Nathan Golovich, Lawrence Livermore National Laboratory; Alex Geringer-Sameth, Lawrence Livermore National Laboratory; Robert Armstrong, Lawrence Livermore National Laboratory; Willem de Vries, Lawrence Livermore National Laboratory; Benjamin Bahney, Lawrence Livermore National Laboratory
https://doi.org/10.64861/UTTV8276
Keywords: SDA constellation, LEO to GEO, wide field surveillance
Abstract:
We present a simulated space domain awareness (SDA) architecture for persistent optical monitoring of the geosynchronous (GEO) regime using wide-field sensors deployed in low Earth orbit (LEO). The concept is built around a compact, monolithic telescope developed at Lawrence Livermore National Laboratory (LLNL), featuring a 12.5 cm aperture, f/3.2 optical design, and a 17 square degree instantaneous field of view (FOV). The objective is to demonstrate, through end-to-end simulation, that a modest-aperture optical payload in sun-synchronous orbit (SSO) can support both wide-area GEO search and rapid orbit refinement via binocular ranging.
The simulation framework models a representative GEO-like population with inclinations extending to 30 degrees and a size distribution from 12u cubesats to large (>1 meter) spacecraft. From LEO, this produces a thick belt rather than a narrow equatorial ring, spanning a substantial sky volume. To first-order, the accessible GEO region may be approximated as ~2π steradians as viewed from LEO. Applying a 45-degree solar exclusion constraint (appropriate for an optical payload) removes approximately one quarter of the belt at any given time, leaving roughly 3π/2 steradians available for search. When surveilled with a 17 deg² FOV, this corresponds to on the order of ~900 pointings for single-epoch coverage of the instantaneous accessible GEO volume, providing a practical basis for scan strategy development and performance assessment with a small constellation of LEO observers.
The simulation incorporates solar phase angle effects, target brightness modeling, and time-dependent line-of-sight geometry between LEO sensors and GEO targets and the Sun. Target brightness is parameterized as a function of size, albedo, and phase angle, enabling realistic detection probability estimates across the GEO domain. Payload sensitivity is treated as a function of aperture, focal ratio, background radiance, and exposure time, producing limiting magnitude and signal-to-noise ratio (SNR) predictions for each pointing. The apparent angular rate of GEO objects as observed from LEO is also included, capturing relative motion and integration-time constraints that directly affect achievable depth.
Two operational modes are investigated to demonstrate complementary capabilities.
In the first mode, two LEO sensors are closely phased within the same orbital plane to enable near-simultaneous observation of common GEO targets. This configuration provides an optical “binocular” baseline for angular parallax. By exploiting measurements from a known LEO separation, instantaneous three-dimensional position estimates can be derived, reducing reliance on purely angle-only initial orbit determination (IOD) and improving state estimation precision by orders of magnitude. The simulation quantifies orbit refinement performance as a function of baseline geometry and centroiding precision, demonstrating accelerated covariance reduction and improved convergence for newly detected or maneuvering objects. This binocular configuration emphasizes rapid IOD refinement and responsive tracking.
In the second mode, the two sensors are phased approximately 180 degrees apart within the same sun-synchronous plane. This geometry maximizes longitudinal diversity and revisit frequency, effectively reducing latency to first detection and enhancing persistence. In this configuration, the constellation operates as a wide-area survey system, tiling the ~3π/2 steradian accessible region through sequential pointings. Performance metrics include time to full coverage of the observable GEO volume, revisit interval, and detection completeness fraction for the simulated population.
A central component of the study is the exploration of exposure-time during a full GEO scan. Because ~900 pointings are required for single-epoch coverage of the accessible belt, exposure time directly determines the trade between scan cadence and sensitivity. The simulation evaluates two dwell-time regimes, including a rapid, shallow survey that prioritizes refresh rate to deeper integrations that enhance faint-object completeness. This enables quantitative assessment of how sensitivity, completeness, and time-to-coverage evolve under realistic geometric constraints.
To extend effective integration time without requiring long single-frame exposures, the architecture incorporates frame stacking with rate tracking. Spacecraft platforms often cannot support exposures significantly longer than 1-second due to stability, jitter, and operational constraints. Instead, the simulation assumes a sequence of short exposures that are co-registered along predicted GEO angular velocities and combined within a stacked data cube. By shifting and adding frames along candidate rate vectors, the system synthetically increases integration time while mitigating trailing losses. This approach preserves survey cadence while enabling deeper sensitivity, particularly for faint or high-phase-angle targets. Stacking and rate-tracking operations are assumed to be executed onboard using emerging edge-computing architectures, enabling real-time detection and data reduction while managing downlink bandwidth. The simulation evaluates how stacking depth, velocity-grid resolution, and onboard processing limits influence limiting magnitude and overall survey completeness during a full-belt scan.
Results are expressed in terms of cumulative detection fraction, time to catalog refresh, orbit-determination performance, and sensitivity as a function of phase angle. Even with a 45-degree solar exclusion limiting instantaneous access to roughly three-quarters of the GEO belt, the combination of wide-field optics, LEO orbital motion, geometric diversity, and onboard stacked integration enables frequent coverage of the accessible volume. In binocular mode, geometric parallax accelerates orbit refinement, while in distributed mode, the constellation delivers persistent wide-area surveillance.
This simulated architecture demonstrates that wide-field optical sensors in LEO can provide meaningful GEO SSA capability by combining rapid search, synthetic deep integration, and angular parallax. The study highlights a scalable pathway in which modest apertures, optimized scan strategies, orbital phasing, and onboard processing are jointly leveraged to balance coverage, sensitivity, and orbit determination performance across a diverse GEO population.
This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. The document number is LLNL-ABS-2016486._x000D_
Date of Conference: September 15-18, 2026
Track: Space-Based Assets