Regarding I and Q measurements on the nRF54L15 development kits

RE: Regarding I and Q measurements on the nRF54L15 development kits

Jack jpeng — Fri, 07 Nov 2025 07:29:22 GMT

Hello Ignatious,
I'm using your method for outputting IQ values, but the terminal displays the issue shown in the image. How did you resolve this? It outputs correctly a few times initially, but subsequent outputs are all zeros.

Best regards

Jack jpeng

RE: Regarding I and Q measurements on the nRF54L15 development kits

ignatious-p — Thu, 19 Jun 2025 09:27:03 GMT

Understood. Thanks for the help, Simon! I believe I have my questions answered, but I don't know which one of your responses to mark as the answer. Could you kindly collate your responses into one message, so I could mark this ticket as resolved?

Regards,

Ignatious

RE: Regarding I and Q measurements on the nRF54L15 development kits

Simonr — Thu, 19 Jun 2025 09:04:55 GMT

2. Yes, the 0s are from channels that aren't used by channel sounding. The content of the cs_de_iq_tones_t is not indexed by channel, and starts at the first valid frequency rather than at channel 0. Note that you should not rely 100% on the index here as it is very likely that there will be changes to this API since it's currently experimental. A PR was drafted today to clarify this in code: https://github.com/nrfconnect/sdk-nrf/pull/22889

Best regards,

Simon

RE: Regarding I and Q measurements on the nRF54L15 development kits

Simonr — Thu, 19 Jun 2025 08:05:27 GMT

1. Correct.

2. This sounds plausible, but I will need to check with the developers of our Channel Sounding solution to confirm what this 0 readings are exactly.

Best regards,

Simon

RE: Regarding I and Q measurements on the nRF54L15 development kits

ignatious-p — Wed, 18 Jun 2025 11:36:52 GMT

Dear Simon,

Thank you very much for your response!

1. I see. It seems that according to the BLE v6.0 Core Specification Vol 6 Part H, Section 4.6 they are complex numbers.

"Each measurement within T_PM results in a single PCT value expressed as complex
number I+jQ."

So according bluetooth/cs_de.h are they complex numbers encoded as 32-bit float values, yes?

2. These '0' readings keep happening on the same indices. I noticed that according to bluetooth/cs_de.h these correspond to the tones. If I am understanding this correctly, this corresponds to the 79 channels that are used for channel sounding? If we look at the indices that are missing data and observe the forbidden channels for CS, the indices that are "missing" correspond to 2 less than the channel number (a.k.a it seems that the index 0 in Nordic's implementation refers to measurements made in channel 2 (2404Mhz) and not channel 0 (2402 MHz) as in the BLE spec). I may be completely off the mark here, what do you think, Simon?

Regards,

Ignatious

RE: Regarding I and Q measurements on the nRF54L15 development kits

Simonr — Wed, 18 Jun 2025 10:39:42 GMT

Hi Ignatious

I'm so sorry, but your ticket fell through our cracks. I'll handle it from here and we'll try to get you sorted.

1. This is raw IQ data (not degrees) and is described in the Bluetooth core specification on the section of IQ data.

2. 0 could be because the ranging wasn't completed or if the buffer overflows or similar I believe. In short it could be multiple reasons for this.

Best regards,

Simon