rt_cov.h File Reference

Real-time noise covariance estimation from streaming MEG / EEG data blocks. More...

#include "../dsp_global.h"
#include <fiff/fiff_cov.h>
#include <fiff/fiff_info.h>
#include <QSharedPointer>
#include <QThread>
#include <Eigen/Core>

Include dependency graph for rt_cov.h:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Classes
struct	RTPROCESSINGLIB::RtCovComputeResult
	Bundled output of a real-time covariance computation step containing the covariance matrix and sample count. More...
class	RTPROCESSINGLIB::RtCov
	Controller that manages background covariance matrix estimation from streaming data. More...

Namespaces
namespace	FIFFLIB
	FIFF file I/O, in-memory data structures and high-level readers/writers.
namespace	RTPROCESSINGLIB

Detailed Description

Real-time noise covariance estimation from streaming MEG / EEG data blocks.

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

Author

Christoph Dinh chris.nosp@m.toph.nosp@m..dinh.nosp@m.@mne.nosp@m.-cpp..nosp@m.org

Since

2.1.0

Date

March 2026

RtCov maintains a running unbiased estimate of the channel–channel covariance matrix used by linear inverse operators (MNE, dSPM, sLORETA, beamformers). Every incoming block contributes its centred outer product X⋅Xᵀ to the accumulator together with the per-block sample count; the final covariance is the weighted sum divided by the total number of samples minus one. Computation is offloaded to a worker QThread so the acquisition pipeline never blocks on the dense matrix multiply.

The RTPROCESSINGLIB::RtCovComputeResult bundle carries both the matrix and the sample count, which lets downstream consumers combine partial estimates, apply rank-corrections, or convert to a FIFFLIB::FiffCov for persistence and inverse-operator construction.

Definition in file rt_cov.h.