Ultrasound-to-ultrasound Volterra filter identification of the parametric array loudspeaker

The Volterra filter is a favorable approach to represent a variety of nonlinear systems, including the parametric array loudspeaker (PAL), which is a weak nonlinear acoustic system to create directional sounds. Using the Volterra filter to model the sound process of the PAL saves the computational cost of solving the nonlinear acoustic equation. In the past studies, the Volterra filter is identified from the audio input to the audio output of the PAL, which is known as the audio-to-audio Volterra filter (A2VF). In this paper, the ultrasound-to-ultrasound Volterra filter (U2VF) is recommended. The experiment results validate that the U2VF outperforms the A2VF, in terms of the robustness to the change of the modulation index in the PAL.