TNW-BIS-BMPI
Horst Complex ZH267
Workdays
even
odd
M
T
W
T
F
on campus
work from home
out of office

If you are interested in Bachelor's or Master's assignments/projects with Ivo Vellekoop's Wavefront Shaping group related to wave propagation simulations with our in-house developed Maxwell solver, WaveSim, you can have a look at this page.

Running even larger wave propagation simulations made possible with latest developments to WaveSim published in article

WaveSim is a free and open-source solver developed in-house at BMPI for simulating wave propagation in large and complex structures. We have been working on improving and accelerating our fast and accurate wave propagation solver, and in December 2025 we published a significant addition to WaveSim’s capabilities (link to publication “Domain decomposition of the modified Born series approach for large-scale wave propagation simulations.”).

Using a strategy called domain decomposition to split the simulations over multiple processors or graphics processing units (GPUs), WaveSim can now simulate light propagation at much larger sizes. This new method maintains the accuracy, memory efficiency, and the guarantee of converging with every iteration of the original method without domain decomposition.

To demonstrate the new improvements, we ran a simulation with an unprecedented 2.1 billion voxels (320×320×320 wavelengths, sampled at 4 points per wavelength) in just 45 minutes on two A40 GPUs. This was over 4500 times the size of a recent simulation using the finite difference time domain method, which is very popular in light propagation simulations. You can try WaveSim for yourself, it is already available on GitHub.

The below image shows a small 2D slice of 100×100 wavelengths from the large 3D problem. Here the light source is a plane wave propagating from the left side of the x-axis through a collection of small spheres. The image is showing 160000 voxels, just 0.007% of the total 2.1 billion that WaveSim simulated!

Organisations

Publications

2023

Temporally variable recurrence regimes of mega-tsunamis in the 6500 years prior to the 2004 Indian Ocean event (2023)Marine geology, 460. Article 107051. Sanwal, J., Rajendran, C. P., Heidarzadeh, M., Mache, S., Anandasabari, K. & Rajendran, K.https://doi.org/10.1016/j.margeo.2023.107051Introducing Nonuniform Sparse Proximal Averaging Network for Seismic Reflectivity Inversion (2023)IEEE Transactions on Computational Imaging, 9, 475-489. Mache, S., Pokala, P. K., Rajendran, K. & Seelamantula, C. S.https://doi.org/10.1109/TCI.2023.3277629

2022

An Ensemble of Proximal Networks for Sparse Coding (2022)In 2022 IEEE International Conference on Image Processing (ICIP). Reddy Nareddy, K. K., Mache, S., Pokala, P. K. & Seelamantula, C. S.https://doi.org/10.1109/ICIP46576.2022.9897607Hilbert–Huang Transform and Energy Rate Functions for Earthquake Source Characterization—A Study from the Japan Trench (2022)Bulletin of the Seismological Society of America, 112(6), 2847–2858. Mache, S., Chatterjee, A., Rajendran, K. & Seelamantula, C. S.https://doi.org/10.1785/0120220099

Research profiles

Address

University of Twente

Horst Complex (building no. 20), room ZH267
De Horst 2
7522 LW Enschede
Netherlands

Navigate to location

Organisations

Scan the QR code or
Download vCard