Frequency Filtering Techniques

  • Boris N. Khoromskij
  • Gabriel Wittum
Part of the Lecture Notes in Computational Science and Engineering book series (LNCSE, volume 36)


In this chapter, we outline the method of frequency filtering decomposition (FFD) proposed in [186] for the construction of factorized preconditioners for the systems of linear equations arising in usual FE conforming approximations for elliptic problems. Here we also consider possible extensions of this technique to approximate the Schur complements on the interface [113]. Further generalizations of this approach by using tangential FFD and adaptive filtering were developed in [183] and in [184]. In the case of symmetric positive definite block-tridiagonal matrices with constant block entries, a new interpretation for the TFFD techniques (which also simplifies a theoretical analysis) was considered in [38].


Probe Vector Test Vector Slave Node Convergence History Tridiagonal Matrice 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Boris N. Khoromskij
    • 1
  • Gabriel Wittum
    • 2
  1. 1.Max-Planck-Institut für Mathematik in den NaturwissenschaftenLeipzigGermany
  2. 2.Interdisziplinäres Zentrum für Wissenschaftliches Rechnen (IWR)Universität HeidelbergHeidelbergGermany

Personalised recommendations