Super-resolution sound field reconstruction method based on sparse Bayesian learning

The invention discloses a super-resolution sound field reconstruction method based on sparse Bayesian learning. M microphones are arranged in a sound source near-field radiation area so as to measure holographic surface sound pressure pH; performing interpolation and extrapolation on the holographic surface sound pressure by using sparse Bayesian learning to obtain holographic surface single interpolation sound pressure, thereby increasing the aperture of the holographic surface and improving the resolution; iterating the interpolation sound pressure for multiple times until the sound pressure of the measuring point in the original aperture tends to be stable, so as to obtain robust holographic surface interpolation sound pressure pI; and the sound pressure pr of the reconstruction surface is calculated by using the statistical optimal near-field acoustical holography. According to the method, the aperture of the holographic surface is expanded and the measuring point density of the aperture of the holographi