Computationally efficient transparent sound source for the finite-difference time-domain method

The standard acoustic finite-difference time-domain methods often employ a transient input waveform as a sound source and introduce the waveform into an update equation of the sound pressure. There are three techniques to introduce the waveform: hard source, soft source, and transparent source. In the present paper, the advantages and disadvantages of these three techniques are firstly illustrated. Although the transparent source is most appropriate for obtaining an impulse response, an existing approach to achieve the transparent source requires heavy precomputation. To overcome this issue, the present paper proposes a computationally efficient approach for introducing the transparent source. It is derived from a mathematical investigation of the relation between an input source signal and the excited wave field. To validate the proposed approach, simple sound fields where diffraction effects can be ignored are considered, and the numerical results of the proposed approach are compared with those of the geometrical prediction. Furthermore, the proposed approach is applied to one-and two-dimensional situations for comparison with the existing methods. Computational efficiency of the proposed approach is shown, and some discussions on the difference of excited sound fields are provided.