A Fast MART Method with Voxel Filtering for GEO SAR-Based Computerized Ionospheric Tomography

Ionospheric monitoring is crucial for space environment assessment and the success of space mission. Traditional computerized ionospheric tomography (CIT) based on Global Navigation Satellite System (GNSS) are constrained ground receivers, while CIT based on geosynchronous synthetic aperture radar (GEO SAR) offers an ionospheric tomography solution that does not rely on ground receivers. This approach leverages the abundant Permanent Scatterer (PS) points in the SAR image for electron density reconstruction, providing strong adaptability to various environments. However, due to the large number of PS points in the large coverage area of GEO SAR and the abundance of total electron density (TEC) data, the conventional multiplicative algebraic reconstruction technique (MART) is inefficient in processing iteration. To this end, this paper proposes a fast MART method based on Voxel Filtering, and verifies the effectiveness of the proposed method through a tomographic experiment. The results show that the proposed method ensures reliable tomographic results, effectively improves the iteration speed, and significantly improves the computational efficiency, which provides strong support for GEO SAR-based CIT with huge TEC data.