Channel Sounding Synchronization Mechanism and Application

[cheng]

Dear Experts,

I would like to better understand what exactly is accomplished by the mode-0 synchronization in Channel Sounding. My current understanding is that it seems to measure the frequency offset between two devices, and sometimes also transmits the FAE table (although in one response, an expert mentioned that Nordic does not require this：nRF54L15 Channel sounding FAE table issue - Nordic Q&A - Nordic DevZone - Nordic DevZone). My question is: does mode-0 synchronization need to be performed in every subevent? If the FAE is a fixed table for a given device, then for multiple measurements, is it sufficient to complete synchronization in just the first subevent and continue using the already obtained synchronization status for subsequent subevents?

Based on this, I have a second question: can the results measured by mode-0 be accessed and used by my code? For example, could I use the measured frequency offset between two devices for other purposes in my application? Does the synchronization achieved by mode-0 allow the two devices to remain in this state continuously? If I am able to keep the PLL (phase-locked loop) running and avoid turning off the crystal oscillator, is it possible for the two devices to maintain their synchronized state over time?

I also noticed a register called CSTONE.PCT, which seems to correspond to the phase correction process defined in the protocol. Is it possible that the value of this register represents the measured frequency offset between the two devices? I am not sure if my understanding is correct.

Best regards,
Cheng

[Asbjørn]

Dear Cheng,

For nordics implementation we support using mode-0 synch, but it is no necessary. This is sometimes referred to as CS without Frequency Actuation Error (FAE). Essentially, it means the initiator transmits on the frequency specified by the channel number without any further frequency adjustments. This can be beneficial for simplifying the ranging process and ensuring consistency across different devices. It can be enabled on Nordic if needed by the other end of the ranging.

I'd encourage you to have a look at Bluetooth SIG own technical summary of the spec 6 features, including channel sounding here: https://www.bluetooth.com/core-specification-6-feature-overview/

To keep the synch you would have to re-synch as they will drift compared to each other.

CSTONE.PCT, where are you finding this register? Are you able to share some link to what you are referring to?

Best regards

Asbjørn

[cheng]

Hi Asbjørn,

Thank you very much for your reply.

When I run channel sounding tests using two 54L15 development boards, how can I confirm whether mode-0 is enabled? In nCS 2.9.0/3.0.0, the minimum step number for mode-0 is 1, which means it should be executed at least once at the start of each subevent. If both sides are using nRF54L15, is it possible that nothing actually happens during mode-0? I’m not sure if my understanding is correct.

Regarding the register I mentioned, it can be found in the nRF54L15 Product Specification (https://docs.nordicsemi.com/bundle/ps_nrf54L15/page/radio.html#ariaid-title159). After reviewing the documentation again, it seems that CSTONES.FFOSET is more relevant for frequency offset measurement.

Best regards,
Cheng

[Asbjørn]

Hello Cheng,

Are you logging any output from the nRF54L15 during this on the RTT? There should be a statement in the beginning on the connection that says "CS capability exchanged completed" or along those line. That would be the indication that the exchange has happened and then the next phase of the measurement is done with multiple packages sent and reflected. 

Best regards

Asbjørn

[cheng]

Hi Asbjon,

I can see this output. Is this the output after mode-0 has finished executing?
I am currently using two nRF 54L15 devices. I only see this output during the connection establishment, and for each subsequent subevent measurement, the results are output directly.
In each subevent, is the mode-0 step also present? According to the protocol, mode-0 must be executed at least once at the beginning of every subevent.

Best,
Cheng

[Asbjørn]

Hi,

it's done at connection establishment as you see. Since it is two Nordic devices, it is not retriggered by default. Mode-0 step is still present, but not renegotiated.. I'd expect you to see different output if you were connected to a different vendors reflector. 

Bestregards

Asbjørn