Public

Radio Aperture Synthesis — Public Tier

(5 scenes)

Full-access development tier with all data visible.

What you get

Measurements (y), ideal forward operator (H), spec ranges, ground truth (x_true), and true mismatch spec.

How to use

Load HDF5 → compare reconstruction vs x_true → check consistency → iterate.

What to submit

Reconstructed signals (x_hat) and corrected spec as HDF5.

Parameter Specifications

✓

True spec visible — use these exact values for Scenario III oracle reconstruction.

Parameter Spec Range True Value Unit
antenna_gain_error 0.99 – 1.02 1.005 -
phase_calibration_error -1.0 – 2.0 0.5 deg
bandpass_slope -0.002 – 0.004 0.001 1/MHz
pointing_offset -1.0 – 2.0 0.5 arcsec

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

14.72 dB

SSIM 0.3946

Scenario II (Mismatch)

13.88 dB

SSIM 0.3702

Scenario III (Oracle)

17.16 dB

SSIM 0.5513

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 14.92 0.4140 13.09 0.3771 16.33 0.5715
scene_01 14.62 0.3920 14.01 0.3722 17.29 0.5458
scene_02 14.51 0.3803 14.30 0.3661 17.41 0.5404
scene_03 14.81 0.3923 14.10 0.3653 17.61 0.5476

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 PRIMO + gradient 0.729 29.42 0.908 0.86 ✓ Certified Medeiros et al., ApJL 2023
2 R2D2 + gradient 0.725 28.37 0.888 0.93 ✓ Certified Aghabiglou et al., ApJS 2024
3 AIRI + gradient 0.627 24.23 0.777 0.87 ✓ Certified Terris et al., MNRAS 2022
4 CLEAN + gradient 0.561 21.24 0.657 0.93 ✓ Certified Hogbom, A&AS 1974

Visible Data Fields

y H_ideal spec_ranges x_true true_spec

Dataset

Format: HDF5
Scenes: 5

Scoring Formula

0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

PSNR: 40% SSIM: 40% Consistency: 20%
Back to Radio Aperture Synthesis