Page 55 - 《应用声学》2022年第6期
P. 55

第 41 卷 第 6 期                                                                       Vol. 41, No. 6
             2022 年 11 月                         Journal of Applied Acoustics                 November, 2022

             ⋄ 研究报告 ⋄



              混响环境下目标声场稀疏分解多域声场重放方法                                                                       ∗





                                             杜博凯 曾向阳             †    洪 汐


                                               (西北工业大学航海学院       西安   710072)

                摘要：对于封闭空间内的声场重放，传统的多点声压匹配方法 (Cov-PM) 依据在目标声场测得的声压直接使
                用最小二乘来计算扬声器权重。然而这种方法要求较多的目标声场采样点以实现足够的精度。对于一类特殊
                的声场景，也就是目标声场是由少量声源辐射产生时，提出一种对目标声场稀疏分解的方法来进行混响环境
                下多域声场重放以降低对目标声场采样数量的要求。该文给出基于目标声场等效源稀疏分解多域重放方法
                (Sparse-ESM) 理论推导，通过数值计算以及实验测试两种方式对比所提方法与最小二乘等效源分解方法以及
                Cov-PM 的声场重放性能。数值结果表明，在 600 Hz 以上的频段，Sparse-ESM 方法的重放误差性能提升明显。
                实验结果也得出了与数值计算相同的结论。同时，还通过数值计算和实验测试两种方法证明了当目标声场声
                源方向波动时，Sparse-ESM 仍然可以保持与其余两类方法相近的声对比度，并且实现较高的亮区重放精度。
                关键词：多区域重放；声对比度；稀疏分解
                中图法分类号: TB53           文献标识码: A          文章编号: 1000-310X(2022)06-0901-10
                DOI: 10.11684/j.issn.1000-310X.2022.06.007





                   Multizone sound field reproduction based on target sound field sparse

                                 decomposition under reverberant environment


                                         DU Bokai    ZENG Xiangyang     HONG Xi

                       (School of marine science and technology, Northwestern Polytechnical University, Xi’an 710072, China)

                 Abstract: Sound field reproduction in reverberant environment is conventionally achieved by multi-point based
                 pressure matching method (Cov-PM). However, this method requires a considerable number of microphones to
                 record the target sound field in order to obtain accurate reproduction. With the assumption that the target
                 sound field is generated by a small number of sources, a sparse equivalent source decomposition based multizone
                 reproduction method (Sparse-ESM) is proposed to reduce the record microphone number. The theory of the
                 proposed method is given in this paper, and the method performance is compared with least-square equivalent
                 source decomposition-based method (LS-ESM) and the Cov-PM by simulation and experiment. Simulation
                 results show that the Sparse-ESM improves the reproduction accuracy above the 600 Hz. Similar conclusion
                 can be drawn from experiment results. At the same time, simulation and experiment results demonstrate that
                 the Sparse-ESM achieve the better reproduction accuracy and close acoustic contrast compared with the other
                 two methods.
                 Keywords: Multizonere production; Acoustic contrast; Sparse decomposition


             2021-09-29 收稿; 2022-03-07 定稿
             国家自然科学基金项目 (11774291)
             ∗
             作者简介: 杜博凯 (1994– ), 男, 山西长治人, 博士研究生, 研究方向: 声场重放。
             † 通信作者 E-mail: zenggxy@nwpu.edu.cn
   50   51   52   53   54   55   56   57   58   59   60