[Paper Review] Characterization of LEO Satellites With All-Sky Photometric Signatures

Harrison Krantz, Eric C. Pearce+1 more|arXiv (Cornell University)|Oct 6, 2022

Economic Growth and Productivity4 citations

TL;DR

This paper introduces a novel all-sky photometric signature method for characterizing low-Earth orbit (LEO) satellites using effective albedo—a range- and phase-corrected brightness metric derived from over 14,000 observations of Starlink and OneWeb satellites. The approach enables satellite population differentiation, anomaly detection, and identification with high accuracy using machine learning, even from a single observation per satellite, offering a scalable, knowledge-agnostic solution for growing mega-constellations.

ABSTRACT

We present novel techniques and methodology for unresolved photometric characterization of low-Earth Orbit (LEO) satellites. With the Pomenis LEO Satellite Photometric Survey our team has made over 14,000 observations of Starlink and OneWeb satellites to measure their apparent brightness. From the apparent brightness of each satellite, we calculate a new metric: the effective albedo, which quantifies the specularity of the reflecting satellite. Unlike stellar magnitude units, the effective albedo accounts for apparent range and phase angle and enables direct comparison of different satellites. Mapping the effective albedo from multiple observations across the sky produces an all-sky photometric signature which is distinct for each population of satellites, including the various sub-models of Starlink satellites. Space Situational Awareness (SSA) practitioners can use all-sky photometric signatures to differentiate populations of satellites, compare their reflection characteristics, identify unknown satellites, and find anomalous members. To test the efficacy of all-sky signatures for satellite identification, we applied a machine learning classifier algorithm which correctly identified the majority of satellites based solely on the effective albedo metric and with as few as one observation per individual satellite. Our new method of LEO satellite photometric characterization requires no prior knowledge of the satellite's properties and is readily scalable to large numbers of satellites such as those expected with developing communications mega-constellations.

Motivation & Objective

• To develop a scalable, knowledge-agnostic method for characterizing LEO satellites without prior orbital or physical data.
• To address the challenge of identifying and differentiating large numbers of satellites in mega-constellations.
• To enable Space Situational Awareness (SSA) practitioners to detect anomalies and classify unknown satellites using photometric data alone.
• To create a standardized metric—effective albedo—that accounts for apparent range and phase angle for direct satellite comparison.
• To validate the method’s efficacy using machine learning on real-world photometric observations of Starlink and OneWeb satellites.

Proposed method

• The authors collected over 14,000 photometric observations of LEO satellites using the Pomenis LEO Satellite Photometric Survey.
• They introduced the effective albedo metric, which normalizes apparent brightness by distance and phase angle to quantify specularity.
• Effective albedo values were mapped across the sky to generate unique all-sky photometric signatures per satellite population.
• The method uses no prior knowledge of satellite design, orbit, or orientation, relying solely on photometric data.
• A machine learning classifier was trained on effective albedo data to identify satellites from as few as one observation per satellite.
• The all-sky signature enables comparison of reflection characteristics across different satellite models and detection of anomalous behavior.

Experimental results

Research questions

• RQ1 Can effective albedo serve as a robust, range- and phase-corrected metric for comparing the photometric properties of LEO satellites?
• RQ2 To what extent can all-sky photometric signatures differentiate between satellite constellations like Starlink and OneWeb?
• RQ3 Can machine learning accurately identify individual satellites using only effective albedo from a single observation?
• RQ4 How effective is the all-sky signature method in detecting anomalous or unknown satellites in a population?
• RQ5 Can this method scale to thousands of satellites in future mega-constellations without prior satellite information?

Key findings

• The effective albedo metric successfully normalizes brightness across varying distances and phase angles, enabling direct comparison of satellite reflection properties.
• All-sky photometric signatures were found to be distinct for different satellite populations, including sub-models of Starlink satellites.
• A machine learning classifier achieved high-accuracy satellite identification using only the effective albedo metric and as few as one observation per satellite.
• The method demonstrated strong performance in identifying unknown satellites and detecting anomalous members within a population.
• The approach is scalable and requires no prior knowledge of satellite characteristics, making it suitable for monitoring large, evolving constellations.
• The technique enables practical, real-time applications in Space Situational Awareness for tracking and characterizing LEO satellite fleets.