XeThru X2M200 Web Explorer

Discussion in 'XeThru Explorer' started by Debanshu, Jun 20, 2017.

  1. Debanshu

    Debanshu New Member


    I am experimenting with the x2m200 to create a meditation room focussing on the breathing exercises synced with philips hue lights. Everything is communicating flawlessly except for the sensor and the breath sync i.e. the values received by the sensor are very jittery and not in sync with the actual breathing. Even in the XeThru explorer the signal quality randomly switches between 0-10 and there are huge gaps in data.

    1. Is there a best practice of setting up the sensor in a physical environment?
    2.Does it have to placed with respect to the user?
    3. Should I not use the Web Explorer as its still in it beta phase.

    Any leads would be very helpful.

    I am using the XeThru web explorer demo and then using processing to listen to the local host and getting the values from there to control the philips hue lights.
  2. Olav Liseth

    Olav Liseth Administrator Staff Member

    Hi Debanshu,

    The sensor is designed for respiration detection during sleep or when the person is sitting still. Make sure that the user is sitting steady and not moving. You will get best results, especially when you have to rely on the breathing pattern, when the sensor points straight to the user's chest. See this app note for more details: https://www.xethru.com/community/resources/x2m200-sleep-monitoring-introduction.66/

    See recent similar project: https://www.xethru.com/community/threads/using-xethru-x2m200-made-an-interactive-work.206/

  3. Laura Leponiemi

    Laura Leponiemi New Member

    We're running some tests too by collecting continuous data with the X4M200 respiration sensor and experiencing also always long gaps in the data around same time. What possible reason may cause these gaps? Gap = no respiration data, not even zero values.

    Thank you for help!
  4. Harvord Sun

    Harvord Sun New Member

    Hello,I am trying to use my own MCU to control the X2 chip,and I wonder if it possible to have an access to the hex file or source code running on the MCU of X2M200 board.Looking forward to your replying~
  5. Charlie Shao

    Charlie Shao Moderator Staff Member

    Hi Sun,
    X2M200 will be replaced by X4M200 soon. Suggest you switch to X4 for your development. X2M200/X4M200 is our commercial sensor product, so we are not able to share the source code, but you can start from X4M03 radar development kit and run open-source XEP(XeThru radar open-source embedded development firmware) on it. XEP help you quickly configure X4 and get radar data, so you can quickly start and add your own algorithm.
  6. Harvord Sun

    Harvord Sun New Member

    Thank you for your replying.Now I have changed to X4M200,and I'm trying to get the raw data from the module by running xt_module_print_record_playback_radar_raw_data_message.cpp where the parameter, downconversion, can be set to 0 or 1.I looked up in the X4 datasheet that when the parameter is 0,the sampling rate is 23Gsps,and I failed to get the exact sampling rate when it turns to be 1,which means the raw data is downconversioned to complex baseband data.Would you help me with this puzzle?Thanks a lot.
  7. Charlie Shao

    Charlie Shao Moderator Staff Member

    There are two types of raw data. We call the original radar raw data RF data. There is hardware circuit in X4 can convert RF data to complex baseband data, if you set downconversion data to 1, X4 will output baseband data, but the original RF data is still sampled with 23Gsps.
  8. Harvord Sun

    Harvord Sun New Member

    Thank you very much~Your answer makes much sense to me, but I still want to know the exact time between any two adjacent points of the baseband data in one frame, which can be calculated into distance in the transmisson of radiowave.
  9. Charlie Shao

    Charlie Shao Moderator Staff Member

    RF bin (or sampling bin number) is 1536 according to X4 datasheet. Given Fs=23.328GHz, the frame length is:
    LightSpeed = 299792458
    rf_interval = LightSpeed/23.328e9/2 = 0.0064255928069272975
    (1536-1)*rf_interval= 9.863284958633402
    The downconversion process uses decimation factor 8 with a 4 bins delay, you get 2.916 GS/s sampling rate and 1536/8-4= 188 bins baseband data.
    bb_interval = LightSpeed/2.916e9/2 = 0.05140474245541838
    (188-1)*bb_interval = 9.8632849586 = 9.664091581618655
  10. Harvord Sun

    Harvord Sun New Member

    That's a very detailed and helpful answer.Appreciate it !