A WSPRnet Propagation Technical Analysis can be downloaded here
The purpose of this technical analysis is to explain how the propagation of WSPRnet signals is used to track an aircraft. A worked example taken from the Qantas flight used in a recent blind test will be used to explain in detail how the system detects and tracks the aircraft.
WSPRnet data allows the detection and tracking of aircraft on a global basis. WSPRnet data is available historically going back to 2008. The accuracy of the tracking is generally within 18 nm. An aircraft can be tracked at low altitude as long as it is not shielded by terrain.
WSPRnet transmissions propagate further than previously assumed. The WSPR protocol allows detections at lower signal levels than previously assumed.
Further tests are planned using data from AMSA MH370 search and rescue flights into the southern Indian Ocean in March/April 2014.
If it can be shown that aircraft can be successfully tracked in 2014 across the Indian Ocean on average at least every 10 minutes using WSPRnet data, then coupled with the Inmarsat satellite data every hour, a much more precise estimate of where MH370 crashed can be made.
@All
The WSPRnet Propagation Technical Analysis is based on the 2007 International Reference Ionosphere (IRI). The IRI is a permanent joint scientific project of the Committee on Space Research (COSPAR) and the International Union of Radio Science (URSI) started in 1968. It is the international standard empirical model for the terrestrial ionosphere since 1999.
I use Proplab Pro V3.1 for tracing radio waves around the globe. Proplab Pro was one of the first ionospheric physics-based HF radio signal ray-tracing engines in the world for the PC and has been maintained and developed for over 30 years. It continues to be used by research organisations, scientists, engineers, students, the military and amateur radio operators around the world. It is one of the only software packages to reliably predict ionospherically refracted HF radio signal behaviour. It uses both three dimensional models of the ionosphere as well as three dimensional topographical data of the Earth to provide unprecedented detail in modelling HF radio signal propagation worldwide.
GDTAAA is a software application that I have developed which uses a Vincenty WGS84 navigation engine to track aircraft, a spherical navigation engine to track WSPRnet data links and anomalies. In addition I use the Proplab Pro V3.1 engine with its embedded IRI 2007 ionospheric model to perform global ionospheric mapping and ray tracing.
This toolset enables me to detect and track aircraft anywhere in the globe and at any time currently or historically going back as far as 2009.
This system is undergoing a number of tests. Currently, I am tracking a search and rescue flight organised by the Australian Maritime Safety Authority (AMSA) and conducted by the Royal New Zealand Air Force (RNZAF) in the vicinity of the 7th Arc in the southern Indian Ocean during March 2014. AMSA have kindly released to me under a confidentiality agreement all the operational planning data from their search and rescue flights. The RNZAF have kindly released all the flight data for a flight on 28th March 2014 from Pearce Air Force Base near Perth to the search area near the 7th Arc between 32°S and 33°S. I am grateful for the help of Brian Anderson who obtained the RNZAF flight data under a Freedom Of Information request.
Once these tests have been successfully completed, I will use this system to track MH370 from departure via the point of diversion to beyond the limited range of the civilian and military radar systems and into the Indian Ocean. This data coupled with the Inmarsat satellite data will give a more accurate picture of the flight path followed by MH370. This information will be passed on to Ocean Infinity, who would like to resume their underwater search for MH370. Perhaps it will be possible to conduct a further underwater search towards the end of 2022 during the southern hemisphere Summer months.
Re -“Perhaps it will be possible to conduct a further underwater search towards the end of 2022 during the southern hemisphere Summer months.”
Excellent news, although seven years is already a very long time for the relatives and friends to have had to wait, so an earlier detailed hydroacoustic analysis of a 50nm radius designated search area might help to direct Ocean Infinity with more precision next year.
If further confirmation of the GDTAAA MH370 analysis is required, it may be that in amongst the vast amount of data exchanged between Australia and South Africa as part of the global Square Kilometre Array project there are recognisable patterns of signal communication anomalies that can be mapped against the Southern Indian Ocean component of the final flight of MH370.
“The cost of the SKA including construction and the first 10 years of operations (2021-2030) is estimated to be around 1.9 billion euros in 2020 euros.”
https://www.skatelescope.org/the-ska-project/
Rather than taking on the burden of processing all that data unilaterally, it might be worth Australia and South Africa inviting the UN Security Council, directly or indirectly, to help facilitate some collaborative international support, especially at a time when the UN Security Council is attempting to up the ante on global disarmament and non-proliferation.
Mary Robinson recently addressed the Security Council on 7th September 2021:
“…she went on to welcome … today’s many references to climate change and its relationship to security. Council members should think hard about that nexus before passing any specific resolution on the matter, “and make it a strong one”, she said. In addition, she welcomed the many references to the NPT and the importance of the broader nuclear issue. She also welcomed plans by Ireland’s Presidency to raise the issue at a signature Council event later in September.”
https://www.un.org/press/en/2021/sc14623.doc.htm
For example, the Swiss researchers who recently made headlines by calculating pi to approximately 62.8 trillion decimal places might be looking for something more useful to do with their computers.
https://www.theguardian.com/science/2021/aug/17/new-mathematical-record-whats-the-point-of-calculating-pi
Using analogous underwater technologies to the Square Kilometre Array, WSPRnet and GDTAAA, it might be possible to home in on the acoustic traces of any bubbles or mini-explosions generated by the consignment of, presumably, after seven years, very decayed lithium-ion batteries in one or more of the cargo containers on MH370.
https://www.theguardian.com/environment/2021/aug/31/robo-penguin-self-propelled-devices-to-help-study-of-eddies
By making the source data available it might be possible for the analysis to be cross-verified by supercomputers in, for example, Japan, the US, Singapore, Russia, India, Israel, Ireland and China.
GCHQ might even choose to organise an international schools and universities project to help with the number-crunching as a way of popularising its networked Raspberry Pi clusters (octopi).
https://www.gchq.gov.uk/information/octapi-goes-school
And if there is a whiff of a commercial gain from such a venture, it might even be that some of the bigger global, privately owned, data mining and processing companies would be able to make a useful contribution to the search.
Thank you once again for your tireless efforts, Richard. To quote Robb Johnson, the English folk singer:
“Singing, singing for the souls that can’t
Though they shut songs up in stadiums
Someone always carries on
Whoah, singing for ther Moon”
@ Richard is is at all possible that parts of the plane would wash up around the British Indian Ocean? Did the plan go down anywhere where the currents could have carried the plane or parts of the plane in that direction?
@Brandon,
Welcome to the blog!
May I refer you to my report on the MH370 Floating Debris Drift Analysis previously published on this website:
https://www.dropbox.com/s/kwp4qm0e08lg487/MH370%20Floating%20Debris%20Drift%20Analysis%20-%2012th%20January%202021.pdf?dl=0
You ask whether it is possible that MH370 debris would wash up around the British Indian Ocean Territory (BIOT)?
The BIOT covers an area around 54,000 km2 out of the 70.6M km2 of the Indian Ocean. There are over 1,000 islands and a total land area of around 60 km2. The islands stretch from 5.2°S to 7.4 °S between 71.2°E and 72.5°E.
This is further north than the South Equatorial Current (SEC) region, which extends as far north as 10°S. The SEC is the main transport system of MH370 floating debris from East to West across the Indian Ocean.
From Figure 6 in my report, you can see it is possible to reach the BIOT from points along the 7th Arc, but unlikely from the crash location I have identified at 33°S near the 7th Arc. Figure 7 shows that it is one of the least likely area to find debris washing up on an island in the Indian Ocean, as most floating debris would pass either to the south going westward and later pass to north going eastward.
How do you handle the following:
(A) IRI is a monthly median model – the real ionosphere on any given day will exhibit substantial variations on timescales of minutes and spatial scales of 10’s to 1000’s of kilometres
(B) The ionosphere can induce Doppler shifts which are also time varying.
(C) Proplab Pro + IRI may provide a monthly median model of propagation conditions but this will be substantially different from that which exists on any given day.
@All,
An article by Geoffrey Thomas explaining how this new technology works has been published at AirlineRatings:
https://bit.ly/3lLbcRd
Geoffrey writes:
“The possibility of a new search for MH370 has been dramatically enhanced by the latest results from trials of a new breakthrough tracking technology called WSPRnet.”
“There is no doubt that Mr Godfrey’s work offers real hope of a significant breakthrough in the search for MH370 with a far more accurate picture of the flight route and last minutes of the flight which was lost on March 8, 2014, with 239 souls aboard.”
If the Inmarsat data is flawed or falsified, then your “new” findings will be just as flawed or false, correct? The authorities need to release the original radar information. Why make it “secret” if there is nothing to hide?
Per Tomnod data, there was tons of debris near the flight path and other locations. Just ask the Cathay Crew that reported the next morning, “a large metal debris field off Vietnam.” Which was ignored and unexplained. If there was just one aircraft missing, then why did Tomnod serve up scans of specific areas around that region? Why did so many witnesses report an aircraft sighting from several different locations around that region? Was it because there was more than one aircraft(military) involved?
Keeping the true information “hidden” on this tragedy is a crime against the 239 souls and their family members in my opinion.
@Cyndi,
Welcome to the blog!
I do not believe the Inmarsat data is flawed or falsified. I have performed hundreds of calculations using satellite data provided by independent sources and the Inmarsat data and their calculations are correct.
The WSPRnet data is provided by radio amateurs around the world and is completely independent of the satellite data from Inmarsat.
Unfortunately the world’s oceans are full of debris. A small number of debris items have been found in the Indian Ocean that have been proven to have originated from MH370.
There are many aircraft travelling the world and it is not surprising that there are many sightings.
I agree with you that Malaysia should release the military radar information.
I still strongly believe since MH370 went missing in March 2014 that there is information being hidden from authorities who have the power to do so. Until I’m proven otherwise, I have heavily scrutinised roughly 15-20 theories and narrowed it down to only 3 possible outcomes. My theory is that MH370 DID NOT crash in the Indian Ocean like authorities and the mainstream media want us to believe. It is also really disrespectful to blame the pilot who can no longer defend himself. My theory is MH370 was shot down accidentally during a military exercise over the South China Sea that night. We are led to believe it disappeared from radar over the South China Sea and veered off left over the straights of Malacca with no scheduled flight plan. I find it highly unlikely that a plane with no flight path can just simply fly close to the US Military base Diego Garcia with no information given from themselves to air traffic control, this silence is very suspicious.
@Matthew
Welcome to the blog!
I agree with you that there is information that the authorities are withholding from the public domain.
I disagree with your theory that MH370 did not crash in the Indian Ocean because:
(1) The Inmarsat satellite data proves that MH370 headed south towards the Indian Ocean and continued for 7 hours 37 minutes in that direction.
(2) 33 debris items have washed up on the shores of the Indian Ocean and some are proven to have come from MH370.
I do not believe MH370 was shot down over the South China Sea because in addition to the two points above:
(1) The civilian radar data shows MH370 tracking back over Malaysia to the Malacca Strait.
(2) The co-pilot’s mobile telephone was detected near Penang.
MH370 did not fly anywhere near Diego Garcia.
Thanks for the detailed reply Richard and keep up the good work.
There has been lots of questionable debris brought to the surface but very little has been confirmed.
If we look back at Air France 447 back in 2009 within days aircraft spotted hundreds if not thousands of debris in the Atlantic Ocean. The Boeing 777 has a greater wingspan, height and length to the A330 that crashed in the Atlantic.
As someone who knows the Boeing 777 extremely well, the only way this large jet would be difficult to find would be if it shattered on the surface creating millions of smaller pieces or it landed in tact both which I doubt in my opinion.
I appreciate the Indian Ocean is more vast and the search teams were originally looking in the South China Sea but I find it very strange no sightings of debris has been seen over the Indian Ocean considering the amount of flights that pass over each day.
The flaperon that was found on Reunion island I find suspicious as why would it just be that piece? Where is the tail fin and bigger parts of the jet?
I really do hope you prove me wrong but I still remain highly confident that sadly we will never find the truth to what happened to MH370 and if we were lucky enough to locate the black box it might no longer provide the key information we need?
@Matthew,
We are going over old ground here.
There are 11 debris items confirmed from MH370 and an additional 10 items likely from MH370.
In the case of AF447 the searchers knew exactly where to look and there was no delay in looking. With regard to MH370 the searchers did not know exactly where to look and there was a delay of 10 days before any search started in the Indian Ocean.
A part of the Vertical Stabiliser was found washed ashore in Central Mozambique at Linga, Linga on 26AUG2016:
https://www.dropbox.com/s/kcf1qm4400tlnlj/MH370%20Part%20of%20Vertical%20Stabiliser%20Mozambique%2026%20AUG2016.png?dl=0
Thanks Richard.
I look forward to hearing how you get on and wish you the very best of luck.
Lets see if your theory is right and you prove me wrong.
Kind Regards.
@Matthew,
Many thanks for your kind words!
You may be interested in my MH370 Floating Debris Drift Analysis published here on 1st January 2021:
https://www.dropbox.com/s/bzpwdjdiyz7yiry/MH370%20Floating%20Debris%20Drift%20Analysis%20-%201st%20January%202021.pdf?dl=0
In this paper I estimate there were 35,740 debris items from MH370 in total based on the weight of the 33 items found, which is neither MH370 landing intact in one piece, nor MH370 shattering into millions of small pieces.
Hi
Do you think that if MH370 has not been “shattering into millions of small pieces” could be because she was flown after passing 7th arc, loosing slowly altitude and a final crash at not very high speed?
If MH370 crashed from cruse altitude without any pilot action, she would did that at very high speed, perhaps enough for some parts fluttered and detached before sea contact and then shattering.
Just an opinion!
@Pax Lambda,
Thousands of pieces of debris is already indicative of a heavy crash.
The Inmarsat satellite data shows an accelerating dive at around 15,000 fpm at 00:19:37 UTC. In my view, it is unlikely that MH370 recovered from this dive to slowly lose altitude and crash at a much slower speed close to a landing speed.
The ATSB have reported the flaps were not extended and for a ditching they normally would be partly extended. Mike Exner and Don Thompson have shown that it is perfectly possible that certain control surfaces or even an entire wing may have detached before impact.
Hi Richard,
Don’t you know that finding a vertical speed with the BFO is too speculative. We don’t know the track of the aircraft at this precise time and we would be able to determine his vertical path? For me, that is being too confident on the BFO.
ATSB reported the flaps were not extended, that is true but the DGA reported it possible to explain the trailing edge part missing.
How considering the flaperon could have detached in flight due to flutter? The flaperon still have hydraulic power even with RAT.
Richard: ” In my view, it is unlikely that MH370 recovered from this dive”
IMHO, Air China 006 has to be considered. Investigation report is there: https://ntsb.gov/investigations/AccidentReports/Reports/AAR8603.pdf
From AAR8603.pdf page 11 (paper) / page 14 (pdf): …/… Between 1014:50 and 1015:23, the DFDR recorded a 10,310_foot descent to 30,132 feet. Between 1014:59 and 1015:06, as the airplane descended from 40,346 feet to 37,102, the recorded data showed a right-wing_down controlwheel deflection. …/…
So, between 1014:50 and 1015:23 (33 seconds) the 747 dove 10,310 feet: that is 18746 fpm
with a “peak” between 1014:59 and 1015:06 at 27806 fpm (40346-37102 = 3244 feet in 7 seconds)!
And they recovered.
With some parts lost as can be seen on this pic:
https://upload.wikimedia.org/wikipedia/commons/8/83/Damaged_empennage_of_China_Airlines_Flight_006-N4522V.JPG
I do think nobody can denied Z could (COULD!) have done the same. It would NOT imply a ditch try later. A dive from some hundreds of feet would have been a perfect end after looking for an empty sea around. I can’t believe, if Z did it (IF!), that, after what he believed a “perfectly stealthy flight” , he could have taken any chance that somebody from a boat saw the plane crashing.
PL, with my apologies for the continuation of the off topic.
@Gilles,
Welcome to the blog!
The BFO is not a subject of speculation, it is a subject of a calculation.
Irrespective of the horizontal velocity or track of the aircraft it is not possible without a significant rate of descent to obtain a BFO of -2.0 Hz at 00:19:37 UTC.
In the Archive section of this web site you will find and can download an Excel spreadsheet with detailed calculations of the BTO and BFO at each stage of the flight of MH370.
Under the category Flight Path Analysis, there is an entry titled:
“9th December 2015 – MH370 Flight Path Model Version 16 Download – Duncan Steel’s Blog View”
In row 387 you will find the calculations for BFO at 00:19:37 UTC.
Starting at column CH you will find the Satellite velocity in x, y and z.
Starting at column CS you will find the Aircraft velocity.
Starting at column DO you will find the Aircraft/Satellite velocity.
Starting at column EA you will find the Doppler Compensation, Uplink and Downlink Doppler, EAFC effect, Eclipse effect and BFO Offset.
You can set a position, altitude, ground speed, track and rate of climb (ROC) as you wish and see for yourself.
Without a ROC of -15,000 fpm you will not obtain a result anywhere close to the measured BFO of -2.0 HZ.
@Gilles,
You stated: “ATSB reported the flaps were not extended, that is true but the DGA reported it possible to explain the trailing edge part missing. How considering the flaperon could have detached in flight due to flutter? The flaperon still have hydraulic power even with RAT.”
You are not giving the full picture. Only one Flaperon is actuated and only on one side. The Flaperon will not droop and the trailing edge will not separate due to impact with water.
The RAT provides hydraulic pressure to a reduced number of Power Control Units (PCUs). Only one Flaperon PCU is active, which is the right outboard PCU. In addition the Flaperon will not droop with only the RAT supplying power.
Under RAT power to the aileron, rudder, flaperon, horizontal stabiliser, elevator and spoiler the PCUs allow only basic control of the aircraft. The slats and flaps of the high lift control system are not in operation. The Flap Slat Electronic Unit (FSEU) is bypassed and the hydraulic system is disengaged. The FSEU will no longer send any signals to the Primary Flight Computer (PFC) that result in the Flaperon drooping in alignment with the trailing edge of the adjacent flaps.
As Don Thompson from the Independent Group points out in his analysis of the RAT deployment and quoting from the Aircraft Maintenance Manual: “When the flaps extend, the PFCs use flap position data from the FSEUs to calculate aileron and flaperon droop. The PFCs add these droop commands with the signals from the aileron and flaperon schedules”. The alternate flap mode necessitated by loss of hydraulic power after dual engine flameout and APU shut down bypasses the FSEUs.
The leading and trailing edge flaps may be controlled by an ‘alternate’ open-loop mode in the event the Flap-Slat Electronic Unit (FSEU) is bypassed and the hydraulic systems disengaged. An electric motor drives a local hydraulic circulation for the extension of the flaps and slats. This would still allow an intermediate flap extension for a ditching.
Examination of damage to the Outboard Flap as well as the Flaperon from the starboard wing shows the Flap was “most likely” in the retracted position, while the Flaperon was probably in its neutral position – a configuration consistent with cruise flight, rather than approach.
The absence of the FSEU data to the PFC is conclusive evidence, that an unpowered glide will not result in a Flaperon trailing edge being torn off because the Flaperon was drooped and was being dragged in the waves following a ditching.
Mike Exner and Don Thompson have also shown in their analysis that it is perfectly possible that one or more control surfaces and even an entire wing detached from the aircraft before impact.
@All,
Steve Kent has written a paper titled “WSPR (Weak Signal Propagation Reporting) and MH370 – Theory”
https://www.dropbox.com/s/wf6z22spssby35i/Steve%20Kent%20WSPR%20Paper.pdf?dl=0
It is an excellent paper and is well worth reading. Steve provides a good overview of the theory behind WSPR.
Unfortunately the paper contains a number of flaws, errors and omissions, which in my view leads Steve to the wrong conclusion. Here are my top ten points of critique:
1. “Introduction” – “The most commonly used frequency for WSPR is 14 MHz.”
Whereas the statement is generally true, during the flight of MH370 the most commonly used frequency for WSPR is 28 MHz. Steve uses 14 MHz for a worked example and for his discussion on practicalities, Doppler rate and milliwatt WSPR.
For MH370 we are interested in WSPR signals that occurred during the flight and not what generally happens at other times.
I use all WSPRnet frequencies in the HF range between 3 MHz and 30 MHz, not just 14 MHz.
I only use WSPR signals every two minutes compared to the predicted position of the aircraft at the exact time of the WSPR signal, not an average of what happens at other times during the flight of MH370 or during the day of 7th March 2014.
2. “Theory Basics” – “A typical long distance is Rtr = 16000 km.”
The statement is true, but …
Steve omits to mention that during the flight of MH370, there were 605 WSPRnet signals with a propagation distance greater than 16,000 km.
3. “Theory Basics” – “The received S/N is well above the minimum detectable signal of -30 dB. To summarize, the minimum detectable received power is 200 dB fainter than the transmitted power of 1 W. There is a loss of about 140 dB between the transmitter and receiver due to free space propagation over 16,000 km. Thus, there is a margin of 60 dB.”
Steve omits to mention that during the flight of MH370, that all 605 WSPRnet signals with a propagation distance greater than 16,000 km, where within his 60 dB margin.
4. “Practicalities” – “The highest S/N during the flight of MH370 for the 14 MHz band for link distances exceeding 16,000 km is for spot 186158668, which achieved a S/N of -4 dB, which matches within 1 dB the calculation above.”
Steve omits to mention that during the flight of MH370, that there were 17,972 WSPRnet signals with a SNR equal to or higher than -4 dB.
5. “Practicalities” – “Overall, then, it appears that, under the most favorable circumstances, a link can be established over long distances with a margin of order 25 db, even with a transmitter of only 1 watt. Note, however, that this link was exceptional. Most links are 20 db or so weaker.”
Steve omits to mention that during the flight of MH370, that there were 40,569 WSPRnet signals with a propagation distance of over 2,000 km where the SNR at the receiver station is above the threshold of the minimum detectable signal of -30 dB.
6. “Practicalities” – “Signals at 14 MHz propagate via multiple reflections off the ionosphere and the surface of the earth. The number of such reflections can only be estimated, but clearly several hops are involved. The losses at each step can only be approximated, and reflections off the ground depend on the type of terrain and thus on the direction of propagation.”
Signals in the HF range between 3 MHz and 30 MHz propagate via multiple refractions in passing through the ionosphere. HF signals do not reflect, they refract through the ionosphere.
I use Proplab Pro V3.1 software and show the number of hops at the exact transmission frequency for a given elevation to reach the target aircraft. The number of hops is not estimated. In addition the software engine has the International Reference Ionosphere (IRI) embedded based on the 2007 ionospheric model to perform global ionospheric mapping and ray tracing. Proplab Pro V3.1 calculates the loss at each step.
Proplab Pro V3.1 maps the propagation path and shows whether the propagation takes place during the hours of darkness, the grey zone or daylight.
Proplab Pro V3.1 uses a topographical model of the Earth’s surface and can calculate tilts that might take place during reflection from the Earth’s surface.
I have shown that interim landing points on calm ocean or flat ice provide the best reflection.
7. “Aircraft Scatter” – “There is some evidence that the cross section is enhanced by up to 20 db when the aircraft is viewed from particular angles. That effect will be ignored for now. The cross-section of a 777 is estimated to be something like 40 m2 (although I have seen much higher values claimed.)“
The radar cross-section (RCS) of an aircraft was traditionally calculated for line-of-sight radar systems. The horizontal RCS of a Boeing 777 may well be 40 m2.
The angle of incidence of an ionospheric propagation of a radio wave will be more likely to reflect from the upper surface of an aircraft. The wing area alone of a Boeing 777-200ER is 427.8 m2.
I have shown that in straight and level flight there are fewer WSPRnet aircraft detections. When an aircraft is in a sharp turn, climb or descent the RCS will change and there are many more WSPRnet detections.
The WSPRnet footprint of an aircraft is much larger than the physical structure. The engine exhaust contains many different types of particle and the wake vortex containing water vaporisation can both disturb radio signals up to 20 nm behind and 4 nm either side of an aircraft’s path.
8. “Path of a WSPR signal” – “However, if one wants to try and use a WSPR signal to geolocate an aircraft, greater accuracy is needed. To a next approximation, the path will be a geodesic on an oblate spheroid.”
An aircraft flying between waypoints will follow a geodesic path around the globe and navigation systems typically use the World Geodetic System 1984 (WGS84) model of an oblate spheroid to perform path calculations. The propagation path of a radio wave will depend on the transmitting antenna’s azimuth and elevation pattern. The path can ideally be considered isotropic and will follow a spherical path not a geodesic. A skywave is not a ground wave, it does not hug the earth’s surface.
I use WGS84 for aircraft flight paths with:
Semi-major Axis: 6,378,137 m.
Semi-minor Axis: 6,356,752.31424518 m.
Inverse Flattening: 298.257223563
Eccentricity: 0.0818191908426215
I use a spherical Earth reference for WSPRnet propagation paths with:
Mean Earth Radius: 6,371,000 m.
I use only WSPR signals with a propagation distance of ≥ 1,000 km and where the transmitter and receiver stations are not in the same 2 character Maidenhead Grid reference. Any line of sight direct path transmissions are excluded and only sky wave ionospheric transmissions are included.
I have shown that when a WSPRnet signal is disturbed, the transmitting and receiving station and the aircraft are all on the same great circle path.
In total I have performed 887 GDTAAA tests. I have verified 149 progress and position indicators of various aircraft. They cannot all be right just by chance.
In the Qantas A330-200 VH-EBQ blind test there were 27 progress and 9 position indicators. Out of 214 indicators, a hit rate of 17%.
In the RNZAF Orion P-3C NZ4204 test there were 21 progress and 6 position indicators. Out of 149 indicators, a hit rate of 18%.
In the Saudi Arabian Airlines B777-268 HZ-AKF test there were 11 progress and 4 position indicators. Out of 70 indicators, a hit rate of 21%.
9. “Path of a WSPR signal” – “Additionally, the Maidenhead scheme only quantizes locations to a grid size of 2.5 miles in latitude and (depending on latitude) up to 5 miles in longitude. The cross-track errors at the aircraft from both effects are typically several nm but in some cases can be as large as over one hundred nm.”
I use only WSPR signals where the transmitter and receiver stations locations are both documented in the WSPRnet database with 6 character Maidenhead Grid references. The maximum error resulting from the quantisation of the 6 character Maidenhead Grid references is 18 nm.
For example, spot 186158668 quoted by Steve in his paper is rejected by me as the receiving station grid reference is only 4 characters.
10. “Does the tripwire effect exist?” – “Further, given the long wavelengths of HF waves, comparable or even exceeding the size of an aircraft, even the “enhanced forward scattered” signal will hardly be enhanced at all. Finally, the beamwidth of the scattered signal will be so large that the “tripwire” won’t be a wire at all, but something much larger.”
At 14 MHz the wavelength is 21.4 m.
At 28 MHz the wavelength is 10.7 m.
The Boeing 777-200ER such as 9M-MRO used to operate the MH370 flight has a length of 64 m, wingspan of 61 m and a height of 19 m.
At 14 MHz the wavelength is a third of the length of a Boeing 777-200ER.
At 28 MHz the wavelength is a sixth of the length of a Boeing 777-200ER.
Steve Kent’s paper however is an excellent rebuttal of various detractors such as Mike Exner, Victor Iannello and Ed Anderson:
1. Mike Exner: “Nuts”.
Steve states: “It was actually interesting to show that, yes, under ideal conditions, WSPR signals can easily be detected half way around the Earth, and even beyond (so yes, long path).”
2. Victor Iannello: “I understand that deep fading is characteristic of all skywave propagation due to multipath”.
Steve states: “The received S/N is well above the minimum detectable signal of -30 dB. To summarize, the minimum detectable received power is 200 dB fainter than the transmitted power of 1 W. There is a loss of about 140 dB between the transmitter and receiver due to free space propagation over 16,000 km. Thus, there is a margin of 60 dB.
3. Ed Anderson: “only short paths have been researched” and “only short paths are proper”.
Steve states: “Depending on the state of the ionosphere and other factors, it is sometimes possible for signals to bounce multiple times between the ionosphere and the ground and be received at large distances from a transmitter, and it is even possible that the “long-path” direction gives a stronger signal than the “short-path” direction.”
The science is not new and is well researched.
1. HF radio transmissions work around the globe. Amateur radio enthusiasts have shown that this works for years.
2. The WSPRnet proves it works on a daily basis since 2009 by exchanging the WSPR protocol between amateur radio transmitters and receivers around the globe.
3. Over-The-Horizon-Radar works and has been used to track aircraft over thousands of kilometres.
GDTAAA uses HF radio transmissions from the global WSPRnet network of links and detects anomalies in the network like an aircraft disturbing an invisible trip wire. The reflections and scatterings caused by an aircraft can be picked up in the global WSPRnet network just like primary radar picks up reflections from an aircraft.
Steve Kent is trying to rewrite history by claiming Marconi first transmission was “bogus”.
In 1901 Marconi received the single letter “S” in morse code over the Atlantic from Cornwall, UK to New Foundland, Canada at a distance of 3,435 km. In 1902 Marconi sailed on the S.S. Philadelphia from the UK to America recording the signals every day from Cornwall. Audio signals were recorded up to a distance of 3,400 km from Cornwall on the other side of the Atlantic.
Marconi is credited with inventing radio, received a Nobel Physics prize, holds 31 US patents and 6 British patents and built a number of businesses. At one of the Marconi companies, I started my career in Aerospace Engineering. It is perfectly possible that you sat in an aircraft whose autopilot, automatic landing system or cross-wind landing gear steering system was developed and tested by me. Your flight arrived, MH370 did not.
For MH370 I am checking the ray propagation path, number of hops, interim landing points, the elevation angle of the transmission and the maximum usable frequency for each progress and position indicator.
I actually prove each WSPRnet radio transmission I use to determine the MH370 flight path is valid between the transmitter, aircraft and receiver.
@BDW,
Welcome to the blog!
I agree with what you say, but you appear to be making an assumption about the role of the software tool Proplab Pro V3.1 in our upcoming paper titled “GDTAAA MH370 WSPRnet Analysis – Technical Report” which is planned for publication in mid March 2022.
There are 290 data points from the Civilian Terminal Area Radar (TAR) at Kota Bharu and Butterworth Air Force Base between 17:30:33 UTC and 18:00:51 UTC. We use a prior based on the position of MH370 as given by the Civilian Radar at 18:00 UTC. We track MH370 from this point forward using WSPRnet data constrained by the Boeing performance data for the aircraft type and MAS Engineering data for the actual Boeing 777-200ER aircraft with the registration 9M-MRO.
We are reliant on the fact, that irrespective of the exact state of the ionosphere at each moment in time during the flight timeframe of MH370 between the time when MH370 took off on 7th March 2014 at 16:41:53 UTC and crashed on 8th March 2014 at 00:20:24 UTC, there are 171 WSPRnet links that were sent and received using the WSPR protocol, which represents a proof of propagation. At the same time we use these WSPRnet links as candidate detections of MH370, where the link shows a SNR and/or drift anomaly. These 171 WSPRnet links are all analysed and included in our upcoming paper. In all cases the propagation path aligns with the great circle path between transmitter, aircraft and receiver. In all cases the propagation path aligns with a putative interim landing point in a multi hop ionospheric propagation.
I agree that the WSPRnet historic database contains a minimum set of secondary data, which has been subject to the processing in a WSPR receiver. We have also run a number of experiments to detect aircraft using live WSPR primary data. Many WSPR detections of aircraft do not make it into the historic database. Our experiments with the live WSPR primary data give us a much richer insight into the use of WSPR technology to detect aircraft.
The role of the software tool Proplab Pro V3.1 in our method is as a plausibility check on the possible ionospheric propagation path not a determination of the actual propagation path. This software incorporates the IRI2007 international reference ionosphere and a topographic database of the Earth’s surface. As you correctly point out, despite being quite sophisticated, Proplab Pro V3.1 and IRI2007 both have certain limitations.
I have passed on your useful comment to my collaborators and co-authors Dr. Robert Westphal, Dr. Karl Herrmann and Dr. Hannes Coetzee who may wish to chip in with their response.
@BDW,
Here is a further comment from Dr. Robert Westphal in answer to your questions about the ionosphere.
There is a very complex answer and a simple one! The complex one will be a book of 1,000 pages or more so and still not comprehensive so I will choose the simple one which is still not one single sentence!
We do not plan to build a sophisticated passive radar network based on OTH-R principles with 99.9% detection probability and 0.000001% false alarm probability irrespective of the space weather we will face!
We use the stored WSPR data for lack of better data. Yes, the ionosphere is a spatially and temporally dynamic and complex medium. There will be errors. There are many papers published with regard to regional special effects in the ionosphere at various frequencies, regions and seasons.
Some of these papers tell us of the excellent space weather conditions in February, March, April 2014.
We are looking for a single dawn, dusk and night in March 2014 during solar cycle 24 at an SFI of approximately 140 or more. If the SFI would have been 50 or lower we would not haven even started these efforts! We search at mid-latitudes so the worst effects such as strong aurora will be diminished.
By looking at SuperDARN ZHO radar data towards Western Australia in that night, studying fixed WSPR links between stationary WSPR stations we can observe if there were any dominant anomalies regarding Doppler induced by the ionosphere, such as Travelling Wave Disturbances etc. We also have the benefit of a multi-static MIMO system so we have transmissions and receptions from very different aspect angles.
Space weather is well documented for that day, of course not in every specific location of the ocean surface reflection along the assumed flight path.
We have done measurements and observations in 2020, 2021, 2022 to demonstrate the general applicability of the primary and secondary WSPR data with regard to aircraft detection. Please see the upcoming report for Southern hemisphere. There have been and there are and will be days and space weather conditions, when observations will not work.
If all three mentioned physical phenomena and IRI errors would be the killing factors, then no OTH-R, being Groundwave or Skywave type would work. And they will not work in heavy geomagnetic storms, strong aurora, or for a couple of hours or days after CMEs (Coronal Mass eruptions) etc. although billions of Dollars have been spent for these systems.
We are only a handful of people doing this work for the first time. After acknowledging our efforts I expect authorities to request agencies such as DSTG in Australia to analyse the assumed flight path with better computers, better space weather and ionosphere models and more brain power to verify or give counter proof of our observations and measurements. Maybe they could do live experiments with JORN on the assumed flight path using a B777 carrying KiwiSDRs, Winradio broadband SDR and other scientific equipment.
The DSTG published papers on the climatological OTH-R performance on the Boeing 777 at the various seasons and various SFI in Australia helped us enormously with our own assumptions and measurements on MH370 and other B777.
We are prepared for an exchange of information and assumptions on a constructive basis with experts in the field, experts, mathematicians, physicists and engineers. Professional experts could provide data and analysis of ionograms, SuperDARN data in the Southern hemisphere (at Virginia Tech these data exist for the 7th and 8th March 2014), existing JORN data in the time period in question, ADS-X and other air traffic data for that day all of Oceania and on and on….
In summary we could determine the possibility and estimate the ionospheric errors induced for that specific night in the regions in question for Oceania. We do not depend on one single second or minute of irregularities as we monitor for approximately 8 hours of flight duration. So singular irregularities will have minor impacts on the search for MH370 by stored WSPR data.