Quantum Art proposes a more efficient quantum method for simulating waves

[BigDog_Fan]

Israeli trapped ion company Quantum Art has outlined a new approach to simulating waves on a quantum computer. The Quantum Insider, carrying the company's announcement, reports on a paper posted to arXiv that focuses on steady state waves at a single frequency

Wave simulations matter for predicting mobile phone coverage, structural engineering, seismic modelling and materials science. Some classical simulations need between 10 to the power 18 and 10 to the power 21 grid points, for example modelling radio waves across tens of kilometres at centimetre accuracy

Existing quantum methods treat wave problems as generic systems of equations, wasting effort on solutions that physics rules out. Quantum Art's method instead builds the set of physically allowed wave directions directly and places each source onto it

The company claims the number of attempts stays roughly constant as the simulated area grows, that effort rises only slowly with added detail, and that success odds improve with more wave sources. It estimates that very large problems could potentially run on around 100 qubits. Chief technology officer Amit Ben Kish said the paper explains the science behind its approach and could extend to sound, radar and seismic waves

Using physics to trim the problem before throwing qubits at it is a sensible idea. The claims come from the company and have not been independently tested, so caution is needed. The 100 qubit figure is also carefully hedged

Could wave simulation be an early practical use for quantum computers? And how much weight do you give to company claims like these?

Quantum Art Research Outlines More Efficient Quantum Approach to Wave Modelingthequantuminsider.com