Signal Averaging
Signal averaging increases the strength of a repeating signal embedded in a recording relative to random noise in the recording. It works by averaging the waveforms of the repeating signal, which requires a fixed time datum indicating the time of occurrence of each signal within the recording. The signal-to-noise ratio improves in proportion to the square root of the number of signals in the average.
In Dataview, signal averaging is a facility in the Scope view. The fixed time datum is the start time of the events used to trigger the sweeps that will be averaged.
There are typically two related but slightly different aims in using signal averaging.
- Signal averaging can be used to "purify" a waveform by removing random variability that contaminates the individual instances of the waveform. An example is given in the Average section of the Evoked Potentials tutorial.
- Signal averaging can be used to reveal waveforms that are phase-locked to events, but which are invisible due to contaminating noise. An example is given here.