Skip to content

Subspace Methods

Subspace methods (including MUSIC variants) exploit the orthogonality between signal and noise subspaces to localize sources. They are particularly effective for localizing a small number of focal sources.

This category contains 12 solvers.

Solvers

Full Name Solver ID Description
Adaptive Alternating Projections AAP Adaptive variant of alternating-projection source localization with flexible-extent patch estimation.
Alternating Projections AP Alternating-projection source localization on the signal subspace, extended here with flexible-extent (FLEX-AP) patch...
Extended Signal Subspace MUSIC ExSo-MUSIC ExSo-MUSIC extended-source localization via pseudo-disk scanning. Supports 2nd-order (q=1, covariance-based, default...
Exhaustive Subspace Maximum-Likelihood ExhaustiveML Source localization via exhaustive maximum-likelihood subset search (k<=3) with beam search extension (k>3) and BIC m...
FLEX Greedy Maximum-Likelihood FLEX-GreedyML Source localization via multi-start greedy ML on an extended dictionary of single-dipole and patch leadfields with BI...
Flexible Extent RAP/TRAP-MUSIC FLEX-MUSIC Recursive MUSIC variant with flexible extent estimation. Iteratively selects sources (RAP/TRAP-MUSIC) while allowing ...
FLEX-MUSIC 2 FLEX-MUSIC2 Experimental FLEX-MUSIC variant with iterative neighborhood expansion and optional distance-weighted patch growth.
Generalized Iterative Subspace Solver GI Generalized iterative framework covering RAP-MUSIC, alternating projections (AP), and signal subspace matching (SSM),...
Greedy Maximum-Likelihood GreedyML Source localization via multi-start greedy forward selection with ML objective, BIC model order selection, and coordi...
Greedy Maximum-Likelihood GreedyML Source localization via multi-start greedy forward selection with ML objective, BIC model order selection, and coordi...
Multiple Signal Classification MUSIC Subspace-based dipole localization using the MUSIC pseudospectrum. Estimates a signal subspace from the data covarian...
Signal Subspace Matching SSM Signal subspace matching (SSM) for robust source localization under model mismatch, extended with flexible-extent pat...