Abstract: A one-step patch near-field acoustic holography technique based on support vector regression is proposed to solve hologram data extrapolation problem.The regression functions are constructed by treating the measured data on the patch hologram as training samples,and the data outside the measured aperture are extrapolated.Finally,the sound field is reconstructed by means of statistically optimized near field acoustic holography.Both numerical and experimental results show that the extrapolation of the measured sound pressure outside the patch hologram aperture is achieved easily.Furthermore,the reconstruction results show that the proposed method is an effective patch near-field acoustic holography technique even the initial hologram data is disturbed by noise.