This paper considers the reconstruction problem in Acousto-ElectricalTomography, i.e., the problem of estimating a spatially varying conductivity ina bounded domain from measurements of the internal power densities resultingfrom different prescribed boundary conditions. Particular emphasis is placed onthe limited angle scenario, in which the boundary conditions are supported onlyon a part of the boundary. The reconstruction problem is formulated as anoptimization problem in a Hilbert space setting and solved using Landweberiteration. The resulting algorithm is implemented numerically in two spatialdimensions and tested on simulated data. The results quantify the intuitionthat features close to the measurement boundary are stably reconstructed andfeatures further away are less well reconstructed. Finally, the ill-posednessof the limited angle problem is quantified numerically using the singular valuedecomposition of the corresponding linearized problem.
展开▼
