Preprocessing Methods of Parametric Array Loudspeakers
摘要
The original work of the parametric array by Westervelt discussed the generation of difference frequency signal from two high-frequency collimated beams, which are referred as primary waves. The directivity of the parametric array is attributed to an end-fire array of virtual sources that are the byproducts of the nonlinear interaction of the primary waves in the medium. A more complete explanation of the parametric array was given by Berktay and his analysis was not limited to two primary waves. He used the concept of the modulation envelope. The Berktay’s farfield solution, which defines the demodulated secondary waveform generated along the axis of the beam, is proportional to the second-time derivative of the square of the envelope. To eliminate the harmonic distortion in the audio frequency obtained by demodulation, scholars have carried out research on the improvement of modulation algorithms. Based on the Berktay far-field solution theory, scholars have proposed various amplitude modulation methods: DSBAM, SRAM, SSB-AM, RSSB-AM and MAM, etc. In practical applications, it has been found that the above modulation methods based on the Berktay far-field solution cannot eliminate the nonlinear distortion in the parametric array loudspeaker system. Moreover, when the sound pressure level of the ultrasonic primary wave signal exceeds a certain value, the performance of these improved modulation algorithms will significantly decrease. Therefore, it is necessary to study and introduce new algorithms to compensate for the nonlinear distortion of the parametric array loudspeaker. Currently, the Volterra Filter-based preprocessing method has been used to some extent to reduce the nonlinear distortion of the parametric array loudspeaker. Therefore, this chapter firstly adopts the KZK equation to conduct a comparative theoretical analysis with and without equalization of the four commonly used preprocessing methods: DSBAM, SRAM, SSB-AM, MAM for reproducing directional speech signals in parametric array loudspeakers, focusing on comparing the simulation results obtained from the demodulated signal between the KZK equation and the Berktay far-field solution. Secondly, this chapter also introduces the ODVF (One Dimensional Volterra filter) model into the preprocessing research of parametric array loudspeakers, and conducts corresponding theoretical and experimental research.