ABSTRACT
The numerical simulation of weakly nonlinear ultrasound is important in treatment planning for focused ultrasound (FUS) therapies. However, the large domain sizes and generation of higher harmonics at the focus make these problems extremely computationally demanding. Numerical methods typically employ a uniform mesh fine enough to resolve the highest harmonic present in the problem, leading to a very large number of degrees of freedom. This paper proposes a more efficient strategy in which each harmonic is approximated on a separate mesh, the size of which is proportional to the wavelength of the harmonic. The increase in resolution required to resolve a smaller wavelength is balanced by a reduction in the domain size. This nested meshing is feasible owing to the increasingly localised nature of higher harmonics near the focus. Numerical experiments are performed for FUS transducers in homogeneous media to determine the size of the meshes required to accurately represent the harmonics. In particular, a fast volume potential approach is proposed and employed to perform convergence experiments as the computation domain size is modified. This approach allows each harmonic to be computed via the evaluation of an integral over the domain. Discretising this integral using the midpoint rule allows the computations to be performed rapidly with the FFT. It is shown that at least an order of magnitude reduction in memory consumption and computation time can be achieved with nested meshing. Finally, it is demonstrated how to generalise this approach to inhomogeneous propagation domains.
ABSTRACT
Recently, the volume integral equation (VIE) approach has been proposed as an efficient simulation tool for silicon photonics applications [J. Lightw. Technol.36, 3765 (2018)JLTEDG0733-872410.1109/JLT.2018.2842054]. However, for the high-frequency and strong-contrast problems arising in photonics, the convergence of iterative solvers for the solution of the linear system can be extremely slow. The uniform discretization of the volume integral operator leads to a three-level Toeplitz matrix, which is well suited to preconditioning via its circulant approximation. In this paper, we describe an effective circulant preconditioning strategy based on the multilevel circulant preconditioner of Chan and Olkin [Numer. Algorithms6, 89 (1994)NUALEG1017-139810.1007/BF02149764]. We show that this approach proves ideal in the canonical photonics problem of propagation within a uniform waveguide, in which the flow is unidirectional. For more complex photonics structures, such as Bragg gratings, directional couplers, and disk resonators, we generalize our preconditioning strategy via geometrical partitioning (leading to a block-diagonal circulant preconditioner) and homogenization (for inhomogeneous structures). Finally, we introduce a novel memory reduction technique enabling the preconditioner's memory footprint to remain manageable, even for extremely long structures. The range of numerical results we present demonstrates that the preconditioned VIE is fast and has great utility for the numerical exploration of prototype photonics devices.
