Raw data timestamps

[cnolan]

Hi folks, I've got one of the original Muse devices, and I'm just trying to understand what is saved in the raw protocol buffer format. With every event (including every EEG sample) there is a timestamp, but the timestamps suggest a very uneven sampling rate. Am I correct in assuming therefore that the timestamps are actually just the time the packet is saved, rather than the actual sample time? If so, is there a known constant sampling rate for the device?

Cheers,
Chris.

[James]

Correct. Timestamps are assigned in the phone after being processed through the Bluetooth packet buffer, so they tend to be bunched up.
This is a known issue that only Interaxon can fix with a firmware update unfortunately.
The 2014 MU-01 model Muse raw eeg data rate is 220Hz, all the other Muse's run at 256Hz.

[cnolan]

Perfect, thanks for the information James, and the timely response.

[bgodde]

James, great tool - thank you very much. Love to use it for teaching psych students in EEG.

Correct. Timestamps are assigned in the phone after being processed through the Bluetooth packet buffer, so they tend to be bunched up.
This is a known issue that only Interaxon can fix with a firmware update unfortunately.
The 2014 MU-01 model Muse raw eeg data rate is 220Hz, all the other Muse's run at 256Hz.

What does that mean for the OSD data stream? Do I get there also each single data package with 256 Hz sampling time and just ignore the time of receiving it?

[James]

The timestamps in the CSV and OSC data will be the same. They're both as accurate as they can be, given the Bluetooth data buffering.