4

Any reasonable priced (sub 20$/100) ways to directly measure indoors velocities between around 0.1 and 15m/s?

Its main purpose would be to reduce speed drift for a 9DOF device (acc/gyro,magnetometer), so I believe only 1D will do.

Now since it's for indoor use, a GPS won't help. I assume this could be done measuring RF/ultrasonic doppler shift, but I don't know how this is measured, nor have I found dedicated chips to do it.

Imbrondir
  • 695
  • 4
  • 17
  • Shopping questions are not allowed on Stack Exchange; however, this question seems like a very good question to be asking. It would be helpful if you were to reword your question such that you say what it is you are trying to make and ask for help with it. – Kellenjb Aug 02 '11 at 13:13
  • @Kellennjb - noted. I've reworded it slightly. Though does no shopping question mean I can't specify what price range I'm after? – Imbrondir Aug 02 '11 at 21:28
  • Price is very important for most projects so it is fully reasonable to say you have a certain price point that you are trying to obtain. – Kellenjb Aug 02 '11 at 22:06
  • RF Doppler shift is tough.. You need ADCs that would cost an arm and a leg. Ultrasonic is easier but still not that easy. Good luck and let us know if you find something, I am particularly interested in ultrasonic. – Frank Aug 03 '11 at 05:01
  • @Frank - the shifts in question - between 7 Hz and 1 KHz for a 10.5 GHz signal - are in the range where you could sample them nicely with an inexpensive audio ADC. – Chris Stratton Aug 03 '11 at 21:30
  • @chris_stratton I would have guess you need to sample at your carrier frequency * 2(nyquist). For a 40Khz audio, you need a 80Khz sampler, am I wrong? For RF, you need several GHz. I like to understand your approach. – Frank Aug 04 '11 at 02:25
  • @Frank - you mix down to a difference frequency in the audio range. In a simple doppler application its likely the transmitted signal is implicitly used as the local oscillator for the receive mixer, such that energy reflected off of stationary objects would be at 0Hz while anything reflected with a shift would have a positive or negative frequency, likely in the audio range. A simple low pass filter and audio ADC should suffice at that point, possibly with a DC blocking cap if you want to amplify the shifted signals without worrying about the unshifted & transmitted energy. – Chris Stratton Aug 04 '11 at 19:02
  • @chris_stratton I like to better understand this. I am going to ask a question, would you be kind enough to explain further. Thx – Frank Aug 05 '11 at 06:07

1 Answers1

4

I suspect that a Gunnplexer as used in Police RADARs and in door openers prior to PIR's taking over would do what you want. Output is Doppler difference of go + motion affected return signal.

enter image description here!

A few examples:

Russell McMahon
  • 147,325
  • 18
  • 210
  • 386