1: \begin{abstract}

2: This paper provides a mathematical framework for Fisher information analysis for inverse problems 

3: based on Gaussian noise on infinite-dimensional Hilbert space. The covariance operator for the

4: Gaussian noise is assumed to be trace class, and the Jacobian of the forward operator Hilbert-Schmidt.

5: We show that the appropriate space for defining the Fisher information is given by the Cameron-Martin space.

6: This is mainly because the range space of the covariance operator always is strictly smaller than the Hilbert space.

7: For the Fisher information to be well-defined, it is furthermore required that the range space of the Jacobian is contained in the Cameron-Martin space.

8: In order for this condition to hold and for the Fisher information to be trace class, a sufficient condition is formulated

9: based on the singular values of the Jacobian as well as of the eigenvalues of the covariance operator, together with some regularity assumptions 

10: regarding their relative rate of convergence. 

11: An explicit example is given regarding an electromagnetic inverse source problem with ``external'' spherically isotropic noise,

12: as well as ``internal'' additive uncorrelated noise.

13: \end{abstract}

14: