Mat-1.3656 Seminar on numerical analysis and computational science

Monday, March 19, 2012, room U322 at 14.15, Eirola & Stenberg

Antti Huhtala, A! Department of Mathematics and Systems Analysis
Numerical convergence of regularized Poisson inverse source problems

When solving inverse source problems numerically, it is of interest to know the magnitude of the error caused by the discretization steps. This can then be compared to other sources of uncertainty and error in the problem, which in turn can be used in choosing a suitable resolution for the discretization.

We have derived a-priori convergence estimates for regularized Poisson inverse problems, in the case where the measurement data is finite dimensional. We present the convergence analysis for a prototype problem and show examples in the case of full regularity and limited regularity.