A-weighting / C-weighting filter
Hello,
I am thinking about implementing an A-weighting and an C-weighting filter into a ADAU1701. In a previous version of a project I implemented these filters in the analog domain. For the next revision I am planing to transfer them to the digital domain. The response will be more accurate and not sensitive for component tolerances.
What do you thing is the best way to implement? Combining various Filters or using FIR or IIR. Maybe anyone out there has done this before and likes to share his experiences. Or is this for some reasons not a good idea to do this.
Thanks for sharing your thoughts
Regards
Bernd
Audio Signal Processors
filter
sigmastudio
a-weithing
SigmaDSP Audio Processors
c-weithing
adau1701
self-boot
Show More
Hello Bernd,
The a-weighting filter is standardized in the ANSI document S1.4, where it is given as a pole-zero specification and can be broken down into an s-domain transfer function.
I imagine that you could take this s-domain transfer function and then perform a z-transform to get it into the z-domain, at which point it would be easy to implement in a digital filter (either one FIR filter or several IIR filters).
SigmaStudio's Stimulus and Probe functions should help you to match your filters to the specification.
I have not had any experience doing this, but perhaps another member can comment if they have more information.
Best Regards,
Brett
Bernd,
The A-weighting filters were originally described as analog filters, so there are some problems with transforming these into simple digital filters, especially as you approach the Nyquist frequency (fs/2). As BrettG mentioned, there are pole/zero specifications available, but it has been very difficult to find information on good digital implementations of these filters (we have looked for these for other applications). I'm sorry that we don't have anything better to offer. If you do find a solution for implementing these on SigmaDSP, though, please let us know here!
Hi JeradL. Are the ITU-R weightings more conveniently specified? I'm about to build an analogue one to assess AD performance but it's not cheap if you don't have the bits to hand (everything * 5 ) and if it can be read within SigmaStudio it eliminates any DA noise.
Cheers
Gordon
Gordon,
I'm not familiar with the ITU-R weightings so I'm not sure how good a digital implementation of them would be. It seems that ITU-R 468 is used more widely outside the US, while A-weighting is preferred for measurements in the US.
I would expect that if these filters were originally specified with an analog implementation that they may have the same potential issues as A-weighting near fs/2.
Being more recent I thought they might be digitally specified, I was hoping you'd know. 'A' weighting is almost universally used for product spec' but for real world assessment ITU-R ARM is probably more useful. Can't get the spec' without buying it ad I don't want to buy it just to find out it's an analogue spec.
Cheers
Gordon
This question has been assumed as answered either offline via email or with a multi-part answer. This question has now been closed out. If you have an inquiry related to this topic please post a new question in the applicable product forum.
Thank you,
EZ Admin
Please log in to reply.