"A convolutional plane wave model for sound field reconstruction"

This repository provides a Python implementation of plane wave models for sound field reconstruction. In particular, global, local and convolutional plane wave models are provided. Please refer to the paper (arXiv) https://arxiv.org/abs/2208.11324 or get in touch for more details. Updates are expected in the publishing process.

Code License:

MIT -- see the file LICENSE for details.

If you find this useful, please cite:

@report{hahmann2022csc,
  doi = {10.48550/ARXIV.2208.11324},
  url = {https://arxiv.org/abs/2208.11324},
  author={Hahmann, Manuel and {Fernandez Grande}, Efren},
  title = {A convolutional plane wave model for sound field reconstruction},
  publisher = {arXiv},
  year = {2022},
}

Structure

Repo is structured as follows:

.
├── Makefile                       # [Makefile to install dependencies, download data, and run the scripts]
├── data                           # [data sets, get it via 'make data' or link below]
├── src                            # [helper functions and tools]
├── figures                        # [folder is generated by scripts]
├── README.rst
...

To generate the figures

Step 1) clone repository and start a fresh python environment Step 2) install python packages dependencies:

make install

Step 3) Download measurement data set (uses wget to download the published dataset https://data.dtu.dk/articles/dataset/Acoustic_frequency_responses_in_a_conventional_classroom/13315286; Results for Figure 7 are included in the repo as it takes a long time to generate).:

make data

Step 4) run the different scripts, generate the graphs and collect the pdfs in ./figures

make