[TowerTalk] If you had a choice

[Jim Lux]

On 10/17/12 2:01 PM, David Gilbert wrote:
>
> Those $300 quad-copters might not have the beef to handle the task, but
> I'm pretty sure that the more expensive octo-copter machines (about
> $3500) could.  They can lift upwards of 2 kilograms, and they not only
> can be controlled from the ground they can be pre-programmed to navigate
> whatever path you want via their own onboard GPS.  I think it would take
> a pretty stiff wind to throw them off course.  The biggest downside to
> using them might be battery life, but pretty much the entire measurement
> process could be automated so that it didn't require much air time.  It
> might even be possible for such a machine to lift its own power cord for
> supply from the ground if the measurement protocol allowed for a fixed
> X-Y position (i.e., the antenna could be rotated).
>
> But in general, this should be possible:
>
> *  pre-programmed, GPS controlled flight plan consisting of concentric
> rings at multiple heights in a spherical configuration.
>
> *  onboard field strength detector with a small sense antenna
>
> *  onboard transmitter to send back real time signal strength readings
> and GPS coordinates to a ground based receiver feeding a laptop for data
> collection.
>
> I suspect that if someone wanted to get fancy they could feed the test
> antenna with broadband noise (pulsed for recognition) and put an SDR in
> the copter.   There are some interesting possibilities out there ....
>
> 73,
> Dave   AB7E
>


I was going to fly a PCR-1000 receiver (which has a reasonably well 
calibrated power detection) and a switch which would select one of 3 
orthogonal "electrically short" dipoles.  The receiver would be tuned in 
a sequence of frequencies across the bands, stepping through the 
sequence at 1 Hz or something like that.

You'd just log the GPS position, orientation, etc. along with the data 
from the receiver.  Then you'd transmit 10 second long CW signals from 
the antenna under test at each frequency.

Then you'd post process.

The receiver part is easy.  The flying part is hard (or, at least it was 
back when I had this idea).  As you say, today, you could probably get a 
several kilobuck quad or hexcopter that has good stationkeeping and 
autopilot capabilities.

The $300 Parrot AR-Drone could not do this.. (I have one.. it's a heck 
of a lot of fun, and can get 50-100 ft up, but it doesn't have the load 
capacity, nor the autopilot sophistication...)

You could make the remote unit a transmitter, which would reduce the 
weight a lot. Something like a PIC or Arduino driving a DDS and the 
appropriate RF switch (e.g. a CMOS gate) to the three dipoles would work 
nicely. Probably do it all for <100 grams.