Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations KootK on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

smoothen out noisy acceleration curve

Status
Not open for further replies.

Kimpan

Mechanical
May 30, 2008
39
I have a very noicy rotational acceleration curve from an impact to a dummy head. The impact has a duration of ~10ms and I wish to smooth it out without compromising the data. What type of filter is suitable for this?
I have been using a "sliding average" and butterworth filter but not sure which, if any, is good?
In the picture I have applied a sliding average that calculates the average of 20 points. The blue line has been filtered again with sliding average of 20 points.

slidingaverage.jpg
 
Replies continue below

Recommended for you

Blue curve looks fine to me. We normally call it a moving average. The other obvious option is to fourier transform the entire signal, zero the high frequency ringing tone, and then IFT it back into the time domain.

However for crash data I think you'll find the regs specify what sort of filtering to use.

Cheers

Greg Locock

SIG:please see FAQ731-376 for tips on how to make the best use of Eng-Tips.
 
" Although Savitzky-Golay filters are more effective at preserving the pertinent high frequency components of the signal, they are less successful than standard averaging FIR filters at rejecting noise when noise levels are particularly high."

Given that he has an S/N ratio of 0 dB I think I'd class that as high noise levels.

Cheers

Greg Locock

SIG:please see FAQ731-376 for tips on how to make the best use of Eng-Tips.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor