EI / SCOPUS / CSCD 收录

中文核心期刊

迭代重加权时域平面波叠加法的非稳态声场重建

Iteratively reweighted time domain plane wave superposition method for reconstructing the non-stationary sound field

  • 摘要: 提出了迭代重加权时域平面波叠加法, 并应用于非稳态声场重建。该方法首先通过时域离散卷积得到全息面时域声压与虚源面时域波数谱离散状态公式; 然后根据稀疏表示和迭代重加权, 得到残差范数与加权解范数最小化问题表达式; 最后将每一对波数下虚源面时域波数谱与时域传播核卷积叠加得到重建面上的时域声压。为了验证该方法的有效性, 给出了激励板声源的仿真验证, 并在半消声室内进行了实验验证。讨论了一些参数对该方法的影响, 并通过与基于Tikhonov正则化的时域平面波叠加法比较, 突出该方法的重建优势。仿真和实验结果表明, 该方法对非稳态声场重建具有很好的效果, 并且重建精度高。

     

    Abstract: An iterative reweighted time domain plane wave superposition method is proposed to reconstruct the non-stationary sound field. The discrete state formulas of hologram plane time-domain pressure and virtual source plane time-wavenumber spectrum are firstly obtained by discrete convolution in the time domain. Then, based on the sparse representation and iteratively reweighting model, an expression for the minimization problem of residual norm and weighted solution norm is estimated. Finally, the time-domain pressure on the reconstruction plane is obtained by a superposition of time convolution between the time-wavenumber spectrum on the virtual source plane and the time-domain propagation kernel at each wavenumber. Numerical simulations with exciting planar plate source demonstrate the feasibility, and experimental validation is carried out in the semi-anechoic chamber. The influences of some parameters of the proposed method are discussed in the paper, and its reconstruction advantages are protruded by comparing to the time domain plane wave superposition method with Tikhonov regularization. The simulation and experimental results show that the method is effective for non-stationary sound field reconstruction with higher reconstruction accuracy.

     

/

返回文章
返回