GPU-Torre and GPU-Torre-3D are open Matlab-based toolboxes for numerical investigation of mathematical non-linear tomographic radar inverse imaging problems in 2D and 3D. They aim at enabling fast and robust inversion of sparse full-wave tomography data. The forward modelling approach applies a graphics processing unit (GPU) accelerated finite element time-domain (FETD) method and Born approximation (BA) to account for complex wave scattering inside the computational domain. In the inversion stage, a GPU-accelerated multigrid-FETD deconvolution routine is utilized to enhance computational performance. GPU-Torre frontend provides a graphical interface to manage the parameters of the toolbox and to create script pipelines for advanced modelling and reconstruction tasks in full wavefield simulation, tomographic reconstruction, signal preprocessing, 3D model creation, as well as in design and manufacturing (3D printing) of experimental analogue objects. GPU-Torre and GPU-Torre-3D packages have been used in the following studies.
Sorsa, L. I., Pursiainen, S., & Eyraud, C. (2021). Analysis of full microwave propagation and backpropagation for a complex asteroid analogue via single-point quasi-monostatic data. Astronomy & Astrophysics, 645, A73.
Sorsa, L. I., Eyraud, C., Hérique, A., Takala, M., Pursiainen, S., & Geffrin, J. M. (2021). Complex-structured 3D-printed wireframes as asteroid analogues for tomographic microwave radar measurements. Materials & Design, 198, 109364. ISO 690
Eyraud, C., Sorsa, L. I., Geffrin, J. M., Takala, M., Henry, G., & Pursiainen, S. (2020). Full wavefield simulation versus measurement of microwave scattering by a complex 3D-printed asteroid analogue. Astronomy & Astrophysics, 643, A68.
Sorsa, L. I., Takala, M., Eyraud, C., & Pursiainen, S. (2020). A time-domain multigrid solver with higher-order born approximation for full-wave radar tomography of a complex-shaped target. IEEE Transactions on Computational Imaging, 6, 579-590.
Sorsa, L. I., Takala, M., Bambach, P., Deller, J., Vilenius, E., & Pursiainen, S. (2019). Bistatic full-wave radar tomography detects deep interior voids, cracks, and boulders in a rubble-pile asteroid model. The Astrophysical Journal, 872(1), 44.
Takala, M., Us, D., & Pursiainen, S. (2018). Multigrid-based inversion for volumetric radar imaging with asteroid interior reconstruction as a potential application. IEEE Transactions on Computational Imaging, 4(2), 228-240.
Pursiainen, S., & Kaasalainen, M. (2016). Orbiter-to-orbiter tomography: a potential approach for small solar system bodies. IEEE Transactions on Aerospace and Electronic Systems, 52(6), 2747-2759.