Electromagnetic 3D subsurface imaging with source sparsity for a synthetic object

Abstract

Abstract. This paper concerns electromagnetic 3D subsurface imaging in connection with sparsity of signal sources. We explored an imaging approach that can be implemented in situations that allow obtaining a large amount of data over a surface or a set of orbits but at the same time require sparsity of the signal sources. Characteristic to such a tomography scenario is that it necessitates the inversion technique to be genuinely three-dimensional: For example, slicing is not possible due to the low number of sources. Here, we primarily focused on astrophysical subsurface exploration purposes. As an example target of our numerical experiments we used a synthetic small planetary object containing three inclusions, e.g. voids, of the size of the wavelength. A tetrahedral arrangement of source positions was used, it being the simplest symmetric point configuration in 3D. Our results suggest that somewhat reliable inversion results can be produced within the present a priori assumptions, if the data can be recorded at a specific resolution. This is valuable early-stage knowledge especially for design of future planetary missions in which the payload needs to be minimized, and potentially also for the development of other lightweight subsurface inspection systems.

Iterative ADMM for Inverse FE–BI Problem: A Potential Solution to Radio Tomography of Asteroids

“…Last but not the least, sensitivity analyses on each individual error source [19], [20] should be included in the future work. Following Pursiainen and Kaasalainen [19], error sources can be categorized into measurement error, model error, and inaccuracy in the a priori regularizer.…”

“…Note that, in (20), J is an underdetermined matrix, and the matrix inversion can be efficiently calculated using the matrix inversion lemma [24] as…”

“…Many programs and methods have been proposed for asteroid interior imaging (e.g., [16]- [20]). The most recent one is the ROSSETA mission [21] to comet 67P, which includes an orbit and a lander both equipped with microwave sensors capable of penetrating the comet nuclei.…”

DISCUS – The Deep Interior Scanning CubeSat mission to a rubble pile near-Earth asteroid

“…In order to ensure that a robust tomographic reconstruction can be produced, we use a 1.5-2.0 spatial oversampling rate with respect to the Nyquist criterion (NC) which was suggested for a monochromatic (travel-time) measurement in (Pursiainen and Kaasalainen, 2015). This means recording the radar signal for 5000 to 20 000 measurement positions, when extrapolating the measurement positions to the close proximity of a targeted 260 to 600 m diameter NEA.…”

The Effect of Hardware-Computed Travel Time on Localization Accuracy in the Inversion of Experimental (Acoustic) Waveform Data

“…The value of the parameter β determines the balance between the regularization matrices. A small value for β leads to inverse estimates with low total variation and larger values can be expected to result in well-localized estimates [15], [36], [37]. C. Test setup and scenario 1) Domain: An acoustic setup with one speaker and two microphones was utilized to gather the experimental data.…”