Limiting the power consumption of distance measurement

[Matthew]

Hi there, 

I'm looking for some advice on limiting the power consumption of distance measurement, as it seems I cannot simply configure the example code to do a single form of distance measurement ie: IFFT, and after searching, I was not able to find any documentation for how the different forms of distance measurement are calculated within this code. 

Is there any way to do this that I am missing?

Is there any documentation towards how calculations are made that I may use?

Thanks for any help 

[Kenneth]

Hello,

Just referring a bit to the documentation for the library here:
https://developer.nordicsemi.com/nRF_Connect_SDK/doc/latest/nrfxlib/nrf_dm/README.html#nrf-dm 

It's a closed library that perform all the mathematical calculations, the only option you have is either use nrf_dm_calc ("faster") vs. nrf_dm_high_precision_calc ("slower"). I expect most of the current consumption here is comming from the radio activity (which will be the same for all), not the time it takes to perform the calculations.

Kenneth

[Matthew]

I appreciate the response. 

Yes I would agree that all of the current consumption here is seemingly coming from the radio activity, my question is, can I limit this radio activity?

Is there a way to make a device on one side of the transaction not process as many calculations? Or just process one type of distance measurement between the two devices?

Thanks

[Kenneth]

Hello,

There are two major methods in how the radio can be used, this is controlled by either the round-trip timing or the multic-carrier phase difference:
https://developer.nordicsemi.com/nRF_Connect_SDK/doc/latest/nrfxlib/nrf_dm/doc/nrf_dm_overview.html

For the various calculation methods using multi-carrier phase the radio usage is the same whether you calculate for one or several of the ifft, phase sloop and friis path loss. (Or maybe running the friis path loss only could have significant reduction in time on air, but that one is very inaccurate and don't add any current (neither radio or calculation) to the other methods.)

Best regards,
Kenenth

[Matthew]

Would you be able to explain how the "quality" field of a distance measurement is measured? 

In the documentation here:

"

1. The initiator transmits a signal to the reflector.

2. The reflector measures the phase difference to its local reference.

3. The reflector transmits a signal back to the initiator on the same frequency.

4. The initiator measures the phase difference to its local reference.

5. The initiator and reflector jump to a new frequency and repeat the measurement.

6. The library then calculates the distance between devices based on the measured phase differences as a function of the frequency.

"

What is the purpose of switching to a new frequency and repeating the measurement?

Thanks 

[Kenneth]

Hello,

Please check out: https://www.youtube.com/watch?v=5vZIILTWmW0

Kenneth

[Matthew]

Thanks Kenneth, 

I have seen this webinar and appreciate your direction towards it.

I guess I am looking for more specifics on how this quality measurement is done, as

"

13:50

of the of the library so there are a lot of nice features with this one one of them is that we do have

13:56

a quality indicator so we do estimation on the performance of the

14:02

data that is coming in and

14:08

this allows the application layer to determine if the data is sound if it's

14:14

if you can trust it implicitly or if you need to handle it with care or if

14:19

you should just discard the data um we also control the tx power as well

" 

comes across as a little vague. 

How is the performance of the data coming in measured?

How is this data measured as 'sound'? 

[Kenneth]

Hi,

I can understand the question, but the algorithm is developed by us and we do not want to expose much details on how it's done. The intention it should just be used as decribed, the short answer I can say is that the NRF_DM_QUALITY_POOR vs NRF_DM_QUALITY_DO_NOT_USE qualities are set when we detect issues with the sampling procedure.

Kenneth

[Matthew]

Fair enough, I really appreciate the help, Kenneth.