Hi, I would like to ask a few things about Channel Sounding (CS): Is there any recommendation which method (IFFT, PBR, RTT) should be used for reading the results? From my experience for short distance indoors under 10 m the IFFT looks the best to me. But outside without reflective surfaces this method seems to fail a lot - it makes big jumps, for example from 30 meters it suddenly drops to 2 meters for a short moment and then goes back. I did not study it deeper yet, but maybe it is needed to choose the method based on the distance, or somehow mix them?!? Or do these IFFT jumps mean there is some mistake in my code or hardware? There is RED curve for IFFT method - we can see the "shot" in data: I found out that the period for getting the measured distance is strongly affected by this setting: .conn_param = BT_LE_CONN_PARAM(80, 120, 1, 100) I did not find anywhere how this is connected with CS procedures - I was tuning it only by try and fail. But the final period when the CS procedure really happens is strongly affected by it (by hundreds of ms) Thank you, Jan.

Channel Sounding – which method for distance measurement?

witc 7 months ago

Hi,

I would like to ask a few things about Channel Sounding (CS):

Is there any recommendation which method (IFFT, PBR, RTT) should be used for reading the results? From my experience for short distance indoors under 10 m the IFFT looks the best to me. But outside without reflective surfaces this method seems to fail a lot - it makes big jumps, for example from 30 meters it suddenly drops to 2 meters for a short moment and then goes back. I did not study it deeper yet, but maybe it is needed to choose the method based on the distance, or somehow mix them?!?
Or do these IFFT jumps mean there is some mistake in my code or hardware?
There is RED curve for IFFT method - we can see the "shot" in data:
I found out that the period for getting the measured distance is strongly affected by this setting:
.conn_param = BT_LE_CONN_PARAM(80, 120, 1, 100)
I did not find anywhere how this is connected with CS procedures - I was tuning it only by try and fail. But the final period when the CS procedure really happens is strongly affected by it (by hundreds of ms)

Thank you, Jan.