5
$\begingroup$

I have a 4 mic array on a tripod that I use to record Pipistrelles flying about 3-5 m above the ground over a lake.

My calls are always 'double' for some reason (two calls, the direct path and some reflection) - which makes acoustic tracking hard.

I'm actually recording in an open space, and don't expect any reflections. Can someone please tell me why I'm getting this 'double call' type data, and what I can do about it?

Edit:

enter image description here Hand sketch of the 4 channel inverted T array on the tripod. The long line on the right of the array is the water surface.

Note: this is a 're-enacted' question. Don't have the spectrograms at hand now, will try to find it in the coming days

Improve this question
$\endgroup$

3 Answers 3

Reset to default
4
$\begingroup$

Even while in open space, the lake that your bat is flying over is a great reflector, and so you should anticipate recording click-echo pairs. If this is frustrating your automated tracking algorithm, adjust your detector so that it has a blanking time (following each detected click) that is longer than your longest interval between click and echo. (This can remove echoes from being detected in the first place). Good luck

Improve this answer
$\endgroup$
4
$\begingroup$

Without having a picture of your setup and environment, we can only assume and speculate.

I guess, you will always have significant reflections when bat is flying over the flat water surface.

To know where the reflection is coming from, I would first measure the time delay, convert it to range difference, and then try to resolve geometry.

IMO, having reflections, makes tracking not only easier, but enables it by multipath analysis. Especially, if you use your 4 microphone array to estimate the direction of the arrivals.

I guess further, the reflected call overlaps the direct call, making direction estimation somewhat complicated. More information on microphone separation and bat signal would be needed or helpful.

Improve this answer
$\endgroup$
1
  • $\begingroup$ Am on the road, and can't access field photos. Hope my quick sketch helps :). $\endgroup$
    – Thejasvi
    Commented Jul 8, 2022 at 5:06
2
$\begingroup$

This happens a lot, and @WMXZ and @chloe have covered the major solutions and ideas to handle the problem.

The echoes do come from the ground/water-surface. What I've found helps is to raise the height of the tripod (say at least >= 1 m). The time delay between the direct call and reflection increase and you can then localise calls due to the temporal separation.

On a related note (and an untested idea), if you can't go any higher - may even get the array as low as possible? Here, the time delay between the direct and indirect would reduce and end up showing as only 'one' call? Not sure how much it would affect localisation though!

Improve this answer
$\endgroup$
1
  • 2
    $\begingroup$ In some engineering applications it is common to use microphones positioned directly on the ground to avoid significant ground/surface reflections. Note however that the acoustic impedance of the soil might absorb some high frequencies, and this will be highly location- and weather-dependent, so it might not be the best option here; hence the mics are often positioned on a metal plate. An example can be found at nasa.gov/centers/armstrong/features/… $\endgroup$
    – fabio_hirono
    Commented Jul 8, 2022 at 10:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.