Skip to main content

EegReference

Namespace: RTPROCESSINGLIB  ·  Library: DSP Library

Module

This page documents a header-level module — a collection of free functions that share an algorithmic topic. There is no enclosing C++ class; the functions live directly in the library namespace.

Python equivalent

mne.set_eeg_reference in MNE-Python.

#include <dsp/eeg_reference.h>

SPDX-License-Identifier: BSD-3-Clause Copyright (c) 2026 MNE-CPP Authors.

Every EEG measurement is implicitly a potential difference between an active electrode and a reference; changing that reference is therefore a linear operation on the data matrix. The functions in this header build and apply the most widely used reference operators: subtraction of a single electrode (or the average of a few), the common-average reference (CAR), and — where a forward model is supplied — the reference electrode standardisation technique (REST) which projects the data to the equivalent reference at infinity.

Re-referencing is implemented as a single matrix multiplication so that channel covariances and inverse operators remain valid after re- referencing, provided the same operator is also applied to the noise covariance.

src/examples/ex_dsp_artifacts/main.cpp
MatrixXd averageRef = data;
setEegReference(averageRef, info); // average of the EEG channels; EOG/ECG rows untouched

Functions​

setEegReference(data, info, refChannels, projection)​

void setEegReference(Eigen::MatrixXd & data, [FiffInfo](/docs/api/fiff/fiff-info) & info, const QStringList & refChannels, bool projection);

Re-reference EEG channels to average, specific channel(s), or REST.

Only modifies rows of good EEG channels (kind == FIFFV_EEG_CH, not in info.bads), as mne.set_eeg_reference does. Bad and non-EEG channels are left untouched.

Parameters:

  • data : Eigen::MatrixXd & Data matrix (n_channels x n_times). Modified in-place.

  • info : FiffInfo & Measurement info (identifies EEG channels; receives the projector).

  • refChannels : const QStringList & Reference channel name(s). - Empty or "average" -> average reference (mean of all good EEG channels). - Single channel name -> re-reference to that channel. - Multiple channel names -> re-reference to mean of those channels.

  • projection : bool Average reference only: append an inactive average-reference projector over the good EEG channels to info.projs (mne make_eeg_average_ref_proj) and leave the data unchanged. Nothing is added if one is already present.

addReferenceChannels(data, info, chNames)​

void addReferenceChannels(Eigen::MatrixXd & data, [FiffInfo](/docs/api/fiff/fiff-info) & info, const QStringList & chNames);

Add reference channel(s) back as zero-filled rows.

When EEG data was recorded with a physical reference electrode, that channel is implicit (all zeros after re-referencing). This function adds it back explicitly so it can participate in average re-referencing or source localization.

Parameters:

  • data : Eigen::MatrixXd & Data matrix. Rows are added at the end.

  • info : FiffInfo & Measurement info. New EEG channels are appended.

  • chNames : const QStringList & Names of the reference channel(s) to add.

setBipolarReference(data, info, anodes, cathodes, dropOriginals)​

void setBipolarReference(Eigen::MatrixXd & data, [FiffInfo](/docs/api/fiff/fiff-info) & info, const QStringList & anodes, const QStringList & cathodes, bool dropOriginals);

Create bipolar derivations from EEG channels.

Creates new bipolar channels by subtracting cathode from anode for each pair. The original channels are replaced with the bipolar derivations.

Parameters:

  • data : Eigen::MatrixXd & Data matrix. Replaced with bipolar derivation rows.

  • info : FiffInfo & Measurement info. Updated with new bipolar channel info.

  • anodes : const QStringList & List of anode channel names.

  • cathodes : const QStringList & List of cathode channel names (same length as anodes).

  • dropOriginals : bool If true, remove the original anode/cathode channels from the output (default: true).

Example​

Authors of this file​