• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 17 subscribers
  • Views 1554 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Determine time of arrival with CC1125. (Hyperbolic trilateration)

Snorre Olsen
Prodigy 60 points
Other Parts Discussed in Thread: CC1125, CC1200, CC1120

Hi everyone! First of all, let me apologize for being very inexperienced in RF design/applications, I am more of a DSP guy. This is also my first forum post here.

My problem is this: I am looking into the possibility of using time-difference-of-arrival (TDOA) to trilaterate a signal source at an unknown location by determining the TDOA between 3 different base stations at known positions.

The base stations and the transmitters will both be using the CC1125 transceiver at 169 MHz. The base stations will use a GPS-based time references. The method i want to investigate is as follows:

- GPS 1 pulse-per-second at base station triggers counter.

- Counter starts counting pulses from a 100MHz OCXO

- Receiver (CC1125) picks up signal from transmitter

- Stop counter/load count value into register.

The count value will represent a time reference for when the transceiver picked up a signal from the transmitter, and comparison between base stations lets me calculate its position.

The crucial point is: how can I implement a "trigger" from the CC1125? Something that can be used to uniquely timestamp the signal.

I am very new to this, but could it be possible to set an output pin high at the end of the preamble, or something to that effect?

I am an electronics student, and this is my final year project.

Thanks for any answers, and kind regards

Snorre

  • 0
    TI__Mastermind 36395 points

    Snorre, 

    These devices all use a "SYNC" word to determine packet synchronization and I believe that is what you need for this type of activity. That being said, the sync word is only determined at 16 times oversampling compared to the Symbol rate used by the device. So I believe the CC1125 is not the best candidate. I would recommend the CC1200 for this type of work and this run it at maximum symbol rate.

    Please see the user guide for the CC1200 you will a signal called "PKT_SYNC", its used also to interrupt the MCU hosting the system indicating that a packet has arrived. (IOCFGx with option 0x06).

    Please let me know if you get this to work, I would like to see it.

    Regards,
    /TA 

  • 0 in reply to TA12012
    Prodigy 60 points

    TA

    Thanks for your reply. The project has evolved a bit now and I've gotten my hands on some CC1120DK development kits. Guidelines for the project require me to use a carrier around 169MHz, and max 50kHz bandwidth. This means, as far as I can understand, that my symbol rate will be limited to 25 ksps. 

    I have done measurements using to 2 receivers with identical setups and a common antenna, transmitting packets from a third board continuously at 25 ksps.

    The offset between the two PKT_SYNC_RXTX signals from the two receivers are on the range of +/- 10µs, suggesting that it is clocked out at a rate 4 times higher than the symbol rate (40 µs per symbol),

    Can I do anything to achieve the 16 times oversampling you mentioned?

    Is there any other signals I could use to determine the time-of-arrival of a signal more accurately? I tried measuring the CARRIER_SENSE signal, but it has the same inaccuracy.

    Also, the offset between the two receivers was not constant, but changed for every packet. I was expecting an offset due to the GPIO pins being set on clocks that were not perfectly synchronised, but I did not expect it to behave so randomly. Any thoughts on this?

    Thanks in advance for any insight provided.

    Snorre 

  • 0 in reply to Snorre Olsen
    TI__Mastermind 36395 points

    I have looked at it a little again, its actually 4 times the RX bandwidth. Therefore 50KHz, * 4 = 200khz update rate.

    To get more accuracy, you need to have a broader signal (higher RX bandwidth). Could your system operate at 2.4GHz using our Zigbee (802.15.4) type radio?

    /TA

  • 0 in reply to TA12012
    Prodigy 60 points

    Thank you for your reply. Ok, that makes a little more sense, although I would expect a maximum of 5 µs delay between the two signals at 200kHz, but I experienced 10 µs.

    At any rate, that corresponds to a propagation distance of about 1.5km which renders my system useless for tracking purposes.

    Is there any way I could get an indicator that a signal has arrived (PLL lock, RSSI above a certain threshold) that can be clocked out using the full 32MHz rate that is delivered by the oscillator on the CC1120EM?

    It would be interesting to try this approach at 2.4GHz, but I'm afraid the goal of the project was to examine how one could best perform radiolocation at VHF frequencies, and 169 MHz is the band that is allowed for these purposes in Norway. 

    Any other ideas?

    Thanks in advance

    Snorre

  • 0 in reply to Snorre Olsen
    Prodigy 140 points

    Hi Olsen,

    I'm working on the same problem with CC1120. I have measure the same problem and working to estimate the better solution by many packets processing on MCU.

    My target is to measure a range multiple of Nx10 meter between two CC1120 or CC1200 for raw distance estimation.

    Have you find any solution and signals able to detect ToF with CC112x?

    Regards

    Maurizio Brignoli