1: \begin{abstract}

2: Several novel imaging and non-destructive testing technologies are based on reconstructing the spatially dependent coefficient in an elliptic partial differential equation

3: from measurements of its solution(s). In practical applications, the unknown coefficient is often assumed to be piecewise constant on a given pixel partition (corresponding to the 

4: desired resolution), and only finitely many measurement can be made. This leads to the problem of inverting a finite-dimensional non-linear forward operator $\FF:\ \mathcal D(\FF)\subseteq \R^n\to \R^m$, where evaluating $\FF$ requires one or several PDE solutions. 

5: 

6: Numerical inversion methods require the implementation of this forward operator and its Jacobian. We show how to efficiently implement both using a standard FEM package and prove convergence of the FEM approximations against their true-solution counterparts. We present simple example codes for Comsol with the Matlab Livelink package, and numerically demonstrate the challenges that arise from 

7: non-uniqueness, non-linearity and instability issues. We also discuss monotonicity and convexity properties of the forward operator that arise for symmetric measurement settings. 

8: 

9: This text assumes the reader to have a basic knowledge on Finite Element Methods, including the variational formulation of elliptic PDEs, 

10: the Lax-Milgram-theorem, and the C\'ea-Lemma. Section \ref{sect:true_solution} also assumes that the reader is familiar with the concept of Fr\'echet differentiability.

11: %

12: \keywords{Finite Element Methods \and Inverse Problems \and Finitely many measurements \and Piecewise-constant coefficient}

13: % \PACS{PACS code1 \and PACS code2 \and more}

14: % \subclass{MSC code1 \and MSC code2 \and more}

15: \end{abstract}

16: