Improve high-rate PPG acquisition - #271
mkuettner97 wants to merge 2 commits into
Conversation
|
Build output available: |
f2380b1 to
7fe2cab
Compare
✅ Unit tests passed2 passed, 0 failed/error, 0 skipped — view workflow run
Download the |
|
Build output available: |
Compiler warningsThe extended-warning build completed successfully. Application compiler warningsNone. |
CodeChecker static analysis✅ No non-style issues found. |
|
At 1024 Hz, red and ambient frequently contain zeros, although all four channels should be available. Recordings at 2048 Hz looked good with red and IR, and 4096 Hz looked good with only IR. |
|
Zero values were caused by PPG samples being split across two FIFO reads. The fix keeps the unfinished sample and completes it on the next read, so only complete samples are returned. Tested at 1024 Hz and the zeros are gone. |
|
Timestamps are irregular because each FIFO batch is timestamped when the firmware reads it, which introduces scheduling jitter. This was not introduced by this PR though |
|
Build output available: |
2 participants
The PPG did not even reach the higher PPG sampling rates before, since the integration time was set to long