Solar Gravitational Lens Framework Achieves SSIM 0.993 Reconstructions for White Dwarf Surfaces at 10 pc
Preprint quantifies scalar inverse conditioning for non-exoplanet SGL targets with SSIM metrics above 0.9 under stated assumptions. Dominant requirements shift from photon collection to PSF knowledge and coronal subtraction. Primary limitation is unvalidated transfer-function model; strengthening requires dedicated solar multipole measurement campaign.
The June 2026 arXiv submission separates the full vector measurement model from simplified convolution kernels used in prior exoplanet studies, propagating four analytic scenes through realistic kernel mismatch, coronal background, and calibration floors. Reconstructions yield SSIM 0.993 for a solar analog, 0.918 for an M87*-scale ring, and 0.923 for a 0.1 AU protoplanetary subfield, with diagnostics highlighting support leakage and information-floor sensitivity rather than raw photon counts. This extends earlier SGL concepts focused on exo-Earth reflected light by demonstrating that self-luminous compact targets are limited instead by ring extraction, detector dynamic range, and focal-line metrology cadence. The paper correctly flags transfer-function characterization as the enabling step, yet understates the required multi-year solar monitoring campaign needed to separate time-variable plasma from static multipoles. Next steps center on ground-based or near-Earth solar coronagraph campaigns to bound the extended-Sun response before any spacecraft architecture study advances.
Turyshev: Dedicated solar multipole monitoring campaign initiated by 2028 reaches 5 nanoradian plasma subtraction accuracy within 36 months or SGL reconstruction SSIM drops below 0.85 for extended sources.
Sources (2)
- [1]Primary Source(https://arxiv.org/abs/2606.18300)
- [2]Supporting Source(https://arxiv.org/abs/2204.04899)