OpenBCI w/ OpenViBE: Jitter issues?

edited February 2019 in Software
I've been using the OpenViBE acquisition server (along with the designer) with the OpenBCI module. 
[1] mentions that sometimes signals arrive delay which could possibly result in jitter in recorded EEG data wrt. provided stimulations. For ERP based projects, this could significantly degrade the performance of the classifiers.
Does the issue still persist with the new version of OpenViBE? Has anyone tested it after [1] was published? Is there any way to test the latencies in OpenBCI/OpenViBE?


Comments

  • With the release of new OpenBCI GUI, having the networking capabilities, is it possible for OpenBCI GUI to receive stimulation externally through TCP port, and synchronize wrt. time while recording the data in an external file for analyzing it later?
  • wjcroftwjcroft Mount Shasta, CA
    edited February 2019
    Me, hi.

    There are two timing anomalies that can occur with EEG recording: jitter and drift. In the case of Cyton and Ganglion the jitter should be minimal since the sample clock is derived from a stabile oscillator. Even though the radio packet arrival time at the laptop can vary slightly, within a few milliseconds, the sample rate is constant.

    'Drift' is how much the actual sample rate differs (on a long term basis) from the specified rate. Because the Cyton is not completely crystal controlled, it's actual sample rate may be slightly above or below the spec. The default spec is 250 samples/sec. The actual rate for a given Cyton might be 250.2. And another Cyton might be 249.8. Etc. But that rate is constant and does not drift further or jitter.

    If you are providing stimulations that evoke potentials, have you considered using the Cyton Aux channel(s) to record your external trigger events?


    Regards,

    William

  • wjcroftwjcroft Mount Shasta, CA
    re: TCP sync. No, instead use the Aux channel to record your event trigger.

Sign In or Register to comment.