Hybrid patch near-field acoustic holography based on Kalman filter

Abstract

The wave superposition method (WSM) is a classical method for reconstructing the sound field of an arbitrary sound source, but its main drawback is the uncertainty of equivalent source configuration. To overcome this problem, a hybrid patch near-field acoustic holography based on Kalman filter (KF-HPNAH) is proposed. The algorithm consists of three steps: (a) the calculation of the three-dimensional (3D) virtual motion trajectory of sound source using the projected coordinates of the sound source on multiple hologram planes; (b) the application of KF to the trajectory, thus improving the accuracy of 3D localization of the sound source; (c) the formulation of the equivalent source configuration strategy of the WSM based on the 3D coordinates. The KF-HPNAH can effectively reduce the dimension of a WSM matrix and improve the accuracy of sound field reconstruction. Numerical simulations and experiments are carried out to verify the effectiveness of the proposed method. The results show that the KF-HPNAH solves the uncertainty problem of equivalent source configuration of WSM successfully. Compared to the existing methods, KF-HPNAH has the best reconstruction accuracy. The proposed method can thus be used to localize the sound source and reconstruct the sound field radiated from an arbitrary sound source at frequencies between 200 Hz and 900 Hz.