Talk:Common spatial pattern
| This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | |||||||||||
| |||||||||||
History
[edit]The invention of CSP is difficult to attribute.
Mathematically, it correspond to simultaneous diagonalization of two matrices or generalized eigenvalue decomposition which exists from very long time.
The "invention" could be associated with the first person using it in signal processing (or at least EEG) which I think is Z. Koles [1], but he did not give it a clear name.
The first paper to clearly name it Common Spatial Patten (CSP) seems to be J. Müller-Gerkinga et al. [2] — Preceding unsigned comment added by Samuelboudet (talk • contribs) 14:12, 20 February 2015 (UTC)
References
- ^ Zoltan J. Koles, Michael S. Lazaret and Steven Z. Zhou, "Spatial patterns underlying population differences in the background EEG", Brain topography, Vol. 2 (4) pp. 275-284, 1990
- ^ Johannes Müller-Gerkinga, Gert Pfurtscheller, Henrik Flyvbjerg, "Designing optimal spatial filters for single-trial EEG classification in a movement task", Clinical Neurophysiology, Vol. 110 (5) pp. 787-798, 1999
Simultaneous Diagonalization vs Generalized Eigenvalue Decomposition
[edit]I believe the term simultaneous diagonalization is misleading here, as it describes the special case where both matrices can be diagonalized using a single transform, which requires then to commute (see link above). This is generally not the case for arbitrary covariance matrices. The generalized eigenvalue decomposition instead diagonalizes the difference (in a sense) between the two matrices, and that's what is being used here. 125.237.79.189 (talk) 04:32, 22 October 2025 (UTC)
