Difference in Data sampling rate for multiple connected sensors

edited June 2016 in C#
Running UWP C# app for windows 10 IoT core on Raspberry pi 2.

Connecting two metawear boards to get gyro data from both and also setting connection interval for both
mbl_mw_settings_set_connection_parameters(cppBoard, 7.5f, 7.5f, 0, 6000);

With ODR_100HZ config for both boards,
getting gyro sampling for one board at rate 90-100 and for 2nd board at rate between 16-18.

Also with ODR_50HZ config for both boards,
getting gyro sampling for one board at rate 47-50 and for 2nd board at rate between 16-18.

Even with config ODR_200HZ getting same result as 90-100 and 16-18.

Any suggestion how to get same sampling rate from both sensor boards at ODR_100HZ?

Comments

  • The Bluetooth LE radio can handle data throughput of up to 100Hz.  If you're attempting to stream from many MetaWears to one pc, you will need to reduce the streaming rate for each board.

    Two of questions for debugging purposes:
    1. Can you stream 100HZ from each board if only one of them is connected at a time?
    2. Are you using the correct "cppBoard" variable for each call to set_conn_parameters?
      • In other words, you aren't setting the connection parameters on the same board twice.
    I'm hoping this is a simple configuration issue and not a bt adapter or Win10 issue with handling multiple btle connections.
  • 1. Yes, I can stream 100Hz from each board if only one of them is connected at a time.
    2. Yes, I am using correct cppBoard for each board to call set_conn_parameters.
  • When I pair one board and then create board, initialize and config, I can stream at around 100Hz from that board.
    And then when I pair second board (and no creation/initialize), the first board rate fall to 16HZ. 
    More I do further step, board creation and initialization then I get stream from 2nd board at 100Hz. But getting from first board at 16Hz at same time.
  • I haven't had a chance to test a multiple mw use case on my Win10 laptop yet nor look further into the issue.  I will let you know if I find anything.
This discussion has been closed.