In this work, new adaptive numerical wavelet algorithms for the solution of elliptic operator equations posed in a bounded domain or on a closed manifold are developed. To circumvent the complicated construction of a wavelet Riesz basis for the solution space, we work with the weaker concept of wavelet frames.
Using an overlapping domain decomposition technique, suitable frames can easily be constructed and implemented. In a first step, we show that classical results on the convergence rates of best N-term approximations of the solution with respect to wavelet Riesz bases essentially carry over to the considered class of wavelet frames. We then develop an adaptive method based on a steepest descent iteration for the frame coordinate representation of the elliptic equation, and, most importantly, we develop algorithms based on multiplicative and additive Schwarz overlapping domain decomposition methods. We prove that our adaptive schemes are of asymptotically optimal complexity, in the sense that they realize the same convergence rate as the sequence of best N-term frame approximations of the solution. Moreover, using special numerical quadrature rules for the computation of the frame representation of the elliptic operator, the overall computational cost stays proportional to the number of wavelets selected by the algorithms.
The results of a series of numerical tests for non-trivial one- and two-dimensional Poisson and biharmonic model problems confirm our theoretical findings and particularly demonstrate the efficiency of the domain decomposition approach. A comparison with a standard adaptive finite element solver shows that our multiplicative Schwarz method potentially generates significantly sparser approximations. In addition, a parallel implementation of the new adaptive additive Schwarz wavelet solver is developed and tested.
Book Description
Product Details
Would you like to update product info or give feedback on images?
|
Customer Reviews
|
There are no customer reviews yet.
|
|||
|
Video reviews
|
Tag this product(What's this?)Think of a tag as a keyword or label you consider is strongly related to this product.
Tags will help all customers organize and find favorite items. |
Sell a Digital Version of This Book in the Kindle Store
If you are a publisher or author and hold the digital rights to a book, you can sell a digital version of it in our Kindle Store.
Learn more
