Butterworth Highpass Filtering IPP

[Keno]

Hello,

I am attempting to recreate a highpass filter that behaves like the `butter()` and `filt()` functions in Matlab, but I am encountering difficulties. The lowpass filter, which is implemented similarly, is working as expected.

I have included the implemented functions for both the highpass (HP) and lowpass (LP) filters, as well as some example data, in the attachments. I've also provided a Matlab script to verify the results. I have removed any debug and error handling code for readability.

The current IPP version in use is 8.0.1 (major, minor, major build), although the same issue persisted with version 2022.9.

My problem seems to be correlating with `rFreq`. Decreasing `rFreq` below 0.01 creates unexpected behavior. Coming closer to 0 increases the error, for which I have included 2 screenshots.

Is there a workaround to still be able to use an `rFreq` of 1e-3 or below since it is inside the required rFreq of 0 to 0.5?

[VarshaS_Intel]

Hi,

Thanks for posting in Intel Communities and Thanks for providing the files.

From the sample reproducer code provided by you(IPP.cpp), when we tried to compile using the ICPX compiler on Ubuntu using the latest version with the '-qipp' option we got some errors. T

here are a missing number of parameters for IppStatus ippsIIRGenLowpass_64f and

IppStatus ippsIIRGenHighpass_64f, could you please verify and provide us with the modified code?

https://www.intel.com/content/www/us/en/docs/ipp/developer-reference/2021-9/iirgenlowpass-iirgenhighpass.html

Also, could you please provide us with the complete steps you followed to run on MATLAB?

Thanks & Regards,

Varsha

[VarshaS_Intel]

Hi,

We have not heard back from you. Could you please provide us with the details and provide us with an update on your issue?

Thanks & Regards,

Varsha

[VarshaS_Intel]

Hi,

We have not heard back from you. Could you please provide us with an update on your issue?

Thanks & Regards,

Varsha