How to detect peaks in a microcontroller?

Question

In the context of a microcontroller, I have much less memory and far less processing power. This means, I can only buffer a reasonable window of samples (e.g. 16 or 32), and may not be able to use more advanced curve-fitting options, such as this one.

This plot shows the raw data along with filtered output. The filter in this trial is a leaky-integrator, with alpha = 8.

enter image description here

How can I reliably detect that first peak (and trough) to enable the accumulation of the area shown?

A cleaner trial of raw data (sampled every 50ms):

Fat32 · Accepted Answer · 2015-01-19 21:27:29Z

2

It seems to me, more like a spike detection than a peak. For the simplest spike detector, just get the first difference of the input samples and test for a threshold. Use multiple samples and adjust your test to have some memory (as much as you can) so that noisy results be eliminated and you would get a more robust detection.

answered Jan 19, 2015 at 21:27

Fat32

28.4k3 gold badges24 silver badges51 bronze badges

Add a comment |

Stack Exchange Network

How to detect peaks in a microcontroller?

1 Answer 1

Your Answer

Not the answer you're looking for? Browse other questions tagged
peak-detection
or ask your own question.

Hot Network Questions

How to detect peaks in a microcontroller?

1 Answer 1

Your Answer

Sign up or log in

Post as a guest

Not the answer you're looking for? Browse other questions tagged peak-detection or ask your own question.

Related

Hot Network Questions

Not the answer you're looking for? Browse other questions tagged
peak-detection
or ask your own question.