Just my 2p:
Alex_ledin - Like the idea of a database-driven approach, but it's maybe susceptible to other inputs giving false readings and exacerbating the feeling of a loss of control of your senses (no pun intended).
bob.arctor - Linux and MacOS don't crash?? Don't make me laugh!
"Laser" anemometers - good idea for a backup that works in a totally different way - shouldn't be too expensive to mass-produce either? Also, if you have a problem with finding particles that far up, create your own! Just stick a smoke generator a few feet in front if necessary.
And now on to my off-the-wall idea for measurement:
Sensors in each jet engine in the compressor stage, measuring the amount of air resistance occurring at the fan blades.
Not sure how sensitive it might be (or rather how insensitive the jet engine is for a given state of fuel entry), but it's one place to try?