabharan

4. "the satellites log the difference between the expected and observed carrier frequencies" - doesn't solve the problem of tuned receivers, i.e. if set to receive a signal of say 10MHz, how would they detect a signal of 8MHz to log the difference? Also WHY log the difference in the first place? One could suggest the satellite has an array of frequency sensors to detect a band of frequencies, but then we would have to ask why?

5. "One could suggest the satellite has an array of frequency sensors to detect a band of frequencies, but then we would have to ask why?"
The satellite itself moves. So every signal it receives is doppler shifted. A satellite that can't handle these doppler shifted signals is of no use.

6. @Concerned: A tuned circuit receives frequencies that are close to the tuned frequency, the signal strength that passes through depends on how closely the signal frequency matches the tuned frequency. For a doppler shift in this case the difference would be small (more like 10.01 instead of 10.00 perhaps) and would vary depending in the current motion of plane and satellite (which deviates N&S of equator by two degrees on a 24 hours cycle - because it is old and short of fuel for fine-tuning its orbit)

2. The science makes sense. The concern I have is that they used a few data points and tried to fit the best match with "known flight paths". I'm not sure it is wise to assume that this flight few on a known flight path. Not being a flight expert, I would guess that there is less options on the northern route from a "known flight path" perspective due to the national personalities in that region. In turn, there would be poorer fit with the "ping" model of northern paths.

The other assumption is probably a common altitude of the flight paths. Would the model be different if there was altitude variance from the "known flight paths." I would expect that angle and altitude put the plane at very different locations.

1. You got it wrong my friend. No one assumed that the flights paths of different airplanes are the same. What the article says is that when they put the equations together, they VERIED that it indeed was correct, by applying it to other flights of known paths. This process is common in many engineering and physics problems, and is called Validation. In short validation verifies that the math model that was used was indeed reflecting the reality, before applying it to sole a problem.

2. I was not talking about validation of the equations for the technique. My concern was focused on the assumptions of applying the equations on the problem. What were the assumptions they made about the flight path that MH370 actually took? That seems key to buying the outcome of the analysis.

3. Your are right to question the process. What InMarSat did was attempt to fit 5 data points with some margin of error to suggested "known" paths. They then decided that the suggested "southern route" was a better fit than the suggested "northern route". They did not say that the data could not fit other routes if any such were to be suggested. They did not state the magnitude of errors on their readings. And on the sole chart that's been published, it's apparent that they fitted the aircraft speed on the suggested southern route to be a better match to their data - an assumption which they are now saying was wrong, making the fit more questionable.

3. TMS, other articles are more detailed than this one, but it has to do with the fact that the Inmarsat satellite is not completely, 100% stationary, but "wobbles" to the north and south, which means both the plane and the satellite are moving. Making the calculations even more heroic. Here's one reference; there are others if you google for them:
http://www.slate.com/articles/technology/future_tense/2014/03/evidence_that_flight_mh370_crashed_in_the_southern_ocean_doppler_effect.2.html

4. That is partially incorrect. Hence that makes it partially correct. Mean of partially incorrect and partially correct information is nothing but just a possibility. Some how they gave up and used this possibility to prove a point and close a case. It is not easy to find the wreck in the so huge area especially in the bed of the Indian Ocean.

5. wouldn't inmarsat's clock and mh 370's clock need to be synchronized, wouldn't the synchronization come from ground stations, wouldn't 370 miss 7 hours of synch with ACARS off ?......could a few milliseconds put inmarsat off by a few thousand miles
mike27ef@gmail.com

1. inmarsat math is cool but did they just connect the dots ? the real ? is Australia's over the horizon radar...that good ?, at 2000 miles ?, Inmarsat took credit and profited

2. There's probably no need for clock sync. The ground stations send a handshake request via the satellite and times how long it takes to receive the reply. The satcom equipment on the aircraft has nothing else to do, so it's time to reply is probably very predictable. In which case the remainder of the latency or round-trip-time is an indicator of distance. It seems Inmarsat verified this by examining handshakes for other flights by similar aircraft operated by the same airline in the relevant period. Inmarsat's service doesn't depend on ACARS, rather the reverse, ACARS uses SITA for transport, SITA uses VHF if over land and Inmarsat of ourt of range for VHF. Inmarsat's handshakes are pert of it's resource management - they can be expected to log and monitor performance of various elements of their satellite systems for service level and anticipating problems.

6. Hi guys
I am an satellite coms engineer and to answer to a few questions:
1) the satellite transponders have wider bandwidths than the individual carriers.
2) the satellite keeps onboard data about various parameters which can be extracted via telemetry. On top of this you have CMS (carrier monitoring system) on the ground which can pic up these variations.

What I dont understand though (and I dont agree) is how Inmarsat on their Doppler shift graph shows that the offser is of the hundreds of Hz.
The Inmarsat-3 satellites have L-band transponders (1.6Ghz) so if you try to apply the doppler shift equation: shift = (speed of plane / speed of light) x 1.6Ghz then you work out that the shift is actually 1.2Hz even if you suppose that the plane was moving directly away from the satellite. Obviously a lot of engineers have check on this but I am trying to see where the error is.

1. Speed of airplane = 600 miles per hour
Speed of light = 670,000,000 miles per hour
SA/SL = 9e-7 * 1,600,000,000 HZ = 1400Hz

2. Besides MH370 moving "away " from the satellite , the satellite , being inclined was , for the first three hours , moving away from MH370, then for the final hours moving back "towards" the aircraft.

At the time the flight took off the sat was 1 deg 11 mins N of the equator
.Three hours later it was at the top of its " wobble " one deg 40 minutes north....and
by 2240 GMT it was back around one deg 11 minutes....having travelled around
the top of the figure of eight.inclination in orbit.

Here's the sat data

INMARSAT 3-F1
1 23839U 96020A 14082.92914473 -.00000008 00000-0 10000-3 0 2891
2 23839 1.6697 73.1023 0005489 286.4812 220.7448 1.00274299 65821

I assume MH370 used the older "non GPS" satellite system as it would be a cheaper deal for the airline operator.
3F1 was launched in 1996

This video http://www.satcom.freeserve.co.uk/hbs1a.GIF which I filmed about ten years ago shows actual geostationary satellites on orbit....these are stable sats , but you can see just how much they move over a period of 80 minuts....Inmarsat 3F1 would be moving much quicker , . perhaps 30 k/hour in a north south manner , 36,000 km out in space

3. David. Thanks for you correction. i missed 3 zeros in my hasty calculation :P So bearing in mind that the aircraft's tangential plane of movement is inclined with respect to the sat therefore the shift will be dependent on cos(theta) these numbers make sense. thanks again

4. ap0s: Thanks for your first post. I was hoping someone from this field could shed some light on the underlying technology. That explains it all. It's a fascinating analysis, especially with subsequent posts about wobble etc. I used to teach 1st year physics. What a practical application of Doppler shift! Profs will be bringing this one up for years to come. Maybe on exams...

7. Why are the first four blue data points after take off not matching?

1. Hi Jason
its has probably to do with the fact that the aircraft accelerates to reach the cruising altitude. as the speed increases then the shift increases (it is moving away from the satellite). Then once it reaches the required altitude the aircraft decelerates to a smaller speed

2. This comment has been removed by the author.

3. If they cannot match the data when acars and the position was known, how can they match it later?

4. http://theaviationist.com/wp-content/uploads/2014/03/BFO-MH370.jpg

8. Hate to ask a dumb question - but this looks like the right place learn why my question is dumb (and hopefully make me a little more educated). Here is the question;

What if the plane was ditched in shallow water (near shore let's say). Could the salt water impact the radio waves (depending upon the plan's depth and speed of north flowing currents) to cause the plane to (a) appear further away and (b) exhibit the same doppler effect? Why/why not?

Thank you (brutal answers appreciated).

1. Nordane, radio waves of almost any frequency, including any frequency used by aircraft, do not travel through water. (An exception is Very Low Frequency, very high power transmissions used by the navy to communicate with submarines.)
So the moment the aircraft antenna is submerged, the radio signals will not be heard by the satellites.

2. Radio waves - even UHF - do travel through water for very limited distances. They just don't penetrate far. To the extent they do travel through water, they refract in a similar fashion to how light refracts. Check out http://books.google.com/books?id=55ERaB8a58YC&pg=PA31&lpg=PA31&dq=uhf+underwater&source=bl&ots=sLo_pZnwWL&sig=_ik08U1Ih-TcTORBcUeqHtaZUGw&hl=en&sa=X&ei=mqwzU_jrMpbMsQT6poGABQ&ved=0CGQQ6AEwCw#v=onepage&q=uhf%20underwater&f=false

3. Without power to the plane's satcom it would not be sending any transmissions. If the plane was underwater, no power no transmissions (pings).

9. Jay Z in MarylandMarch 26, 2014 at 8:29 PM

Did Inmarsat say their analysis was based on the Doppler effect??, which would require their recording of the frequency of the signal and detection of a small shift. I think it can be worked out geometrically considering that the satellite wobbles N and S of equator in a figure 8. The time delay on the handshaking determines a satellite to airplane distance, and plotting that distance from varying locations on the fig 8 to locations on possible N and S flight paths gives a series of pts N and S that presumably were more consistent and realistic on the S path with a plane flying at constant speed, for example, than on the N path. Comments?

10. can anyone describe the frame structure which transfers data from Inmarsat 3f1 to ground station ? I am not sure if frequency shift is detected on board by 3f1 receiver and embedded in the frame to ground.

11. Don't you need the signal LIVE to be able to apply the Doppler effect?
Did those satellites record the frequency etc of the signal or did they just record the messages it carried (in some form of text or computer language)

12. Now suggesting a new search site because APPARENTLY the aircraft is now considered to have used its fuel faster by flying faster, AND a flight path more East than orinally thought. WOW so many variables - maybe they can solve the unified force theort as part of their (mis) calculations!

13. They have not done any calculations my friend. The whole thing is a hoax. Too sad so many scientists have fallen in the obvious trap. You need the signal LIVE to make such calculations. You need an oscilloscope or other device to record the signal's frequency/wavelength etc LIVE to do such calculations.None of the commercial satellites has such equipment on board. The inmarsat satellites only recorded the event, the data the signal carried in the form of text or computer language.
This is a huge HOAX

1. And, if it flew faster, in the range of 10% (before reaching trans sonic zone), its original presumed path could be shifted about 130 km to the left or to the right of every hourly equidistant from the satellite. This times 7 hours flight, gives a potential total shift in the range of 1,000 Km to the east or to the west.

2. Maybe during their calculations/investigation somehow they manage to analyse based on a LIVE signal from other MAS flight against the recorded signal.

I think that's possible and logical.

3. Has Inmarsat every confirmed, without a doubt, that they were tracking the MH370 plane, and not another plane, perhaps a drone or shadow 777, to entice the world to believe the plane is in the Indian Ocean, instead of within 500 miles of its original flight path where it changed directions?

14. As I'm partial to the northern path. I'm wondering whether anyone can tell me if such things as wave reflection off of the ocean surface or the (?) ionosphere occur with such microwave frequencies, and whether these might be a significant source of error in the calculations?

1. No need to seek factors such as those. The InMarSat readings could just as easily be produced by an aircraft flying in a circle, flying a zig-zag route, or flying a straight route angled differently than they currently think. At best you could do a relative likelihood calculation for each possibility with a broad set of assumptions - but they haven't done that, they've assumed what they want to believe.

15. Here is something else to add to the pot....
https://www.youtube.com/watch?v=hNZtz-HVy6c
This video shows that FlighRadar24 had another version of the plane disappearing. This time it made it across Vietnam then was tailed by another plane before disappearing off the screen.
So if the original point of disappearance is changed so would all further calculations /guesses.
Anyone have any solid thought about this scenario?

1. Any newly presumed disappearing point will quite accurately shift the presumed results (which are far from being accurate by themselves) by the same shift. That is because the small change in angle to the satellite will create smaller differences than possible shifts in speed or distances.
It doesn't make sense to me, that if there were disappearance attempts, the plane would go over Malaysian land after it disappeared, and westerly direction is the most likely it went in, at least, for going around land before it turns back to the south (if it did).
Since last spotted until the next ping there could be enough time to shift the reference point between 0 and 1000 Km, so if we presume a 500 KM westerly shift, and the previously estimated (in a response above) 1,000 Km of potential total westerly shift, the estimated northern corridor given by Inmarsat would pivot by 10-15 degrees on the newly presumed location where the first ping after the disappearance occurs
In numbers, the last position was tracked at 02:15. The last hourly ping occurred at 08:11, meaning, Inmarsat got the first ping after last known position at 03:11. That is, almost an hour after disappearing, a fact that could position the plane 900 Km to the west when it pinged. That is half way to the tip of India. If we position the corridor at this point, the estimated rout could go over the Himalaya to Afghanistan and with the possible 1000 Km possible shift due to increased speed, it would be Iran.
Another scenario is that If the plane continued further west for another ping, it would reach a point of turning north, west of India, before the 3rd ping. superposition the the northern corridor at this turning point, will lead the plane to the Gulf of Iran.
Now, the 777 is a very long range plane and some versions of it can fly 16 hours! I did not study this specific Malaysian plane and there are no details of its fuel content. I believe the plane should be able to fly more than 8 hours because to destination it need to fly 6 hours and in case of a problem it should be able to reach an alternative airport, or fly over for some time.
So, the fact that it stopped pinging after 7 hours implies that it reach "destination" rather than burned all its fuel.
Landing spot can be anywhere between the Gobi desert and the Arabian Desert, including all the mountain range of the Mammalia, then of course the Australian desert, north to Perth

16. Here are the thoughts of this stroppy old fart:

We will NEVER know the truth as to what actually happened to this flight. There are so many possibilities involving so many entities (all of which have one or more reasons for not telling the truth).

One of the most frustrating things about this incident is that all official reports give the impression they are only half-truths or 'stories' designed to contribute to the confusion. This allows other 'theories' to be suggested by every one - from 'experts' to 'couch sleuths'.

I can't remember an incident of recent times that got the attention of so many so quickly that lent itself to so many possibilities (unless of course, you want to includes Justin Bieber's recent exploits haha).

But seriously, there is a theme running though all theories of distrust. No one trusts an official, a corporation or an entity anymore - clearly because history tells us we can't.

I have much sympathy for the relatives of the passengers and crew (if there were any), but I feel more for all us good guys who just want to live in a world where we can trust those who have been given 'power' can be trusted to use it in a way that benefits us all.

We are now all interested in just receiving the truth - no matter who it embarrasses. Surely all the shills out there must want the same thing.......

17. Just found this website in search of the calculations discussed above. However, maybe someone can answer my more basic questions.

Before you can make any calculations based on doppler shift, you must have an estimate for the distance from the satellite to the plane. It has been mentioned many times that this was based on the "ping" time, i.e. how much time elapsed between the ping signal being sent out and the reply. So my questions/comments are on this calculation.

The first question is: Was this ping sent from the satellite itself, or from the ground station through the satellite?

Assuming it was at the ground station, how accurately was this time measured? We have a tolerance here, +- some time value.

Would there be turnaround time in the satellite, or is the ping signal retransmitted as it is being received? I would think the latter.

What is the turnaround time in the SATCOM transciever in the plane, i.e. the time the ping message is received, and the time it decides to make a reply, and again what is the tolerance?

Note that these "ping" messages are not designed for positional information, but only to see if there is a customer out there. It surprises me that the transit times would even be measured and recorded, but of course they must have been.

However, the total time tolerance is the sum of both the tolerances measured above. I have seen NO figures for the ping times (such as 0.23452 seconds +- .001 seconds). Note that an error of .001 seconds (1 millisecond) woud give a positional error in the predicted circle of about 1000 miles! In any case, the band is not a line, but has a definite width.

Can anyone enlighten me on this?

1. You don't need to know the travelled distance to do the Doppler calculations. They only depend on what you detect at the point of arrival. You get a Doppler effect when listening to a police car siren pass by, but you didn't send that sound originally or measure how far away the car is. Astronomers use Doppler shift to figure out if a galaxy is moving towards or away from us, which is another example of just looking at what we receive as light from the distant object.

2. Of course you do. The measured doppler is a result of the cosine of the angle between the satellite and the plane. The plane is not going directly away from the satellite, but at an angle. Without knowing this angle your doppler data is meaningless.

3. Sorry, anonymous, I stand corrected. I finally figured out what they did. However, the analysis only said which were the more likely of two possible flight paths, it did not determine the possible flight paths at all.

These possible flight paths were determined by the response time of the pings, as mentioned above, and without more information as to how these were measured, I find them hard to believe.

I also would question how they were measuring these burst frequencies. The accuracies of at least 0.00003% at 1GH would be very good. I would also question why someone would want to do this measurement and record it routinely anyway. Note this was not real-time but analysis of data that had occurred a couple of weeks earlier.

4. 35K/300,000K =0.0001 seconds.(1/10th of milisecond)
I have my doubts that the events were recorded on the satellite data with such accuracy. Yet this is what they presumably did to draw the 2 arcs...

5. I am not sure what your figures represent. Based on the published altitude of the satellite of 22,236 miles, my basic trigonometry says a point on the earth 1000 miles from the satellite position would be approximately 22,385 miles, or 149 miles away. Using the round figure of 186,000 miles/sec for c, this means it would take 0.000801 seconds more for the signal to travel the longer distance.

Now if you throw in all the variables, like the transponder turnaround time, the planes altitude, etc. they must have had some very exact time measurements to establish the lines that they draw.

6. 22 point 385 not 22385.
I was using SI units....

7. @tobert. Yes they were not real time measurements. Someone said they do record every event that occurs on a satellite, including frequency shifts of signals on a log file for troubleshooting purposes

8. I don't know how the satellite might have measured the shifted frequency, but a standard technique in laser work is to mix the sent and received frequencies. This mixing gives us two frequencies: the sum and difference. The difference (aka the beat frequency) is the doppler shift. One problem with this measurement is that the beat frequency is really the absolute value of the doppler shift. That is, you can't determine which way (toward or away) the object is moving. It is simple, and accurate, but requires another input to determine the direction.

18. Could someone enlighten me on this. What would cause there to be a difference of doppler data between the north and south routes, the only thing I can think of is air currents.

19. Another anomoly. The diagram provided by Malaysia - MH370: Burst Frequency Offset Analysis (450 knots) - is not even consistent with other data released pertaining to ping times.

We have been told that there was a normal transmission at 1:07 then there were pings every hour thereafter at 11 minutes past the hour, i.e. at 1:11, 2:11, 3:11, etc. with the last one at 8:11, or maybe one more at 8:18.

The above mentioned chart shows doppler readings at approximately 1:07 (corresponding to the last normal transmission. Then it shows doppler readings at approximately 2:25, 2:26, 2:27, 3:40, 4:40, 5:40, 6:40 and 8:14. Either very sloppy plot or something is amiss.

1. Are you referring to this chart
http://theaviationist.com/wp-content/uploads/2014/03/BFO-MH370.jpg
or something else?

2. That is the chart. Note the times (my times were UTC+8). Especially note the three points on the blue line at almost the same time (approx 1827, 1828 and 1829 UTC) with large differences. How come?

3. What can I say? Presumably they got those data from a log file which was supposed to be used in case they needed to troubleshoot their system.We don't have the log data, we don't have the exact times. Initially they drew two arches one going to the Northern Hemisphere another to the Southern.To do that you need time measurements with accuracy down to at least 1/10th of a millisecond. 1st question:Are computers able to create/save text/log files within 1/10th of a millisecond? b)Can they append big log files within 1/10th of a millisecond? 3) Is their system recording times on log files with such accuracy?

I scratched my head a lot with all that stuff and the rest of Inmarsats calcluations, until I gave up.I don't have the data, I don't have the times so what the heck :-)

I think I can explain the 3 points on the blue curve though.It seems each handshake procedure lasted for 2-3 minutes.This particular one occurred at the time the plane was turning 360 degrees hence the 3 different readings.
If this is true then the other data points are not just one reading but possibly 3-4 readings that were identical or very very close and averaged.

I also agree with you that the recordings were not consistent in time.Others 1 hour others 1.07 others more. In fact the last one is 1.5 hours away from the previous one.

While on one hand Inmarsats data is the only meaningful tool they ever had to find the plane, on the other they might have got it all wrong.
I am more inclined in waiting until the 30 days pass.
Finding the plane and the black box earlier means they will have to release information they don't need anyone to know
6 more days to go :-)

20. If at every hourly ping, this plane was traveling in a large orbit, each hour the ping occurring whilst plane southbound, what would be results? Only arc remain constant unless orbit ellipses enough towards south each rotation.

21. It's not possible to triangulate or otherwise extrapolate a direction from just a Doppler measurement. Can only show arc. Surely some wise guy can prove otherwise, in a lab. Repeatable, measure a radio frequency, an object in water, anything, moving away from a receiver, in a straight line or curved, zigzag or otherwise. It just can't be done, assumptions sure, even compared to known routes, not possible. HOGWASH!

22. In my laymen's mind. The ping frequency was measured. The signal was either compressed or widened compared to its (original signal?). That compared to inmarsats known location lets them know whether the plane is moving away from or towards the satellite. So for each ping they can publish a suspected arc. If another satellite picked up the same ping, well now there's another arc, or actually a circle. Where the two circles overlap, kinda egg shape, we narrow it down. Add a third satellite and the the intersecting circles will have a common spot, x marks that spot. Now eliminate satellites two n three n explain how u can compare those ping fluctuations to known pings traveling in known directions can do the same as triangulation. Post a YouTube animation of your best interpretation of this pseudoscience, it's not possible.

1. 2 or more sats then the arcs would coincide not cross.
Comparing with known pings means they also had the amplitude of the signal (it's loudness let's say) not just frequency shift.
Who knows, they don't tell much.

23. And y is it so convenient only one satellite happened to pick up ping and record it. And if it recorded it y is there no analysis of its full data, ie; s data. Maybe some old analog satellite recorded this ping as well.

1. The ping is not done by the satellite, it is done by a ground station in Australia. The satellite just amplifies and retransmits signals between ground stations, in this case between ACARS station in Australia and SATCOM transceiver in plane.

24. If the plane did in fact end up in the Indian Ocean then my personal theory is this a)Killed all passengers/pilots/crew by going to 45000 ft, depressurizing the cabin.b) Descended very low over land possibly Malaysian land or adaman islands or Maldives.( Needed land to drop down something + the hijacker/s). c) Had programmed the autopilot to take the plane to it's grave.

25. Sort of by accident I found a thread on a website where a professional pilot from Maylasia was asking for information on how to change the airplane identifier in a 777 ACARS system. Although this is an old post, it is an interesting question. Why would he want to do that? The URL:

http://www.pprune.org/engineers-technicians/472227-boeing-777-acars-aims-user-guide.html

All I could find out about the poster was that he was from Malaysia and was a BMW enthusiast. I would think the FBI or another organization would want to check it out. Does anyone know how to get them this information?

1. Send a message to Mr Hussein the Minister of Transport of Malaysia he is on Facebook.He will relay it. Or login to that pilots forum and spread the news. I am sure they will tell everyone.

2. Or contact AMSA directly
https://www.amsa.gov.au/contact-us/index.asp

26. Has there been any validation of this method by corelating the analysis to known aircraft ? Seems the analysis is very questionable and is being published as exact science. At the speed of light any small error is quite large of distance.

27. From reading up on Inmarsat I concluded that the signal (ping) was started at the earth station, probably in Australia, and that an aircraft is assigned to only one earth station at a time. If the plane doesn't answer, the earth station sends out a signal every hour through the satellite and makes a handshake with the equipment in the plane. This is the part of ACARS that they didn't turn off. The same happens when two dial-up modems start communicating. Yes, the bit rate used is very low like we had it back in the sixties, about 4800 bits per second for a pulse width of about 0.2 ms. The periodic bits act like radar pulses sent over the system and back to the earth station. There one could precisely measure the pulse delay and the Doppler shift of the pulses. The carrier frequency of the pulses is changed many times in up and down-converters but is eventually brought back to exactly what it was when it started the trip, assuming no Doppler shift occurred. Measuring now the frequency shift gives the Doppler, and measuring the pulse delay allows you to calculate the distance of the plane from the satellite which is a circle on the globe. In conventional radar a pulse width of 0.2 ms would give a resolution of about 60 km but this can be reduced by at least a factor of ten by comparing the shift of the transmitted and returned pulse shapes on a scope. The night of the event there was probably nobody with a scope looking at waveforms but it is conceivable that delay and Doppler are measured automatically and recorded by the existing equipment. At least it would be easy to implement this in future equipment.

1. You are correct about what you say above. I spent some time with an expert this morning and he assured me that the frequency and time measurements (which are made at the ground station in Australia) are possible with todays technology. The basic problem I still have is with the turnaround time in the aircraft. Lets say we measure very accurately the time from the starting edge of the first ping character to the starting edge of the reply. A ping would go like this:

A. Ping message sent to satellite
B. Satellite amplifies and rebroadcasts ping
C. Message sent from satellite to aircraft SATCOM receiver
D. SATCOM determines it is for ACARS and routes it to ACARS unit
E. ACARS determines it is a ping and formulates a reply
F. Reply is sent to SATCOM
G. SATCOM transmits reply to satellite
H. satellite amplifies and retransmits reply
I. Reply is sent to ground station
J. Ground station receives reply and records transit time.

So the transit time measured is the total time for items A through I. What we need to know, in order to determine the circle where the airplane may be, is time C and G (which should be the same), which determines the hypotenuse of our triangle. Now items A, B, H and I are precisely known since the positions of the satellite and ground station are well known. The possible variables are in D, E. and F. Without knowing the internal software of these units, I would have no idea whether this time is always a constant, or would vary depending on what else may be transpiring at the time in the units.
Now they presumably had a ping at 1:11 where the position of the plane was known, therefore C and G was known, therefore the DEF time could be measured and used in later calculations. However, if the first ping was at 2:11 it is much more iffy, since they did not know exactly where the plane was then.

Anyway, I have seen no published data that indicates the tolerance associated with the ping times. This tolerance would indicate the width of the arcs they drew, and as I mentioned before, about 0.8 millisecond error would translate to an error of about 1000 miles!

2. Cyril Alexander @Cressidom = twitter?

3. I believe that you are correct that the frequency, which translates in to the average measured pulse width, must have been measure and recorded at the satellite in some form of log record that was sent back to the ground station. The resolution of the frequency measure could be very good. I did hear a mention of nanosecond resolution on the news although I cannot quote a source. The overall response time as measured by the satellite is highly dependent on the response time from the plane. If high level software is involved in the ping response, this could be highly variable even with a higher end processor. Add some type of custom embedded OS and you could be talking 10s of microseconds slop in the system even if the response was formulated immediately.Anonymous (who is not myself) is correct by my calculations, toberd, you must be off a bit with your .8ms versus 1000 mile calculation. I believe that .8ms would equate to about 149 statute miles assuming speed of light travel. Remember that you have to divide .8ms by 3600 to convert the time to hours before dividing in to c.

4. I should have added, Inmarsat gave a figure of about 310 miles (statute miles) of slop on either side of their projected flight path. That should give you some idea of the possible variability in the overall time stamps. Given the glaring errors in the ping data and the predicted Northern and Southern routes, anything out of Inmarsat should be viewed with a great deal of skepticism. Keep in mind these are not the guys that designed the Inmarsat system, they are the users of it. The designers of the system have expressed interest in seeing the raw data that has not yet been shared with them. That being the case, the statements from the Prime Minister of Australia must be viewed with a skeptical eye in that the best and brightest (designers of the satellite system) have not had a chance to comment on the raw ping data. Until that happens, expect the search areas to move on a daily basis.

5. In reply to the 4/3/14 6:04 note, my .8ms is of course about 149 miles. However, what we are measuring is the hypotenuse of a triangle that has one side of 22236 miles. For example, a right triangle with one side of 22236 miles and the hypotenuse of 22385 miles would work out a base value of 2578 miles. Of course we are dealing with spherical trig here, so the analogy is not absolute, the point is a difference of 149 miles in the hypotenuse (which is the distance from plane to satellite, would make a large difference in the radius of the circle plotted for the plane. 1000 miles is actually very conservative.

28. In reply to tobert, April 1, 5:31 PM. The delay of DEF is a well known parameter of the circuit design and shows up in any handshaking operation of two digital modems. It could certainly be easily measured as a fixed number. It is a characteristic of this particular "radar reflector". As I said before, the pulses are about 0.2 ms long which is 60 km in space, but the measuring accuracy is much better if the shape of the returned pulses is matched to the shape of the outgoing pulses in the Australian earth station.

1. I don't know where you are coming from. DEF is not a well know parameter. It depends on the software design and function in two different units built by two different manufacturers. Anyway, the DEF time has a tolerance of +- x, and we have not been told what this is.

All we know is that they calculated the circle diameter based on the total time measured between transmission and reception at the ground station. Note that each phase of the operation, as I described above, has a tolerance. And the tolerance of the estimate of position is the sum of all of the individual tolerances.

All engineers know that all measurments come with a tolerance. The estimate used to draw the circles is some value +- x. If we are not given x, and the values used to compute x, we don't know whether the information is valid or not.

As I said before, if x is 0.0008 seconds, the error is about 1000 miles.

By the way, we are talking about a digital communications link, and pulse width and pulse shape have nothing to do with it, and there is no "radar reflector" involved.

29. Cressidom posted curious question, dude just a atc techie, bimmer owner and pet owmer

1. I thought it was very curious also. Presumably the only 777s running around in Malaysia are owned by the airline. If he was tech working for the airline, why would he have to ask a question like that on something called a "rumor network"? Could be a completely innocent question, but also could be that even back then (Dec 2011) someone was thinking about the same situation we have now and doing some investigation and planning. Possibly if guy just a techie someone asked him the question and he, not being familiar with the ACARS system, decided on this route in order to answer his friends question. In any case someone should check it out.

Cressidom was probably not on 370 himself, as he has made posts subsequent to the dissapearance.

30. For each hourly handshake there will be an arc, the length relative to the time of the last known radar signal. So there exists a series of arcs of ever growing length that will be of finite thickness relative to the vagaries described in this thread but relatively thin. It would seem that a review of these arcs with the doppler data would be informative and I presume that the analysis of these arcs are why they have the various speed paths coincide until they move to the last arc when we do not know the end length. Has anyone seen these arcs posted?

31. If you could change call sign , Id of plane in flight, wow, who are you tracking now? Has anyone seen flight data of flights on radar that night. Are any not correlating to schedules? This curious question dude also tweeted about aircraft sales!

32. It looks like Inmarsat HAD published all the data in their website in a pdf file, which later on they removed.
Fortunately somebody saved the file. It's all here:
http://www.duncansteel.com/

Numerous calculations and graphs....

33. As I asked elsewhere - As retired airline pilot in easy communication with several current 777 drivers, I wonder what the chance is that the handshakes were just random noise having nothing to do with 370. We think the airplane suffered an explosive decompression caused by a windshield blowout during climb. This would convert the flight deck from a calm environment into a hurricane at many degrees below zero making finding and donning oxygen masks problematic. Soon the flight crew was dead, having disengaged the autopilot during the circus. The airplane had a climb power setting and was trimmed to climb. It continued to do so through its assigned altitude, above its maximum altitude for its weight with fuel for Beijing, and far above its maximum service ceiling for any weight, anytime. It stalled, pointed its nose at the Gulf Of Thailand, and went into the water at above the speed of sound. Wreckage will eventually be found in tiny pieces on beaches in the Gulf. It is impossible to explain the airplane leveling off and flying unattended for several hours after it suffered whatever failure caused it to dive 40,000 feet, from 45,000 feet, in one minute as was first reported.

1. Being totally naive about the way autopilot programmes are written: is there any possibility that a destination could be calculated along a reverse vector, i.e. at 180 degrees to that intended? If the crew were incapacitated somehow as you and other 777 pilots suspect, could the plane have flown itself on autopilot to the reverse vector location, the same distance as Beijing from KL? By rough measurement on atlas maps the resultant location would be in the range of the Inmarsat predicted southern corridor. If there is any credibility to this suggestion, the equation used in the autopilot software could be used to determine an accurate location for potential searching.

34. It appears to me that the analysis of satelite data based on wobble of satellite would also have a solution in the Northwest quadrant in addition to the Southeast quadrant. I think it is a mistake to assume that the aircraft would run out of fuel. Based 777 max landing weight the aircraft could have had much more than 8 hours of fuel on board. Price of jet fuel in Beijing is quite high.

35. The pilot was known as a techie, is it possible that he knew about the satellite ping? Why did he not disable the ping output from his plane, as he did with other communications? Is it easy to fool Doppler ping data, like adding a simple delay to the ping, or by using altitude and speed?

36. @JDBishop5.
Most pilots are looking for a standard explanation in the form of mechanical failure, or fire etc.I doubt this case will ever fall within that Standard explanation.
Btw the plane was tracked my military radar of Malaysia + 3 other countries returning back in a zig-zag move.

37. The calculations Inmarsat are fraudulent. The original observation indicated that the plane flew for 7 hours without changing the distance between the satellite and the plane and interpreted that to indicate that the plane flew along that circular path to the south, This initiated the search in the south Indian Ocean. The Doppler data disagreed with the original conclusion so they picked the one of the 14 possible paths that supported their original erroneous conclusion and the search continued. For details see marvinsthinking.blogspot.com

38. This comment has been removed by the author.

39. This comment has been removed by the author.

40. How would changes in altitude at time of ping effect the Doppler calculation? If heading away from satellite while increasing altitude or heading towards while descending, could the results be effected sine we do not know the variable of altitude changes?

41. This comment has been removed by the author.

1. You wrote: "Inmarsat don't able to determine Distance to the MH370 by its ping. ACARS-SATCOM SYSTEM have INDEFINITE INTERNAL DELAY that don't allow such calculations."
How do you know what internal delay the system has? Is it a random parameter?

2. This comment has been removed by the author.

42. This comment has been removed by the author.

1. This comment has been removed by the author.

2. This comment has been removed by the author.

43. Emil, I listen, "purposeful lie" "distortion and adjustment of truth" . These are your unrelated quotes. I believe we are being duped, at a mass scale. Why, anyone's theory plausible. Interesting Inmarsat launches new sat phone since incident. Also launches ability to pinpoint from a singularity not designed for this with convenient recordings. Sat made in 90s. No other sat picked up this data. We are led to believe. We don't discover. Proof is offered up. Or is it covered. Way too many distractions. So much SAR effort we will no doubt find a debris field in south corridor they searching now. Did it crash? Is it 9hmro? You can record thousands of flight peramiters for 25 hrs but voice for only 2? 2 stolen passports. 2 Iranians. Chinese and yugiers

1. This comment has been removed by the author.

2. This comment has been removed by the author.

44. MarvinVestel. Inmarsat did not say the plane flew in this arc. What they said is, in this arc is somewhere where the plane was last detected according to ping data. And this is plausible. It's possible if they have a known location of plane with a known ping frequency. And it can only detect how far away from the satellite the pings are, thus you get an arc. One can only postulate that they only show two partial arcs rather than a Full circle with the sat at center because a; they searched the Malay area already b; anything further to west out of fuel range or they don't want to find it there? Burst frequency is NOT an exact science! It's never been done before or recreated to use Doppler shift to pinpoint like they are claiming to do. You r correct to doubt because it is hogwash.

45. "THEY" do not tell us all info! They say 7 pings every hour at 211,311,411,511,611,711,811 and an additional at 819. So y is there not a ping in first 2:11 minutes. At 1:11 there should have been a first ping at lat 5.65 long 102.97 or somewhere close there of. Do your Doppler math, this gives you first ping arc. Now math out the distance travelled from kl . You should be able to determine speed, flight radar help u. At 2:11 somewhere in Malay peninsula. (first?) ping. Fast forward to arcs "THEY" show us. Guesstimate missing arcs. Does this jive?

1. INMARSAT published a graph that shows time shifts of the pings. The time dots are not exactely ploted at the expected time x.11 dots. As far as I understand their method is to tripple hand shakes for security reasons. However, a hand shake last approx. 2 min. each. In my opinion INMARSAT has ploted only the time dots to justify its arbitrary assumption of the south coridor.
Am I right?

2. This comment has been removed by the author.

46. This comment has been removed by the author.

1. This comment has been removed by the author.

47. Congrats to Emil and Marvin for their detailed Physicist-like posts and work.
Presumably Inmarsat got all their data from log files that are usually stored on sats for troubleshooting purposes. They never released that information to the public, and the Malaysian government says they are "sealed". Physicists around the world could have not only checked the validity of their conclusions, but also provide the percentage possibility for each and every scenario. They haven’t ask for anyone’s help so what can we do?

With all this stupidity going around I wouldn’t exclude the possibility that the Malaysians or the Australians did in fact pass all that data to a group of Physics professors for advice. Who in turn said "hmmm…we don’t know" or "hmmm… interesting" or "hmmm…. possibly". These are very common responses from Professors. Then again they probably did check everything scientifically and confirmed or even refined the results. WE WILL NEVER KNOW….

I suggest we leave aside the Inmarsat story for the moment and move into something new
http://edition.cnn.com/2014/04/09/world/asia/malaysia-airlines-plane/index.html?hpt=ias_c1

The first signal, at 4:45 p.m. Perth Time on Saturday, lasted 2 hours 20 minutes.

The second, at 9:27 p.m. Saturday, lasted 13 minutes.

The third signal was picked up Tuesday at 4:27 p.m. That lasted 5 minutes 32 seconds.

The fourth, at 10:17 p.m. Tuesday, was 7 minutes long.

4 Signals so far of so much substantial duration and they have not yet been able to make triangulation and pinpoint the exact location of the source?
What prevents them from using the Doppler effect principle in this case?

48. This comment has been removed by the author.

1. This comment has been removed by the author.

49. Of course they aren't. BUT don't forget that's a computer simulation, the real signal is not even audible to the human ear.
From their own data:12Khz windows, centered at 37.5 spike is 32.3 Khz.

1. Also if you notice, the guy has a sidebar on it's PC program that shows the REAL pulse duration, but it is so small we can't read it. Please have a look at the frequency spikes and pulse length, they could well be within specs.

2. This comment has been removed by the author.

3. This comment has been removed by the author.

50. I have constructed a 10000 Hz audio file square waveform 95ms duration, using Nero wave Editor version 3.8.3.0,. Almost all my audio programs can play it including Windows media player, and yes it is quite audible and irritating like you said. Out of that I made another 3 files 30, 20 and 10ms.
Download link:
http://www.2shared.com/file/j5oo3vnD/MilisecsAudio.html

The 30ms one cannot be played with any other program other than Nero wave editor and it it still quite audible and irritating like the previous one. The 20ms, and 10 ms ones cannot be played with any of my programs, it seems my computer or the programs are not fast enough. I wonder if anyone else in this forum could try them. Anyway judging from the duration of the 30 ms one, I am sure my "human and a little deaf" J ear would be able to hear 1/3rd of that, 10ms sounds i.e

Turning myself temporarily to becoming the advocate of the devil, I would -for the moment- excuse the Australians of not having fast enough computers or software to hear those 10ms signals without some sort of modification. We will see what happens in the next few days.

Btw where did you find that those beacons signals pulse is of 10ms? Any link please?

1. On 2shared.com, press the small "download" button not the big one.

51. This comment has been removed by the author.

52. Thanks for doing that demo on youtube, it wasn’t necessary, it was obvious the pulse of the signal was huge. I am still reluctant to call it fake though, I mean it carried a lot of noise, and we do need the original signal to draw safe conclusions. Having the original signal we could see if at the spike frequency of 32.3Khz, the PURE pulse duration was that high or not, because the PURE pulse would be quite distinguishable in the wave pattern of the signal.
Patience my friend. We will find out in the next few days.

1. This comment has been removed by the author.

2. This comment has been removed by the author.

53. Why Emil deleted all his posts?

1. This comment has been removed by the author.

54. Emil enchev, Sofia , Bulgaria . Where did u go? U like a clue to mh370. U appear than fade away. U have no data confirmed about u. Now the report is this plane skirted indo at 4000' . Is it possible to access flight dr and cockpit vr while in flight and ditch? Is it possible to scan plane for signals sent out from plane in flight such as engine pings and cut source after discovery, maybe even discover thru search effort reportings or computer programming? Looks like black boxes may be found, but is the plane there? What are scenarios to support this? Too much effort to evade detection to dump deep sea without debris. Nothing adds up here, . Inmarsat data is convenient, why is that?

55. Emil enchev, from Sofia Bulgaria. Occupation unknown. I looked him up in web. Dude joins a variety of blogs online. Seems to wait and read. Picks a blogger and attacks his postulates with sometimes incoherent psycho-babble( prob due to broken English), yet at times more articulate than authors themselves. Some topics range from mh370 theories, complex mathematics and psychology . I assume he is self read intellectual, his paranoia is fed from his discoveries that mainstream ideology is propagated from underlying alterior motives. Dude has almost completely erased all evidence he ever existed. Why? Come back Emil Enchev!

1. This comment has been removed by the author.

56. Dude is problem solver, mathematician. Highly logical thinker. Maybe his posts are gone to silence him, but by whom? Why?

1. From ANOM.
He deleted them and the reason (for those who read them) was inside his first post. Do you know the poem that says "anything you say can and will be used against you" :-)
Emil you should have written anonymously man!

57. This comment has been removed by the author.

58. This comment has been removed by the author.

59. From ANOM
I will call myself ANOM from now on, cause there are too many anonymous in here.
I am the one who said "Congrats to Emil and Marvin for their detailed Physicist-like posts and work." I then exchanged some correspondence with Emil about the signals from the beacons.
I personally appreciate the man, the same way I appreciate everyone who can contribute something scientific and logical.
Emil I was looking for the link you gave me regarding the Manufacturers specs for the Black box beacon. You deleted everything and can’t find it man!
Could you please post it again?

60. I am curious. If the inmarsats final calculations are based on the known local of mh370 earlier in its flight? After all, the last truly known local was where this plane shut off its acars. Everything else has been assumptions based on assumptions have they not? So if the last truly known ping location matched radar just north of Malaysia, is the Inmarsat data remotely possible?

61. Did this blog close or did u all get silenced

62. So just how many tonnes of Avgas (aviation gasoline) were loaded onto this flight MH370 and with that tonnage amount, what is the absolute maximum range that could be flown by MH370 including (very unlikely) tail-winds ? ? ?

63. JDBishop5April 4, 2014 at 5:34 AM.

You say,"It stalled, pointed its nose at the Gulf Of Thailand, and went into the water at above the speed of sound. Wreckage will eventually be found in tiny pieces on beaches in the Gulf".

Lottsa fishermen in that small area from the countries littoral to it; Thailand, Malaysia, Vietnam, Cambodia, meaning the unsinkable floating stuff would already have been found.

But, yes, it is in the Gulf of Thailand, intact, brought down gently by remote control.

Why ? Malaysia convened a War Crimes Tribunal last year which brought down the verdict that Israel was guilty of war crimes and genocide in Lebanon in 1982 and in Gaza in 2008/09. Nuf said.

64. cant they just re-fly a 777 down the route and adjust the route until the ping data from this plane matches up????

65. Inmarsat's ping are on the engines. These ping's are not on the plane. Why would pings locate a plane that had crashed and wings torn off thus engines ping are of no use because they could be anywhere. Beside's the plane has been found here is Proof: http://www.lagrandeobserver.com/News/Local-News/Mystery-of-missing-airliner-solved

66. All of this is is incorrect. You all have violated the fundmental premise of science. Claiming hyposthesis is fact before there has been any validation. Unfortunitely this has sucked in all of the investigators.

67. How do we know that the frequency of the transmitter on the plane didn't vary? The described method seems fairly sensitive to accurately knowing that frequency, while it seems unlikely that it would be that accurate through temperature changes in flight.

68. This is one of the best discussions I have seen about the subject of the Inmarsat data. One more thing to add- the possible refractive effects of layers of cold dense air and warmer less-dense air on the direction of "radio-wave" transmission.

On a tangential subject, can the wave length/frequency of an electromagnetic signal change when it passes through colder/denser or warmer/less-dense air?

1. Exe summary: at 1.6 GHz, the changes in the Electromagnetic signal as it goes from the ground to satellite is probably smaller than the other sources of uncertainty.

Although optical light can change due to colder/denser – warner less dense air, radio frequencies (RF) are not effected. However, what does affect RF is variations in the free electron density in the ionosphere. Low frequencies are delayed more than high frequencies, the delay scaling mostly as the frequency-squared. So the signal does not move at the speed of light and different frequencies move at different speeds (causing dispersion). In simplified situations, these differences in speeds can be estimated from the Appleton-Hartree equation knowing just the integrated line-of-sight column density of the free electrons (the so-called total electron content, or TEC).

An important consideration is that the RF does not get to the satellite in a straight line. Think of the ionosphere as layers of an onion where each layer has a different electron density. Each layer will have a different index of refraction (given by the Appleton-Hartree eq). Applying Snell's law, the ray bends at each layer. Compared to the line-of-sight (LOS) path, the actual path that hits the satellite starts out on one side of the LOS, bends towards the LOS as it goes through the layers with increasing electron densities, crosses the LOS, and then bends towards the satellite as it goes through the layers with decreasing electron densities. This refracted ray can have a significantly longer path, increasing the time delays.

How strong these effects are depends on the time of day because the sun causes enhanced electron densities near local noon. One can get the historical electron densities from NeQuick2. However, the pings occurred at night when the longitudinal electron density gradients are usually small. I hope when they used other aircraft to calibrate the delays, they did it at the same time of day.

I've spent the past few years writing a code that calculates all this. It can be used to locate lightning since lighting makes electromagnetic pulses which are much like pings. I noticed the two line element (TLE) for the satellite was given above. That is one of the inputs for my code. Does anyone know the coordinates of the ground station?

However, since the effects scale as frequency-squared, I suspect the corrections are very small at 1.6 GHz, probably at most tens of microseconds whereas other sources of errors in the delays are probably 0.1 millisecond.

2. Ray banding is well known to terrestrial radio engineers, where it can have a significant effect on microwave reception. Also the speed of light slows down in the atmosphere, and this will effect the time taken for signals to get from the plane to the satellite, which in turn will effect the estimated location. I'm not master physicist but both these effects should be factored into the equations.

3. bending not banding !!!!!!!!

69. Updated BFO analysis here:
https://dl.dropboxusercontent.com/u/32349391/Inmarsat_BFO_Analysis_2014-04-27.pdf

70. First of all the amount of fuel that was loaded into the airplane,should be determine, this would determine the the true flight range of MH370

71. 1st off, the plane can not emit any signals without power for the engine health check to take place, so therefore Boeing has no additional data. 2nd, Inmarsat can not determine any position of the plane after the ACARS has stopped responding to requests sent by the ground station to the plane if the plane can not respond. No more pings will be detected and all the additional pings are strictly fabricated. Transponder ping signals are connected to the black boxes and should not be confused with the ping's from the engines or the communication link's between the ground stations and the plane.

72. I have been an engineer for 40 years and I'll only say this. I have seen many cases where a small, incorrect assumption at the beginning of a message becomes group think gospel, causing supposedly "iron clad" information, accepted and even peer reviewed by many to be completely false.

73. Hi folks, you need to ask the right question. The right question is: how INMARSAT gathered the 14 numbers instead asking how to calculate with them.

You have to understand the method of the data collection as a first and most important step. Then one be able to judge if the "raw data" are valid only then you start to compute with. Otherwise you finest calculation is a game with numbers.

Anyway, the raw data are useless as the entire INMARSAT COM system does not allow to get data in real time. Raw data must be repeatable and valid, in this case 1-F3 SAT and the SATCOM on MH370 has a load dependent delay time of 320ms. The result of such a calculation leads to random numbers far away from reality.

74. This comment has been removed by the author.

75. This comment has been removed by the author.

76. This comment has been removed by the author.

77. This comment has been removed by the author.

1. This comment has been removed by the author.

2. This comment has been removed by the author.

78. This comment has been removed by the author.

79. This comment has been removed by the author.

80. This comment has been removed by the author.

81. Emil, again the so called experts do not ask the right question. Before i am interested in data i compute with i must be sure data are valid and repeatable. Mean, if its useless data I only produce more useless data. The big question is delay time between SATCOM and ACARS. Always the same? I minor error in variation leads to eliminate the calculated Doppler.
Did you ever answer this question beside the valid question you have asked?

82. This comment has been removed by the author.

83. This comment has been removed by the author.

84. This comment has been removed by the author.

85. This comment has been removed by the author.

1. This comment has been removed by the author.

2. This comment has been removed by the author.

86. This comment has been removed by the author.

1. This comment has been removed by the author.

87. This comment has been removed by the author.

88. This comment has been removed by the author.

1. This comment has been removed by the author.

2. You don't know me Emil..but Now is not the time to get angry..but to get even. The report looked pretty cleaned up and dummied down. You are up against Boeing and Immarsat not to mention their active interests they are shielding with the NRO..they work together..This is fact. By leaving like this..you do their dirty work..Don't.
There is no room for emotion Emil..Just shed light so proper questions can be asked.that the Journalists are not asking..
I hope you reconsider and come back..

89. Dang Emils gone..what a gift he has..

90. Treat everyone with politeness, even those who are rude to you – not because they are nice, but because you are.

91. This comment has been removed by the author.

1. This comment has been removed by the author.

2. This comment has been removed by the author.

3. Yes they are Emil.. he same guy on CNN Richard Quest who flew with the Young Copilot Fariq..
about 2 weeks before the plane went poof.. I am curious about the hangman's pose .
http://images.says.com/uploads/story_source/source_image/262828/6197.jpg

Quest was arrested in a park caught with a rope around his neck tied to his genitals. amazingly he was still kept on. Imagine the physics of that..This is true of many in the news..and in power..

Now..the search is off...for months.. Boeing can finish making its satellite for Immersat..and finish selling its BUA prepped planes and the rest its toys. Raytheon which provides Network centric radar tech to Malaysis and others in the region for the last few years as well as the Veri system from Chzecs can go back to play with their toys..
War games or not there were 40 active radars..in the region..yet..nobody's seemed to be looking at the sky that night.
Sherlock Holmes who did did not have a slide ruler or calculator..said .rule out the impossible and what is ever left must be the truth..
Its not over....as they say...the Devil is in the Details..
Regards
KuriousKat

4. This comment has been removed by the author.

5. This comment has been removed by the author.

92. ,Welcome back Emil..
,its apparent even from that the source site of the following that what some of us were waiting for ..a raw steak..was a Very well done steak ..to use the exact words of malaysian authorities for "public consumption".

https://www.metabunk.org/threads/inmarsat-mh370-raw-data-released-unlocked-xlsx-csv.3721/

locked
https://www.metabunk.org/attachments/flight-370-data-logs-pdf.7431/

unlocked..allows for cut and paste just like the pros..
https://www.metabunk.org/attachments/flight-370-data-logs_unlocked2-pdf.7434/

excel version with highlights in yellow for comment by the MB admin

https://www.metabunk.org/attachments/mh370-data-log-xlsx.7435/

Bon Apetite!

93. Emil,

Would you be willing to share your theory about what went down with this plane? Would be interested to hear, you seem to have a firm grasp on everything scientific, technical and the likes on the whole situation.

1. This comment has been removed by the author.

94. Maybe this might help in theory
http://i215.photobucket.com/albums/cc191/sys_config/sys_config172/ScreenHunter_04Apr221905_zps43d34f13.gif
Sonar layout of areas by the cbto..add 1400 others dropped in wrong place
http://www.ctbto.org/map/#

satellites flying about..8000+ known military 500+
http://www.n2yo.com/satellites/

now..

Google network centric warfare Raytheon check out some of the videos.or even Saabs ...these systems during wargames depend on air satellite marine surface ..they also can distinguish like Raytheon-Boeings also Immarsat partners any plane flowing low level..stealth and in shadow of another plane like an airliner..They can jam and commandeer if necessary..

If shot down someone is buying time to investigate how this happen given the equipment can distinguish..foe and friend..that means someone did not operate correctly..or the equipment did not operate correctly..We know the MH370 Pilot knows how to operate correctly his end. During American Thailand Cobra other countries present..including China as observers.over 17 ships and 40 plus radars engaged...These are all client states of these companies..
This is billions and billions of dollars in current and future contracts..as well as reputation.

If intentionally shot down..that raises other problems such as not only passengers but cargo as well. It would have to be tremendously important to justify..
Accident is more probable..
imo this is why you have the silence..until amongst themselves until they pin the tail on the donkey.
Even though the Allianz Insurance company paid out..it can come back and collect from them..
Remember the back and forth during the USS Vincenns shootdown of Iranian Airliner in Iranian waters..almost 300 people 60 children..died..The US admitted after legthy investigations and settled..but never apologized ..in fact..The Capt later got a medal
Such is the world..
whether this is what happened or not..Emil is correct on the PDF..they should have titled the report....How We Spent our Spring Break..because it is not Raw data , using 19th century wave analysis technique that the sat was not intended for.. ..I would not doubt even Rhune stones..

Regards
Sys

1. This comment has been removed by the author.

2. This comment has been removed by the author.

95. This comment has been removed by the author.

1. You can't say interesting things like that and not expand on them..Tell us..Tell us all

2. This comment has been removed by the author.

96. I thought it interesting the parallel between mh370 and alitalia 870 both incidents involved nearby wargames..its stilled debated bomb or missile.....I thought the Russian General who commanded the radar that saw the enire thing was interesting. All the witnesses that were involved that knew something...are dead..
there we had debris radar images of fast fliyng objects..here 30 years later...instant tracking and identification network centric warfare ...some really astounding videos by the way from Boeing..Lockeed..Raytheon..
Of course if something goes wrong you can't find them
and all we get is a pdf with stick figures

http://en.wikipedia.org/wiki/Aerolinee_Itavia_Flight_870
Regards
Sys

1. This comment has been removed by the author.

2. This comment has been removed by the author.

97. As much as I will miss your view. until then .it is a very good posture to take Emil..
I believe ..there are still good people and cops left

Peace and Justice

Sys

98. new witness sailor claims saw plane fire..provides sailing charts etc..
http://www.phuketgazette.net/phuket-news/I-thought-saw-MH370-fire-says-Phuket/29654#ad-image-0

also claims 2 aircraft higher up..

99. just enuff detail to make plausible..my problem whith at is she saw a fantastic scene..like a ufo..and did not wake husband..or even the crewman..then 2 months later the husband provides the detail of charts.. like the pdf..looks good..to be truthful..I am not convinced too much time pass..too much detail..I believe the island people that saw plane fly over..before I believe this one..

I dont think it was any of flight crew.but might be.If there were terrorists on board..it would be with passengers..2 we know had Iranian connection...and 2 or 4 others last minute standbuys..and we know the passports not scrutinized good.. .most terror cells are about 12 like disciples at most.....and probably a few about 8 went in disguised as mangoteens..(hahaha)..I need to know about access to cargo from passenger cabins. or cockpit or cabin crew .suppose somebody can set of a smoke device . fake .passenger would yell smoke.. of sorts that would force captain or engineer to check..the cockpit door opens..boom its over..suppose everything they need was in the mangoteen crates..explosives..guns.phone jammers...etc..
The fake passengers let others in like a trojan horse..very quietly..
and if cargo inspections were lax..its late at night..pre flight lax. as capn was called in last second....who knows..

but easier way..
[color=yellow]http://boards.straightdope.com/sdmb/showthread.php?t=705576
Join Date: Mar 2002
Location: Covington, WA
Posts: 5,324
The 777 has an access hatches to both the forward and aft cargos. The forward hatch is just aft of the flight deck door, the aft access is between doors 3R and 3L. The 737, the plane I have worked on the most in my 30 years at Boeing only has an emergency access panel to the forward cargo. The gain access, the carpets would need to be pulled up and possibly a seat moved to gain access. It is something that would be difficult, if not impossible, to do while the plane is in the air.
Reply With Quote

GargoyleWB GargoyleWB is offline
Guest

Join Date: Mar 2001
The really fun "secret" route on the 777 is the escape hatch from the aft overhead crew rest area (bunks and private area for the crew). One of the stowage bins above the passenger seats is a dummy (locked on the outside) that is really an emergency escape from the crew rest. An attendant could, from inside the crew rest, open the stowbin door, flip out a little ladder, and climb down right into the lap of a passenger "Don't mind me. sir, I'm just a carry-on"
Reply With Quote
[/color]

so we know accessible..it would be good to know what rest of cargo 2 tons was..I saw these large containers which could hold a few people each..

people say ohh but nobody took credit! well nobody took credit for the nerve gas in Syria either did they.. 8-) and if there was an attempt to commandeer...it may have been a tug of war..override the override..and why plane zig zagged.. being that communications jammed the only signal would be visual from a fighter escort ..wagging the tail ordering it down..that would tie in with Sara Bajics statement...

and with this search and pings remember flight 007 south korea..accidently or deliberately..over Russia..cold war was very high then.. maybe little like now..but Russans set false pings to throw off West..because they wanted to get plane first..
Here China running around ocean with its own pinger..if water had touched that....then we have more false signals..

100. This comment has been removed by the author.

101. This comment has been removed by the author.

102. Old coldwar tactic..invented by Russians..starburst..confuse pursuers go in several directions at once. probably West think they invented it like smoke and mirrors and dangling the carrot.but truthfully ..I think the Racoons were first
. What is the authorities are gaining here?..protection of reputations.and Time.....withhold information on cargo..everyone chase whats in the box..throw in as you say instant experts..that start as critics ..then sing alike like childrens choir....
News play along even call call pilot fanatic..
but Pilot is only one with deep history. a decent man..more secular than friends I have.
The players are all connected closely.
and by virtue of being tied together as client and customer. then
if one fall they all fall . That is the Global Truth.

http://www.youtube.com/watch?v=5JpbZZKqxy0

in less than 24 hours changed by unknown but caught

http://www.youtube.com/watch?v=hNZtz-HVy6c

1. This comment has been removed by the author.

103. This comment has been removed by the author.

104. This comment has been removed by the author.

105. @Emil :Brilliant..

Reddit was interesting.."Comms between Satellite and A/C are determined by highly technical procedures. So its not easy for the public to understand the issues involved.

I found the following information about emergency comms that should have taken place on march 8th. It is stated, that Boeing implemented a system in its 777, that will transmit all necessary and available emergency data of the A/C with the handshakes to the Provider. These data should be somewhere in a Data Dump at INMARSAT.

found this in

quote
The Inmarsat Data collection while praised is fatally flawed to the extent that Inmarsat “screwed the pooch” on its translation, however the data has long been refined over and again into its proper form by independent individual(s).

Furthermore, the companies who reviewed the Inmarsat Data did not include the one single company who could have properly placed the data into its required and needed format as the data is more than signal propagation, it actually provided ACARS from the Primary FMS ELT and Emergency Transmission and much more.

Inmarsat has a duty by design to receive and notify emergency transmissions immediately as if it was a 911 call and dispatch first responders in an air emergency as Boeing designed the 777 to do that very action, however Inmarsat failed to build the system for reception and dispatch alerts in emergency circumstances. Only now are they offering free services, however many years later and several air accidents overdue.

Moreover, largely overlooked and misrepresented is the notion of 7 Inmarsat Handshakes. In actuality there is 8 handshakes on the initial Inmarsat report. Hereto, the Inmarsat report demonstrates that #MH370 continuously from the vanish point from radar via SatCom immediately began transmitting at that time until impact at 8:19MYT with repeated Emergency Alerts and Data transfer attempts of an estimated 100+ or more times in flight.

..... it’s more than possible the raw data resides in more than the Inmarsat system or other systems unknown in the network. Additionally, the Inmarsat satellite used to analyze the data is a piece meal analysis. What Inmarsat believes they do not have lies within the system as an unknown to their knowledge as a cookie or temp file as it would appear in a home computer. Also on this point, most satellite operators have a mass amount of raw data that is determined as junk and removed in a regular maintenance period.

More importantly, the fallacy of the Inmarsat 3f-1 satellite standard frequency use, as Inmarsat has stated “Signal Handshakes are routinely completed if the receiving or transmitting device is powered on and a connection ping is completed on an hourly basis on average to establish and refresh the signal propagation to that device for use if needed” is simply NOT all correct.

In order for Inmarsat 3f-1 to achieve the handshake operation, the signal array from the satellite would have to be in bounce mode, and Inmarsat 3f-1 Standard setting is Burst Mode or Microburst Mode. In Burst Mode, satellites send continuous signal burst outward and receive a connection when a device is transmitting outward to the satellite receiver grabbing a burst cell.

Therefore, #MH370 would have to be sending a transmission outward collecting a burst signal for the Handshake to bounce and return the signal; otherwise burst mode sends out signal cells continuously until a transmission is attempted or connection completed by a device itself.

MH370 continuously attempted these transmissions in Emergency Mode; however the calls were dismissed and uninterrupted leaving only the data stream un-decoded in the system with full Data being dumped into the cloud as the Emergency Network never got around to being designed by Inmarsat to address such a circumstance.

The International Community should seize all information Inmarsat may have with or without knowledge to address this issue on the Iridium System as it was designed 20 years ago to interpret these safety emergencies and issues involving commercial aircraft.’

1. This comment has been removed by the author.

106. This comment has been removed by the author.

How 4,000 Physicists Gave a Vegas Casino its Worst Week Ever

What happens when several thousand distinguished physicists, researchers, and students descend on the nation’s gambling capital for a conference? The answer is "a bad week for the casino"—but you'd never guess why.

Ask a Physicist: Phone Flash Sharpie Shock!

Lexie and Xavier, from Orlando, FL want to know:
"What's going on in this video? Our science teacher claims that the pain comes from a small electrical shock, but we believe that this is due to the absorption of light. Please help us resolve this dispute!"

The Science of Ice Cream: Part One

Even though it's been a warm couple of months already, it's officially summer. A delicious, science-filled way to beat the heat? Making homemade ice cream.

(We've since updated this article to include the science behind vegan ice cream. To learn more about ice cream science, check out The Science of Ice Cream, Redux)

Over at Physics@Home there's an easy recipe for homemade ice cream. But what kind of milk should you use to make ice cream? And do you really need to chill the ice cream base before making it? Why do ice cream recipes always call for salt on ice?