IQ Sampling without antenna array

[SDK_User_771]

Hi,

I got a question on IQ Sampling with the direction_finding_connectionless_tx/rx example.

Settings:

Sample Spacing = 125 ns

Sample Sapcing (Reference Period) = 125 ns

DFEPACKET.MAXCNT = 656

CTE Length = 160 us

Switch Slot/Sample Slot Length: 2 us

=> Number of Sample Slots: 37 

So 1 Sample/125 ns means 16 Samples/2 us Sample Slot and additionally 64 Samples during the reference period.

Problem:

Without the guard period and the switching slots, I should be able to get  (Number of Sample Slots)*16 + 64 Samples, which evaluates to 656 Samples with 37 Sample Slots.

But I only get 101 Samples. 

I am not using an antenna array (so only the chip antenna should do the sampling) , and have reduced the number of antenna patterns to {0x1, 0x2, 0x3} since 3 are mandatory for the code to work. My assumption is that still some antenna switching is performed in the code and I am looseing samples to that. 

Question:

What are the correct settings to get IQ samples only from the chip antenna, and if this is not possible, how can this number of 101 Samples be explained that I receive ?

(I also don't necessarily need the  IQ  samples in the context of direction finding, so I am also happy with a solution that disables the funcionality of this feature.)

Best regards 

[Simonr]

Hi

I don't think you'll be able to disable the switching periods etc. while using the Direction Finding sample applications like this, but I have asked the developers internally if they have any clever ideas to purely maximize the number of IQ samples you can achieve.

AFAIK the only documentation we have on doing this manually is the Direction Finding whitepaper that shows you how to set up every part of IQ sampling separately, so if you set up IQ sampling without antenna switching from scratch that would be my guess as to what the best procedure forward would be. I'll let you know if I hear anything from the developers on the matter.

Best regards,
Simon

[Simonr]

Hi

Some input from one of our developers working on the Direction Finding features:

It's possible to do IQ samples with the manual setup from the Direction Finding whitepaper. If you use a single antenna, it should already skip using the reference, sample and switch periods and only take samples.

The guard period can not be disabled, but it is possible to use proprietary configuration for TSAMPLESPACINGREF and TSAMPLESPACING. They also can setup TSWITCHSPACING to 8us to make the radio attempt to switch antennas not so often.

Best regards,

Simon

[mariakenneth]

Interesting challenge with IQ sampling! It seems like you're navigating some complex settings here. If only there were an iq test free of these technical hurdles to clarify the sampling process! Good luck with your adjustments!