Using autocorrelation model for breath detection

Posted on Feb 14, 2018 | 0 comments

We have noticed markable improvements in breath detection in accelerometer signal when we tested autocorrelation model. It’s similar to autoregressive (AR) model, but in this article we call this model autocorrelation.

The autocorrelation is defined as:

Screen Shot 2018-02-14 at 9.17.52 PM


Where N is the total window size, N win is the AR window size, i is the AR shift, and Ncor is the number of steps.

In tests we have noticed that autocorrelation works a lot better than other methods, including FFT., though it requires significant calibration and tuning. See the graphs with different breathing states below.

Screen Shot 2018-02-14 at 9.25.14 PM

Screen Shot 2018-02-14 at 9.25.22 PM

Screen Shot 2018-02-14 at 9.25.35 PM


The final plot depicted below is the result of running breath detection using autocorrelation on live production data collected from one of our customers baby,

The three colored axis show row accelerometer signal and the dotted line shows detected breath in breaths per minute. The bottom graph shows state enumeration with different movement classification states detected. A thing to note that no-breath detection is corresponds to flag=-2, which is an alarming condition to alert parents. In the sample below, the alarm have not been detected.

Screen Shot 2018-02-14 at 9.04.59 PM

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>