Malaysia Airlines Flight 370

=== Analysis of satellite communication ===
{{main|Malaysia Airlines Flight 370 satellite communications}}
The communications between Flight 370 and the satellite communication network operated by Inmarsat, which were relayed by the Inmarsat-3&nbsp;F1 satellite, provide the only significant clues to the location of Flight 370 after disappearing from Malaysian military radar at 02:22 MYT. These communications have also been used to infer possible in-flight events. The investigative team was challenged with reconstructing the flight path of Flight 370 from a limited set of transmissions with no explicit information about the aircraft's location, heading, or speed.<ref name=ATSB />{{Rp|16–17}}<ref name="Telegraph-delays"/>

==== Technical background ====
[[File:Inmarsat-3 crop.jpg|thumb|A depiction of an [[Inmarsat#Satellites|Inmarsat-3 series]] satellite. Flight 370 was in contact with Inmarsat-3 F1 (also known as "IOR" for Indian Ocean Region).|alt=A depiction of a satellite in space.]]
Aeronautical satellite communication (SATCOM) systems are used to transmit messages sent from the aircraft cockpit, as well as automated data signals from onboard equipment, using the [[ACARS]] communications protocol. SATCOM may also be used for the transmission of [[Future Air Navigation System|FANS]] and [[Aeronautical Telecommunication Network|ATN]] messages, and for providing voice, fax and data links<ref name="classicaero">{{cite web|title=Classic Aero services and SwiftBroadband|url=http://www.inmarsat.com/service/aviation-safety/|work=[[Inmarsat]]|access-date=28 March 2014|archive-date=28 March 2014|archive-url=https://web.archive.org/web/20140328221911/http://www.inmarsat.com/service/aviation-safety/|url-status=live}}</ref> using other protocols.<ref name="Telegraph-delays"/><ref name=RGN>{{cite web|last=Kirby|first=Mary|title=SITA aids MH370 investigation; expert explains|url=http://www.runwaygirlnetwork.com/2014/03/14/sita-aids-mh370-investigation-expert-explains/|website=Runway Girl Network|date=14 March 2014|access-date=26 March 2014|archive-date=26 March 2014|archive-url=https://web.archive.org/web/20140326105450/http://www.runwaygirlnetwork.com/2014/03/14/sita-aids-mh370-investigation-expert-explains/|url-status=live}}</ref><ref name=ATM>{{cite web|last=Turner|first=Aimee|title=Malaysian MH370: SATCOMS 101 (Part One)|url=http://www.airtrafficmanagement.net/2014/03/mh370-satcoms-101/|website=AirTrafficManagement.net|date=16 March 2014|access-date=26 March 2014|archive-url=https://web.archive.org/web/20140322143315/http://www.airtrafficmanagement.net/2014/03/mh370-satcoms-101/|archive-date=22 March 2014|url-status=dead}}</ref> The aircraft uses a [[satellite data unit]] (SDU) to send and receive signals over the satellite communications network; this operates independently from the other onboard systems that communicate via SATCOM, mostly using the ACARS protocol. Signals from the SDU are transmitted to a communications satellite, which [[RF power amplifier|amplifies]] the signal and changes its [[Radio frequency|frequency]] before relaying it to a [[ground station]], where the signal is processed and, if applicable, routed to its intended destination (e.g. Malaysia Airlines' operations centre); signals are sent from the ground to the aircraft in reverse order.

When the SDU is first powered on, it attempts to connect with the Inmarsat network by transmitting a log-on request, which is acknowledged by the ground station.<ref name=ATSB />{{rp|17}}<ref name="ATM"/> This is partly to determine whether the SDU belongs to an active service subscriber, and also to identify which satellite should be used for transmitting messages to the SDU.<ref name="ATM"/> After connecting, if no further contact has been received from the data terminal (the SDU) for one hour,{{efn|The timing of the log-on interrogation message is determined by an inactivity timer, which was set to one hour at the time of Flight 370's disappearance (it was later reduced to 15&nbsp;minutes).<ref name=ATSB />{{rp|18}}}} the ground station transmits a "log-on interrogation" message, commonly referred to as a "ping";<ref name=ATSB />{{rp|18}} if the terminal is active, it will respond to the ping automatically. The entire process of interrogating the terminal is referred to as a "[[Handshake (computing)|handshake]]".<ref name="Inmarsat(26 March)" /><ref name="myDCA_20140325">{{cite web|title=Information provided to MH370 investigation by UK Air Accidents Investigation Branch (AAIB) |url=http://mh370.dca.gov.my/information-provided-to-mh370-by-aaib/ |date=25 March 2014 |author=Prime Minister's Department (Malaysia) |author-link=Prime Minister's Department (Malaysia) |publisher=[[Department of Civil Aviation Malaysia]] |access-date=6 May 2014 |url-status=dead |archive-url=https://web.archive.org/web/20140406010301/http://mh370.dca.gov.my/information-provided-to-mh370-by-aaib/ |archive-date=6 April 2014}}</ref>

==== Communications from 02:25 to 08:19 MYT ====
Although the ACARS data link on Flight 370 stopped functioning between 01:07 and 02:03 MYT (most likely around the same time the plane lost contact by secondary radar),<ref name="Ground log" />{{Rp|36}} the SDU remained operative.<ref name=ATSB /> After last contact by primary radar west of Malaysia, the following events were recorded in the log of Inmarsat's ground station at [[Perth]], Western Australia (all times are [[Time in Malaysia|MYT]]/UTC+8):<ref name=ATSB />{{rp|18}}<ref name="Ground log"/>{{efn|Information released and reported publicly about SATCOM transmissions from Flight 370 have been inconsistent, especially the use of the terms "ping" and "handshake". It was initially reported as six "handshakes" or "pings" with one "partial handshake or ping" sent at 00:19&nbsp;UTC by Flight 370, unprovoked by the ground station. The events listed may consist of several "transmissions" between the aircraft and ground station over the course of a few seconds. A readable copy of the ground station log of transmissions to and from Flight 370 is available [https://web.archive.org/web/20160304185716/http://www.dca.gov.my/mainpage/MH370%20Data%20Communication%20Logs.pdf online]}}
* 02:25:27&nbsp;– First handshake ("log-on request" initiated by aircraft)
* 02:39:52&nbsp;– Ground to aircraft telephone call, acknowledged by SDU, unanswered
* 03:41:00&nbsp;– Second handshake (initiated by ground station)
* 04:41:02&nbsp;– Third handshake (initiated by ground station)
* 05:41:24&nbsp;– Fourth handshake (initiated by ground station)
* 06:41:19&nbsp;– Fifth handshake (initiated by ground station)
* 07:13:58&nbsp;– Ground to aircraft telephone call, acknowledged by SDU, unanswered
* 08:10:58&nbsp;– Sixth handshake (initiated by ground station)
* 08:19:29&nbsp;– Seventh handshake (initiated by aircraft); widely reported as a "partial handshake'", consisting of the following two transmissions:
:* 08:19:29.416&nbsp;– "log-on request" message transmitted by aircraft (seventh "partial" handshake)
:* 08:19:37.443&nbsp;– "log-on acknowledge" message transmitted by aircraft (last transmission received from Flight 370)
The aircraft did not respond to a ping at 09:15.<ref name="Ground log" />

==== Inferences ====
A few inferences can be made from the satellite communications. The first is that the aircraft remained operational until at least 08:19 MYT—seven hours after final contact was made with air traffic control over the South China Sea. The varying burst frequency offset (BFO) values indicate the aircraft was moving at speed. The aircraft's SDU needs location and track information to keep its antenna pointed towards the satellite, so it can also be inferred that the aircraft's navigation system was operational.<ref name=JN-Inmarsat />{{rp|4}}

Since the aircraft did not respond to a ping at 09:15, it can be concluded that at some point between 08:19 and 09:15, the aircraft lost the ability to communicate with the ground station.<ref name="Inmarsat(26 March)"/><ref name=Fox-Inmarsat /><ref name="myDCA_20140325" /> The log-on message sent from the aircraft at 08:19:29 was "log-on request"; there are only a few reasons the SDU would transmit this request, such as a power interruption, software failure, loss of critical systems providing input to the SDU, or a loss of the link due to the aircraft's [[:wiktionary:aircraft attitude|attitude]]<!-- ATTitude NOT ALTitude. This has been changed many times. Because there is no concise Wikipedia article/section to point to, a link to the definition on Wiktionary is appropriate -->.<ref name=ATSB />{{rp|22}} Investigators consider the most likely reason to be that it was sent during power-up after an electrical outage.

At 08:19, the aircraft had been airborne for 7&nbsp;hours and 38&nbsp;minutes; the typical Kuala Lumpur-Beijing flight is 5{{frac|1|2}}&nbsp;hours, so [[fuel starvation|fuel exhaustion]] was likely.<ref name=ATSB />{{Rp|33}}<ref>{{cite web|title=Considerations on defining the search area&nbsp;– MH370|url=http://www.atsb.gov.au/publications/2014/considerations-on-defining-the-search-area-mh370.aspx|publisher=ATSB&nbsp;– Australian Transport Safety Bureau|access-date=28 May 2014|archive-date=26 May 2014|archive-url=https://web.archive.org/web/20140526210049/http://www.atsb.gov.au/publications/2014/considerations-on-defining-the-search-area-mh370.aspx|url-status=dead}}</ref> In the event of fuel exhaustion and engine flame-out, which would eliminate power to the SDU, the aircraft's [[ram air turbine]] (RAT) would deploy, providing power to some instruments and flight controls, including the SDU.<ref name=ATSB />{{Rp|33}} Approximately 90&nbsp;seconds after the 02:25 handshake—also a log-on request—communications from the aircraft's [[in-flight entertainment]] system were recorded in the ground station log. Similar messages would be expected following the 08:19 handshake, but none were received, supporting the fuel-exhaustion scenario.<ref name=ATSB />{{rp|22}}

==== Analysis ====
[[File:MH370 location probability heat map per DST Group analysis.jpg|thumb|A [[heat map]] indicating the probable location of missing Flight 370 based on a [[Bayesian search theory|Bayesian analysis]] of possible flight paths by Australia's [[Defence Science and Technology Group]]<!-- Note: Original file name included "HeatMap", which is why it's being added here. Original file URL: http://jacc.gov.au/media/maps/files/20151201_IndicativeArea_Bathy_DSTO_HeatMap_A4.jpg --><ref>{{Cite report |date=3 December 2015 |title=MH370 – Definition of Underwater Search Areas |url=http://www.atsb.gov.au/media/5733650/AE-2014-054_MH370-Definition%20of%20Underwater%20Search%20Areas_3Dec2015.pdf |publisher=[[Australian Transport Safety Bureau]] |access-date=3 December 2015 |archive-date=12 April 2021 |archive-url=https://web.archive.org/web/20210412232900/http://www.atsb.gov.au/media/5733650/AE-2014-054_MH370-Definition%20of%20Underwater%20Search%20Areas_3Dec2015.pdf |url-status=live }}</ref>]]
Two parameters associated with these transmissions that were recorded in a log at the ground station were key to the investigation:
* Burst time offset (BTO) – the time difference between when a signal is sent from the ground station and when the response is received. This measure is proportional to twice the distance from the ground station via the satellite to the aircraft and includes the time that the SDU takes between receiving and responding to the message and time between reception and processing at the ground station. This measure was analysed to determine the distance between the satellite and the aircraft at the time each of the seven handshakes occurred, and thereby defining seven circles on the Earth's surface<!---(conceptually it should be up in the air at flight level)--> the points on whose circumference are equidistant from the satellite at the calculated distance. Those circles were then reduced to arcs by eliminating those parts of each circle that lay outside the aircraft's range.<ref name=ATSB />{{Rp|18}}<ref name=JN-Inmarsat />{{rp|4–6}}
* Burst frequency offset (BFO) – the difference between the expected and received frequency of transmissions. The difference is caused by [[Doppler effect|Doppler shifts]] as the signals travelled from the aircraft to the satellite to the ground station; the frequency translations made in the satellite and at the ground station; a small, constant error (bias) in the SDU that results from drift and ageing; and compensation applied by the SDU to counter the Doppler shift on the uplink. This measure was analysed to determine the aircraft's speed and heading, but multiple combinations of speed and heading can be valid solutions.<ref name=ATSB />{{Rp|18}}<ref name=JN-Inmarsat />{{rp|9–11}}
By combining the distance between the aircraft and satellite, speed, and heading with aircraft performance constraints (e.g. fuel consumption, possible speeds and altitudes), investigators generated candidate paths that were analysed separately by two methods. The first assumed the aircraft was flying on one of the three [[autopilot]] modes (two are further affected by whether the navigation system used magnetic north or true north as a reference), calculated the BTO and BFO values along these routes, and compared them with the values recorded from Flight 370. The second method generated paths which had the aircraft's speed and heading adjusted at the time of each handshake to minimise the difference between the calculated BFO of the path and the values recorded from Flight 370.<ref name=ATSB />{{Rp|18, 25–28}}<ref name="ATSB Flight path update"/>{{Rp|10–11}} A [[probability distribution]] for each method at the BTO arc of the sixth handshake of the two methods was created and then compared; 80% of the highest probability paths for both analyses combined intersect the BTO arc of the sixth handshake between 32.5°S and 38.1°S, which can be extrapolated to 33.5°S and 38.3°S along the BTO arc of the seventh handshake.<ref name="ATSB Flight path update" />{{Rp|12}}