Using historical WSPR data to try and track MH370…
My instinctive view is :
There is a third option which is to use it as a test for known aircraft tracks and see how well it's works.
I suspect this has already been done 😉
@simonzerafa yes I believe this is what the newer paper has done. I really should read it.
My preconception is it will be drawing targets around bullet holes. But maybe that preconception will prove wrong.
@bloor @simonzerafa
The ‘author’ tried that a number of times.
He had previously built a flight path prediction tool. Using regularly operated routes, which generally follow airways (perhaps with offsets), and knowing weather (head/xwinds), paths could be predicted but a number of times his predictions/alleged WSPR derived tracks were contradicted by ADS-C data logged at adsbexchange.
How accurate is ASD-C and what was the margin of error when compared with WSPR? 🤔
@simonzerafa @bloor ADS-C is typically involves reports every 15mins (C = contract, the contracted reporting period may vary, ANSP to ANSP). The reports are generated by the FMS using its inertial source, at very worst a few 100m from GNSS.
The ‘author’ alleged that he’d tracked a QF B787 from Sth Africa to Aus. Cross referencing what his WSPR mining had divined to ADS-C reports, errors in ‘4D’. IIRC it was about 50nm.
50 nm isn't as accurate as would be really desired but might be better than nothing? 😟🤷♂️
@simonzerafa @bloor
It’s a guess.
Guided by predictions of a flight path using other known information.
The first ‘test’ was a total fail - the target aircraft was allegedly tracked to Melbourne. However, it landed at Perth (after crossing the Indian Ocean from Jo’burg).
