The Clutter Simulation of a Known Terrain by the 3D Parabolic Equation and RCS Computation

Zhiyi Wang, Jieru Ding, Min Wang, Shuyuan Yang2022 · DOI

TL;DR

This paper introduces a novel 3D parabolic equation (3DPE) based method for simulating ground clutter in known terrain. By combining 3D wave propagation modeling with a cell-based radar cross-section (RCS) calculation, it provides a more realistic prediction of clutter compared to traditional statistical models, crucial for optimizing radar deployment.

What Problem Does This Paper Address?

Accurately simulating ground clutter is essential for effective radar operation, particularly in complex environments like urban areas where detecting targets like drones is critical. Traditional clutter simulation methods often rely on statistical models that fail to capture the spatial distribution of ground objects and terrain features. This limitation hinders the ability to predict ideal radar placement and optimize deployment, especially when detailed terrain data is available. The paper addresses this by developing a method that leverages known terrain information to generate more realistic clutter maps.

Methods

The core of this work is the development of a 3D split-step Fourier transform (3D-SSFT) method based on the three-dimensional parabolic equation (3DPE) model. This approach models radio wave propagation in three dimensions, accounting for diffraction and scattering effects.

Key methodological contributions include:

  • 3D Parabolic Equation (3DPE) Model: Extends the standard parabolic equation to a 3D space, enabling more comprehensive wave propagation simulation.
  • 3D Split-Step Fourier Transform (3D-SSFT): An efficient algorithm for solving the 3DPE, allowing for step-by-step propagation calculations.
  • Cell-Based Radar Cross Section (RCS) Calculation: Instead of relying on statistical models, this method calculates the RCS of individual spatial cells that represent ground objects. This allows for the incorporation of detailed terrain and building geometry into the clutter simulation.
  • Perfectly Matched Layer (PML): Used as an absorbing boundary condition to prevent artificial reflections in the simulation.
  • Gaussian Antenna Pattern: The initial field is modeled using a Gaussian antenna pattern, which is then transformed into the wavenumber domain.

Datasets

The paper utilizes a digital elevation map (DEM) of an urban area in China. This map is described as a superposition of a topographic height map and a building height map, covering a 3000 × 3000 m² area with altitudes ranging from 450 to 700 m. This DEM is divided into 1 m × 1 m × 1 m cells for simulation. No new benchmark datasets are introduced.

Key Insights and Ideas

  • Spatial Distribution Matters: The paper strongly emphasizes that accurately modeling the spatial distribution of ground objects and terrain is crucial for realistic clutter simulation, a limitation of purely statistical approaches.
  • 3DPE for Realistic Propagation: The 3D parabolic equation model, coupled with the SSFT, provides a more physically accurate representation of radio wave propagation in complex terrain compared to 2D models.
  • Cellular RCS is Key: Calculating the RCS of individual cells representing terrain features and objects allows for the direct incorporation of geometric information, leading to more detailed clutter maps.
  • Radar Height is Critical: The simulation results highlight that the radar's erection height significantly impacts the signal-to-clutter ratio and the detectability of targets, underscoring the importance of accurate clutter prediction for optimal deployment.
  • 2D PE Limitations: The comparison with 2D PE demonstrates that it fails to account for radar beamwidth, making it insufficient for real-world applications requiring accurate clutter prediction.

Key Results

  • The proposed 3DPE-based simulation method generates ground clutter maps that have good reference value and meet the demand for strong clutter area prediction under known terrain conditions.
  • Simulations show that the field strength decreases with distance from the radar and is significantly affected by building shielding.
  • The experiments with different cell sizes confirm that larger objects require more cells to be accurately represented, and that combining smaller cells can effectively model larger objects.
  • The clutter maps generated at different radar erection heights show that a higher erection height closer to the target's flight path improves the signal-to-clutter ratio.
  • A comparison with 2D PE shows that the 3D PE method provides more distinct range profiles, especially in adjacent azimuths, due to its consideration of radar beamwidth.

Limitations and Future Work

  • Near-Field Effect and Sidelobes: The paper acknowledges a "hollow" phenomenon in the central area of some clutter maps, which is attributed to the failure to consider the influence of the near-field effect and radar sidelobes. This suggests an area for future improvement.
  • Computational Complexity: While 3D-SSFT is presented as an efficient solution, simulating large-scale 3D environments can still be computationally intensive.
  • Scattering Coefficient Simplification: The paper uses a simplified scattering coefficient (0.1) for terrain and targets. Future work could explore more complex and varied scattering models.
  • Real-world Validation: While the simulation results are presented as having good reference value, direct validation against real-world radar measurements in diverse scenarios would further strengthen the findings.

Summary updated

BibTeX
@misc{wang2022clutter,
  author = {Zhiyi Wang and Jieru Ding and Min Wang and Shuyuan Yang},
  title = {{The Clutter Simulation of a Known Terrain by the 3D Parabolic Equation and RCS Computation}},
  year = {2022},
  doi = {10.3390/s22197452},
}
Figure