WEBVTT 00:00:01.229 --> 00:00:05.410 For years, investigators, engineers, mathematicians, 00:00:05.730 --> 00:00:08.630 and independent analysts worked to understand 00:00:08.630 --> 00:00:11.349 what these satellite signals were trying to say. 00:00:12.650 --> 00:00:15.070 This graph represents one of the most important 00:00:15.070 --> 00:00:19.769 breakthroughs in the MH370 investigation, the 00:00:19.769 --> 00:00:22.469 Doppler behavior derived from Inmarsat burst 00:00:22.469 --> 00:00:26.250 frequency offsets. What made this graph so important 00:00:26.250 --> 00:00:28.750 was that it proved the aircraft was not simply 00:00:28.750 --> 00:00:32.409 stationary. It was moving relative to the satellite. 00:00:32.990 --> 00:00:35.950 And more importantly, the motion itself could 00:00:35.950 --> 00:00:39.450 be measured. Michael Exner and others helped 00:00:39.450 --> 00:00:42.030 demonstrate that the Doppler shift contained 00:00:42.030 --> 00:00:45.210 directional information hidden inside the signal 00:00:45.210 --> 00:00:50.390 structure. That changed everything. But even 00:00:50.390 --> 00:00:53.009 with this breakthrough, one major limitation 00:00:53.009 --> 00:00:56.689 remained. The signals were still primarily being 00:00:56.689 --> 00:01:00.100 treated as isolated measurements. points on a 00:01:00.100 --> 00:01:04.879 graph, corrections, frequency offsets. What you 00:01:04.879 --> 00:01:08.000 are about to see is the next stage beyond that. 00:01:08.420 --> 00:01:12.859 Instead of treating BTO and BFO separately, we 00:01:12.859 --> 00:01:15.500 begin treating them as a unified motion system, 00:01:16.400 --> 00:01:19.700 a live constraint environment, where timing, 00:01:20.319 --> 00:01:24.359 distance, direction, and continuity must all 00:01:24.359 --> 00:01:28.079 remain satisfied simultaneously. This is where 00:01:28.079 --> 00:01:31.239 the reconstruction stops behaving like disconnected 00:01:31.239 --> 00:01:36.799 points and starts behaving like motion. We start 00:01:36.799 --> 00:01:40.379 with BTO signals, not assumptions, not search 00:01:40.379 --> 00:01:44.760 zones, just signals. The first signal is the 00:01:44.760 --> 00:01:48.340 measured BTO error field. This graph represents 00:01:48.340 --> 00:01:50.719 the difference between the measured burst timing 00:01:50.719 --> 00:01:54.099 offset and the calculated satellite path timing 00:01:54.099 --> 00:01:57.079 between the aircraft and the ground Earth station. 00:01:58.040 --> 00:02:02.579 At first glance, it appears chaotic. Noise, random 00:02:02.579 --> 00:02:06.760 variation. But inside that apparent chaos is 00:02:06.760 --> 00:02:11.379 something important. Continuity. The signal never 00:02:11.379 --> 00:02:15.699 completely loses structure. Even inside the airfield, 00:02:15.900 --> 00:02:18.979 The aircraft remains constrained to a measurable 00:02:18.979 --> 00:02:21.780 timing relationship with the satellite system. 00:02:22.840 --> 00:02:25.300 The second signal is the distribution profile. 00:02:25.960 --> 00:02:29.000 The histogram shows us how the BTO error behaves 00:02:29.000 --> 00:02:32.439 statistically over time. What matters here is 00:02:32.439 --> 00:02:35.639 not perfection. What matters is containment. 00:02:36.599 --> 00:02:39.460 The signal errors remain centered around a measurable 00:02:39.460 --> 00:02:42.960 mean. That tells us the system itself remains 00:02:42.960 --> 00:02:46.379 stable enough to reconstruct motion from. The 00:02:46.379 --> 00:02:49.960 satellite system is not randomly drifting. It 00:02:49.960 --> 00:02:53.319 is operating inside a constrained timing environment. 00:02:56.180 --> 00:02:59.819 The third signal is the detailed breakdown. This 00:02:59.819 --> 00:03:02.560 is where we begin seeing the microstructure of 00:03:02.560 --> 00:03:06.979 the timing behavior itself. Short bursts, small 00:03:06.979 --> 00:03:10.699 oscillations, continuous corrections. The signal 00:03:10.699 --> 00:03:13.439 is constantly adjusting while still remaining 00:03:13.439 --> 00:03:17.129 inside a stable envelope. This becomes extremely 00:03:17.129 --> 00:03:20.509 important later, because once motion begins to 00:03:20.509 --> 00:03:23.430 organize, these same small timing structures 00:03:23.430 --> 00:03:28.030 begin revealing directional continuity. This 00:03:28.030 --> 00:03:31.009 is the foundation of the reconstruction before 00:03:31.009 --> 00:03:34.349 headings, before turns, before flight paths. 00:03:35.129 --> 00:03:38.409 We begin with the raw timing environment itself, 00:03:39.009 --> 00:03:42.150 and once these signals are stabilized and constrained, 00:03:42.349 --> 00:03:45.569 we can begin combining them with Doppler behavior, 00:03:45.909 --> 00:03:49.150 motion continuity, and fixed anchor geometry. 00:03:50.050 --> 00:03:52.110 This is where the system begins transitioning 00:03:52.110 --> 00:03:55.590 from isolated measurements into live motion. 00:03:57.250 --> 00:04:00.229 Then we break the system down even further. Now 00:04:00.229 --> 00:04:03.750 we isolate only the BTO and BFO noise environment 00:04:03.750 --> 00:04:07.469 itself. No flight path yet, no assumptions about 00:04:07.469 --> 00:04:11.009 destination, no reconstructed trajectory, just 00:04:11.009 --> 00:04:16.290 signal behavior. we begin reducing the BTO noise 00:04:16.290 --> 00:04:20.750 field. The raw burst timing offset contains transmission 00:04:20.750 --> 00:04:24.230 variability, timing jitter, satellite timing 00:04:24.230 --> 00:04:28.110 uncertainty, and environmental noise. At first 00:04:28.110 --> 00:04:31.310 glance, the signal appears unstable, but once 00:04:31.310 --> 00:04:34.329 the non -essential variation is reduced, the 00:04:34.329 --> 00:04:37.629 underlying timing continuity begins to emerge. 00:04:38.410 --> 00:04:40.990 The important part is not removing the signal. 00:04:41.290 --> 00:04:44.670 the important part is revealing the stable structure 00:04:44.670 --> 00:04:48.290 underneath it. This is where the Doppler environment 00:04:48.290 --> 00:04:53.269 becomes visible. RAW BFO contains oscillator 00:04:53.269 --> 00:04:56.949 drift, satellite motion effects, aircraft motion 00:04:56.949 --> 00:05:00.769 effects, and system correction behavior. Once 00:05:00.769 --> 00:05:04.209 reduced, the directional continuity begins separating 00:05:04.209 --> 00:05:06.149 itself from the surrounding noise.