• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 31 subscribers
  • Views 577 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.

IWR6843ISK: data transfer implementation?

Konstantin Mikhaylov
Prodigy 130 points
Part Number: IWR6843ISK

Dear TI engineers,

Sorry for somewhat a strange question, but is this technically possible to make mmWave sensors (for example – the IWR6843ISK + BOOST dev. kit) transmit and receive the data? I understand pretty well that these systems have been designed and optimized as radars, and such use has been (probably) never planned for them, but for me this looks like their capabilities can enable such a use case. Can you, please, comment whether the following ideas make any sense?

  1. mmWave radar as a transmitter. First, probably, the slope of the chirp needs to be set to 0. Then, the data might be encoded as OOK (by varying the idle time between the chirps and chirp durations), the BFSK/QFSK (by changing the starting frequency of profiles) or BPSK (using BPM feature). Obviously, the limited IF bandwidth of mmWave radar in receive is a limiting factor and should be accounted for.
  2. mmWave radar as a receiver. I see two major challenges here.
    1. The first one is the loopback between the transmitter and the receiver. Thus, in order to be able to receive something, the device will have to transmit something. After this, on the receive antenna, we will have both the useful data signal and the refractions from objects in the environment of the transmitted signal. If to go with 2FSK mode, the TX can, for example, send chirps with F0+4MHz and F0-4MHz (assuming 10 MHz IF bandwidth at receiver), while receiver may be configured to F0 (either carrier wave or chirps with 0 slope). Then the signals should be possible to detect.
    2. The second problem is the fact that, as far as I understand, the ADC at the receive side starts only after a chirp is transmitted. An easy solution would be to have a carrier wave mode with ADC periodically sampling the channel, but, as far as I understand, this is not the case for the current TI devices. A workaround can be to use at transmitter frames with one single chirp with frame period equal or higher than the period of 2FSK “symbols”. Given the high flexibility of TI mmWave radars for chirp duration, this seems to be possible.

Does this makes sense? Maybe there are some other hidden stones, like, e.g., the thermal management?

With best regards,

-Konstantin

  • 0
    TI__Mastermind 30405 points

    Hi Konstantin,

    As you already know, these chips were designed as sensors only and so such a use case is not supported from a hardware and software/systems perspective from TI so we will not be able to provide any support on this effort.

    Regarding some of your questions:

    1. mmWave radar as a transmitter. First, probably, the slope of the chirp needs to be set to 0. Then, the data might be encoded as OOK (by varying the idle time between the chirps and chirp durations), the BFSK/QFSK (by changing the starting frequency of profiles) or BPSK (using BPM feature).

    [NS]: While these sensors do support programming up to 512 different chirps in the system, the chirps are programmed once before the sensor starts chirping. In other words, you cannot change the chirp parameters (such as idle time, start frequency, ramp slope etc) dynamically without stopping and re-configuring the sensor. So it is not possible to encode data using chirp characteristics the way you mentioned.

    2. mmWave radar as a receiver. I see two major challenges here.

    [NS]: Besides the fact that you cannot work in receive only mode as you noted, there is also the issue of synchronization between the transmitter and receiver. Also note that the received chirp signal (at 76-81GHz) is sampled at 900MHz or 1.8Ghz and goes through a series of down-conversions in order to get the IF signal at 10MHz bandwidth (not sure if the integrity of your transmitted data will be maintained if there were a way to overcome the other issues listed above).

    We do not recommend using TI mmWave sensors for this use case.

    Regards

    -Nitin