Improve decision boundary between time domain and frequency filtering #273
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Benchmarking reveals that time domain filtering is faster than frequency domain filtering if the unrolled-loop functions are used.
In my testing it is faster to use time domain filtering for filter sizes of length 54 or smaller.
galenlynch
changed the title
galenlynch
added a commit
to galenlynch/DSP.jl
that referenced
this pull request
Apr 19, 2019
|
Ok to merge? |
This reverts commit a59adfb.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
In my testing,
filtis choosing to use frequency domain filtering in instances when time domain filtering would be faster. I have moved the decision boundary between the two algorithms to be better reflect benchmarking results from my laptop. To be honest, I'm not sure how well these benchmarks generalize.In the process, I have also changed how
nfftis chosen for the overlap-save algorithm. The new heuristic is simpler, and seems to follow the cost estimates provided by FFTW.The choice between time and frequency domain filtering is still not perfect. Right now the decision seems to be correct when filtering large amounts of data. However, for smaller amounts of data, it may be advantageous to do time domain filtering to avoid the overhead of frequency domain filtering, even if it would be more expensive per output sample for large amounts of data.