标题： 基于传统Born级数公式对真实介质进行GPU加速的换能器场计算 显示英文标题

标题： GPU Accelerated Transducer-Field Calculation using the Traditional Born Series Formulation for Realistic Media

摘要： 本研究数值求解了描述均匀且无损耗、吸收性和色散性、非均匀性和非线性介质中声波传播的非齐次Helmholtz方程。 传统的Born级数（TBS）方法已被用于求解此类方程。 该方法中的全波解表示为未受扰方程解的无限和，其权重随着势能的更高次幂增加。 通过TBS过程估算的线性声源阵列（线源）的模拟压力场模式与标准时域方法（k-Wave工具箱）的结果表现出良好的一致性。 尽管TBS方案是迭代的，但是一种非常快速的方法。 例如，实现TBS过程的GPU启用CUDA C代码在计算均匀且无损耗介质的压力场时仅需5秒，而后续模块则需要近500秒。 相应CPU代码的执行时间约为20秒。 本研究的发现表明，TBS方法在求解非齐次Helmholtz方程方面是有效的，而基于GPU的实现显著减少了计算时间。 该方法可以实际应用于计算为不同应用设计的真实换能器产生的压力场。 显示英文摘要

摘要： This study numerically solves inhomogeneous Helmholtz equations modeling acoustic wave propagation in homogeneous and lossless, absorbing and dispersive, inhomogeneous and nonlinear media. The traditional Born series (TBS) method has been employed to solve such equations. The full wave solution in this methodology is expressed as an infinite sum of the solution of the unperturbed equation weighted by increasing power of the potential. Simulated pressure field patterns for a linear array of acoustic sources (a line source) estimated by the TBS procedure exhibit excellent agreement with that of a standard time domain approach (k-Wave toolbox). The TBS scheme though iterative but is a very fast method. For example, GPU enabled CUDA C code implementing the TBS procedure takes 5 s to calculate the pressure field for the homogeneous and lossless medium whereas nearly 500 s is taken by the later module. The execution time for the corresponding CPU code is about 20 s. The findings of this study demonstrate the effectiveness of the TBS method for solving inhomogeneous Helmholtz equation, while the GPU-based implementation significantly reduces the computation time. This method can be explored in practice for calculation of pressure fields generated by real transducers designed for diverse applications.