Will X2M200 work whilst on exercise bike?

Discussion in 'Respiration Sensor' started by Andrew Walsh, Oct 18, 2016.

  1. Andrew Walsh

    Andrew Walsh New Member

    I understand you need to be still but I am wondering if it could work whilst on exercise bike if you try and take care to keep torso very still (eg. by locking arms on bars). Has anyone used it like this?
     
  2. Josh Gardiner

    Josh Gardiner New Member

    I have not tried, and I don't have a bike so I can't strap some x2m200's to and look and see what happens. Since the field of view of the sensor is quite large, my guess is it will pick the leg movement which would probably confuse the respiration detection application. So out of the box, I doubt it will work.

    But if your happy to write your own specific filters and detection for the baseband I can't see why it would not be possible. More than one x2m200 might be required but that is just a guess.
     
  3. Andrew Walsh

    Andrew Walsh New Member

    Thanks Josh.
    Yeah I was thinking I would have to have the sensor mounted as close as possible to the chest and angled upwards to avoid the legs or even to use a recumbent bike (lying on back).
    Why is the minimum distance 0.5m? If it were mounted on the bars the distance could be more like 10-20 cm and focused on the chest. Can I write my own code to bypass this limit or is it inherent in the module?
    Also I figure it is impossible to keep torso perfectly still even if trying very hard.
    I am happy to write some Python code to interface to the sensor.
    Two sensors is an idea - Shame I just ordered one to try out ... thanks for the feedback anyway.
     
  4. Josh Gardiner

    Josh Gardiner New Member

    I don't know the actual reason for the minimum distance, But you cant override the minimum anyway; which on the baseband is 0.4m. I have find that 0.4 to 0.439m frame is almost always garbage data in my setups anyway. The actual useful field of view is more like 160 degrees out to 1.8m but SnR does drop off once you are out of the 120 degrees.

    I was thinking more along the lines of spacial clustering, and getting good locks on torso rotations so they can be filtered out. As thats kind of stuff I've tinkering with (spacial clustering), but hey what ever works :)