multitaper_psd.h File Reference

Thomson multitaper power-spectral-density estimator. More...

#include "dsp_global.h"
#include <Eigen/Core>

Include dependency graph for multitaper_psd.h:

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

Go to the source code of this file.

Classes
struct	UTILSLIB::MultitaperPsdResult
	Result of a multitaper PSD computation. More...
class	UTILSLIB::MultitaperPsd
	Multitaper power spectral density estimator using DPSS (Slepian) tapers. More...

Namespaces
namespace	UTILSLIB
	Shared utilities (I/O helpers, spectral analysis, layout management, warp algorithms).

Detailed Description

Thomson multitaper power-spectral-density estimator.

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

2.2.0

Date

April 2026

The multitaper estimator (Thomson, 1982) computes K independent eigenspectra by tapering the input with the first K Discrete Prolate Spheroidal Sequences (Dpss) of time-half-bandwidth product NW, then averages them. Because each Slepian taper concentrates almost all of its energy inside the design bandwidth, the resulting estimate has dramatically lower spectral leakage than a single-tapered periodogram while keeping the bias–variance trade-off explicit: variance shrinks roughly as 1/K, and frequency resolution degrades to ±W cycles per sample.

Adaptive eigenvalue weighting (Thomson's adaptive scheme) is supported so tapers whose concentration λ_k is degraded by signal energy outside the design band contribute less to the final spectrum. The estimator mirrors MNE-Python's mne.time_frequency.psd_array_multitaper so MEG / EEG analysis results stay reproducible across the two stacks.

Definition in file multitaper_psd.h.