Backscattering From Randomly Placed Dirichlet Particles: Simulated Wave Data

A simulated dataset of backscattered waves governed by the 2D wave equation from randomly placed Dirichlet particles. The data was created by Artur L. Gower of the University of Manchester and includes subsets for incident waves with different wavenumber (k) and time (t) ranges. The dataset is provided via the paperswithcode platform.

Use Cases

• Training machine learning models to predict backscattering patterns based on particle radius and volume fraction ranges.
• Benchmarking numerical solvers for the 2D wave equation against simulated data with known parameters.
• Studying the effects of multiple scattering in media with randomly placed circular obstacles.
• Developing inverse scattering methods to infer particle properties from backscattered wave data.

Strengths

• Simulated data provides a controlled environment with known parameters, including particle radius (0.2 to 2.0) and volume fraction (1% to 20%).
• Includes distinct subsets (bunny, bunnytest, train, test) for different incident wave conditions, facilitating training and evaluation.

Limitations

• Row count and dataset size are unknown, which may limit suitability assessment for large-scale modeling.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

University of Manchester

Collection Method

Simulation governed by the 2D wave equation.

Time Range

null

Freshness

null

Geography

null