The duo behind SoundSim360 - Winners of VentureChallenge 2024

Ken Mattsson, Professor of Computational Science, and Gustav Eriksson, PhD, have developed the sound simulation tool SoundSim360, which can provide precise sound level predictions. SoundSim360 handles complex sound simulations both indoors and outdoors, accounting for reflections and transmissions through walls, as well as considering atmosphere, topography, and other relevant physical factors.

The ability to manage low-frequency sounds, which are often perceived as disturbing, is unique to their product. Simulating low-frequency sounds is essential for minimizing noise, for example, when siting wind turbines, improving acoustics in an arena, or achieving optimal setup for a home theater system.

Success Factors

– It was extremely fortunate to find a student who is so interested and who can also make it work in practice. Gustav learned to program GPUs during his time as a doctoral student with me, as we were researching environmental noise. That work is largely the foundation for what we do now. It was previously considered too expensive and time-consuming to use the method we developed, but with technological advancements in GPUs for AI applications, it’s now possible to perform these calculations without supercomputers, explains Ken Mattsson, Professor.

– What we do is actually better suited to today’s GPUs than machine learning, adds Gustav Eriksson, PhD.

– Of course, there must be a solid mathematical foundation, and it’s the combination of all these factors that now allows us to perform advanced optimization calculations and inverse modeling, making our solution unique and enabling us to measure low-frequency sounds with precision, continues Ken.

Short-term goals and bigger visions

Currently, the team, together with UU Invest, is focusing on finding product-market fit to generate revenue. There are several application areas where noise simulations are an important avenue to explore in the market.

– Today, there is no product that can simulate transmission through façades below 50Hz, but regulations go down to 30Hz, which makes us unique, explains Ken.

A potential future direction is creating an app that could be used by private individuals, not just acoustic consultants and other professionals, providing greater scalability. Another interesting area is marine applications, such as simulating noise propagation from offshore wind farms and ships. Currently, no relevant solutions exist in this domain.

Right now, the focus is on acoustic waves, but our computational method could potentially be applied to seismic, electromagnetic, or even quantum mechanical waves.

– It’s a different type of physics, but mathematically, it’s very similar, says Ken.

Time as a Challenge

Time management, as with many start-ups, is a major challenge. Long-term funding through customers or investors is needed to scale up the team.

– We need to conduct measurements and experiments to obtain the correct parameters for our calculations, and for that, we need both funding and collaboration partners, says Ken.

– We don’t really see any major technical challenges now, adds Gustav.

A Long Journey to Today’s Solution

Twenty-five years of research form the basis for the mathematical calculations used in SoundSim360. Ken Mattsson, Professor, began his journey as a student in Uppsala and has worked at Stanford University, in nuclear power plants, and at FOI (Swedish Defence Research Agency) before returning to Uppsala University as a professor.

– I attended a program at UU Innovation to develop this product 12 years ago, but the timing simply wasn’t right then, says Ken.

Gustav Eriksson, PhD, studied engineering physics at Uppsala University. During his studies, Gustav took Ken’s Advanced Numerical Methods course before his thesis project and was later offered a doctoral position in Ken’s group.