Time-of-flight in TWR protocol

[alexpic]

Hi everybody
I'm studying the communication between Crazyflie and LPS Anchor; i'm studying the Crazyflie firmware, in particular the script lpsTwrTag.c; in this script, after studying the TWR protocol, i observe that t_prop (https://github.com/bitcraze/crazyflie-f ... #L177-L181), that i think is time_propagation,

Code: Select all

     tprop_ctn = ((tround1*tround2) - (treply1*treply2)) / (tround1 + tround2 + treply1 + treply2);
      tprop = tprop_ctn / LOCODECK_TS_FREQ;
      options->distance[current_anchor] = SPEED_OF_LIGHT * tprop;

has a different form compared to the time-of flight descripted in your wiki:

So my questions are:

• What is mean LOCODECK_TS_FREQ?

• What is the difference between tprop_ctn and the theoretycal t_f?

• Why tprop_ctn is calculated in that way?

• Finally, Why, for calculate tprop, tprop_ctn is divided by LOCODECK_TS_FREQ?

Thanks a lot for your help and sorry for so much questions
Alex

[arnaud]

Hi,

For the formula, there is a typo in the wiki. I fixed it, thanks for reporting the discrepancy!

As for your questions:

• LOCODECK_TS_FREQ is the frequency of the UWB radio timer, it is a 40bit timer that counts up at about 64GHz and that is sampled when a packet leaves or is received by the radio
• tprop_ctn is t_f measured in UWB timer ticks
• This way of calculating T_prop allows to calculate the time of flight in a 'virtual clock' that is between the clock of the sender and the receiver drift-wise, this is done to avoid most of the error related to clock drift. there is a white board demonstration in this thread: viewtopic.php?f=6&t=1944&p=9965#p9959.
• tprop_ctn is a timer count, dividing it by the timer frequency gives us tprop in seconds (this is t_f).

If you can share your study when you are done I feel it would be a great addition to the documentation for the LPS

[alexpic]

Hi Arnaud!
Thanks for your reply!! Now it's more clear!!

For the formula, there is a typo in the wiki. I fixed it, thanks for reporting the discrepancy!

You're welcome! It was a pleasure!

If you can share your study when you are done I feel it would be a great addition to the documentation for the LPS

Of course! it'll be a pleasure to be helpful!

PS: Thanks for the analytical demostration, i was searching just that!!!!!