Radio propagation model and tuning algorithm

[Balt]

Hi Hardy

Can you provide a little more detail about how you are calculating the VOR/DME signal propagation in the sim? Are you emulating the actual FMS behaviour for the auto tune algorithm or is this a best guess from observed behaviour?

Cheers

Balt

[Hardy Heinlin]

Hi Balt,

the models are based on various documents and formulas. I don't want to provide more code details at the moment -- too much work ...


Cheers,

|-|ardy

[Balt]

Hi Hardy,

understandable, no problem. Based on some experiments it would seem that you are modelling propagation in altitude dependence (of course). What's of more importance to me is whether the tuning algorithm for the DME DME position crosscheck is reasonably close to the original, can you comment on that?

A little bit of background information: In my day job I'm an astrophysicist, working on building a new telescope in Western Australia called ASKAP (https://en.wikipedia.org/wiki/Australian_Square_Kilometre_Array_Pathfinder for more info). This telescope lives in a radio quiet zone, an area the size of the Netherlands that's been put aside by the government to not have any radio transmitters in it in the bands where this traditionally causes a lot of problems for low frequency radio astronomy, so between about 30 kHz to 20 GHz. Above that frequency, radio frequency interference (RFI) becomes rather directional, so unless you're staring at it with the antennas, you're not likely to be affected by it.

Other than space based RFI (satellites), aviation is the second largest impact factor for the frequencies we're observing in, from 700 - 1700 MHz. There's ADS-B at 1090, Mode S at 1030, and a veritable forest of DME frequencies (very narrow band) between 952 - 1150 MHz.

Prediction is half the war in RFI world, so if we can characterise better what to expect and when to expect it, we can win the battle. I've made a flight plan for flying the airways around the ASKAP site with PSX and am recording the autotune DME frequencies, where they are switched etc. The plan then is to then match this with actual flight movements (we have ADS-B receivers at the site for that purpose) as well as the DME induced RFI data we're recording, we should see when the "real deal" switches the DME channels and should then be able to reverse match that with what PSX has produced.

Even if nothing comes of it, at least it is a fun project and links aviation with astronomy, combining my two callings...

So Hardy, if you could just try and remember whether you implemented code as per some Honeywell documentation (or Pegasus or whoever manufactures the FMS?) or whether you're just looking for the nearest two DME's, that'd be much appreciated.
Cheers

- Balt

[Hardy Heinlin]

The primary factor is the nearest distance. At places where two stations are close together, one of the two is typically of a different power class, in which case the higher power class is used. Geometry may be a factor as well, but this is rarely a problem because the two nearest stations typically have a great arc angle relative to the aircraft anyway. And if there really are two stations almost in line, then there is typically no better third station anyway, e.g. near an oceanic coast line.


Cheers,

|-|ardy


If you're wondering why it tunes GEL instead of MEK in your area, although GEL is 200 nm, MEK is 100 nm away:  GEL is in the high power class, MEK is in the low power class. It tunes low power stations only if they are closer than 70 nm. (I can't remember where I got this figure from.)

[Balt]

Thanks Hardy... indeed, one of the mysteries thus far has been the absence of seeing MEK being used (in real life). PSX is behaving in the same way, and looking at high alt charts, MEK VORDME isn't even shown only the NDB. Things are looking fairly consistent for the time being.

Cheers

- Balt