Proposal of a criterion to select the layout of bolometric diagnostics for robust plasma radiation reconstruction in fusion devices

• Maximum likelihood tomography approach helps to define acquisition geometry. • Accuracy of a detection system can be evaluated in terms of uncertainty and reconstruction error. • Two indicators are defined for synthetically analysed an acquisition geometry. Nuclear fusion stands out as a promising energy source, with numerous experimental reactors currently operational and larger ones in various stages of construction. As reactor dimensions increase, the precise measurement of the plasma total emitted radiation becomes increasingly crucial. This quantity not only helps in evaluating tokamak power balances but also enhances the understanding of the underlying physics in different processes. To optimize the performance of large nuclear fusion reactors, for example, various radiative scenarios have been proposed. These involve injecting gas into the plasma to induce radiation emission, thereby reducing the heat load on the reactor plasma facing components. Additionally, analysing the radiation pattern within the plasma provides valuable insights into impurity transport. A key technique for extracting information on plasma radiation is tomography reconstruction based on bolometric measurements. Addressing the resulting inverse problem, an effective approach involves utilizing maximum likelihood methods, which not only facilitates the reconstruction but also offers an estimation of the reconstruction errors. This study aims to investigate the impact of the bolometer diagnostic geometry on the reconstruction performance, a critical aspect for designing bolometers for future fusion machines. Due to limited space availability, bolometer systems present a restricted number of projections, rendering the optimization of the geometry fundamental for minimizing the sources of error. The analysis performed consists of reconstructing synthetic images in different geometries, varying angles of view and the number of lines of sight. Finally, an optimal criterion is defined to guide the selection of the geometry, ensuring robust reconstruction with minimal susceptibility to measurement errors.