Size Estimation of Space Object Based on Narrowband Radar

Chunrui Yu, Diannong Liang, Zhen Dong, Anxi YuChinese 2010 · DOI

TL;DR

Researchers have developed a novel method to estimate the shape and size of space objects using narrowband radar data. By modeling satellites and debris as ellipsoids and analyzing the statistical variance of their Radar Cross Section (RCS) time series, the team can reliably determine an object's dimensions without needing high-resolution imaging.

What Problem Does This Paper Address?

As Earth's orbit becomes increasingly crowded, the risk of collisions between operational satellites and space debris grows. To assess this risk, we need to know the size and shape of these objects. While wideband radar can provide high-resolution images, narrowband radar is much more common and cost-effective.

The challenge is that narrowband radar only provides a Radar Cross Section (RCS)—a measure of how detectable an object is by radar. RCS is notoriously difficult to interpret because it fluctuates wildly based on the object's material, complex geometry, and, most importantly, its orientation (aspect angle) relative to the radar. Traditional methods often require massive databases of known objects to make a guess; this paper seeks a more direct, analytical approach.

Methods

The authors propose a two-step estimation process based on the assumption that many space objects can be effectively modeled as ellipsoids.

  1. Shape Estimation: The researchers discovered that the standard deviation of a normalized RCS time series is primarily a function of the object's axial ratio (the ratio of its length to its width) and is largely independent of its absolute size. By calculating the standard deviation of the received signal, they can look up the corresponding axial ratio on a pre-calculated curve.
  2. Size Estimation: Once the shape (ratio) is known, the absolute size must be determined. The authors propose a thresholding method. They set a threshold based on the mean and standard deviation of the RCS sequence ($\alpha = \bar{x} + \sigma$). By averaging the RCS values that exceed this threshold (representing the "peaks" when the radar sees the object's largest profile), they can calculate the effective radius of the long axis and, subsequently, the short axis using the previously determined axial ratio.

Datasets

The study validates its model using real-world narrowband radar data rather than just simulations:

  • Space Debris: Six different data segments of unidentified space debris were used to test shape estimation.
  • Three-Axis Stabilized Satellite: Ten segments of data from a known satellite were used to evaluate the accuracy of the two-dimensional size estimation.

Key Insights and Ideas

  • The Invariance of Variance: The most significant "aha" moment is that the standard deviation of the RCS signal is a "stable feature parameter." Because it relates to the shape regardless of the scale, it provides a shortcut to identifying whether an object is long and thin or spherical.
  • Moving Beyond the Mean: While previous models (like NASA’s Size Estimation Model) often rely on the average RCS, this paper shows that the statistical distribution (the spread of the data) contains vital information about the target's geometry that the average obscures.
  • Thresholding for Peak Detection: By intelligently filtering the RCS time series to focus on the highest-return segments, the authors can isolate the dimensions of the object's primary axes more accurately than by looking at the entire noisy signal.

Key Results

  • Debris Analysis: For the six debris segments, the method produced consistent axial ratio estimates (ranging from approximately 2.2 to 3.1), demonstrating the method's stability across different observation windows.
  • Satellite Dimensions: When tested on the stabilized satellite, the method yielded highly consistent results across 10 different data segments. The average estimated long-axis size was 2.64 meters and the short-axis was 1.13 meters.
  • Feasibility: The results closely matched existing estimation methods (like those by Huang et al.), but the authors' method proved more versatile as it does not require the target to be spinning (it works for non-spinning targets as well).

Limitations and Future Work

  • The Ellipsoid Assumption: The method assumes the target is roughly ellipsoidal. While this is a good approximation for many debris fragments and satellite buses, it may struggle with highly complex geometries, such as satellites with large, protruding solar arrays or parabolic antennas.
  • Aspect Angle Coverage: The accuracy of the estimation depends on the radar seeing enough of the object's rotation or movement to capture a representative sample of its RCS. If the radar only sees the object from a single, unchanging angle, the statistical variance won't represent the true shape.
  • Future Directions: The authors suggest that these extracted features (shape and size) are excellent candidates for input into automated target recognition systems to classify space objects more effectively.

Summary updated

Abstract

Earth orbiting spacecrafts have become an integral part of our everyday lives. A real and growing concern for the safety and reliability of these satellites is the threat from collision with other orbiting objects including space debris. Size estimation of space object is an important task in evaluating its danger of a possible collision with operational spacecraft. RCS (Radar Cross Section) includes much information of space object, and extracting the feature of the object size from the RCS time series of narrowband radar is very helpful to radar target identification. Unfortunately, it is very difficult to get the information of target shape and size from its RCS. In this paper, based on the research of the characteristic of the ellipsoid's RCS, a shape estimation method of space object is proposed, and the problem of size estimation is studied further with a new method of size estimation proposed. The result using real radar data shows that this method is feasible to estimate the shape and two-dimensional sizes of the space object.
BibTeX
@inproceedings{yu2010size,
  author = {Chunrui Yu and Diannong Liang and Zhen Dong and Anxi Yu},
  title = {{Size Estimation of Space Object Based on Narrowband Radar}},
  booktitle = {Chinese},
  year = {2010},
  doi = {10.11728/cjss2010.03.250},
}
Figure