复域压缩感知近场声图测量方法

针对频域压缩感知近场声图测量方法性能下降的问题，基于阵列信号处理中相移与时延关系，按近场扫描位置在复域对各通道数据进行时延补偿和相关处理，建立复域感知矩阵和观测序列构建方法，提出复域感知矩阵近场声图测量模型。以数值仿真和某近场目标定位试验中测得数据为案例，将该模型与频域压缩感知模型处理结果进行对比。结果表明，相比频域压缩感知方法，通过改变观测序列和感知矩阵构建方式，最低信噪比的适应性得到近10lgM dB（M为通道数）的提升，提高了近场声图测量性能。

A Near-field Acoustic Image Measurement Method Based on Compressed Sensing in Complex Domain

In order to solve the performance degradation problem of near-field acoustic image measurement method based on compressed sensing in frequency domain, the time delay compensation and correlation are processed based on the relationship between phase shift and time delay in array signal processing according to the near-field scanning position in the complex domain. A near-field acoustic image measurement model of sensing matrix in complex domain is established. Taking the numerical simulation data and the data measured in a near-field target location experiment as an example, the processed results of the proposed model and the frequency domain compressed sensing model are compared. The results show that, compared with frequency domain compressed sensing, the tolerance of lowest signal-to-noise ratio of this method is improved nearly 10lgM dB (where M is the channel number of array) by changing the construction way of sensing matrix and observation sequence, and the ability of near-field acoustic image measurement is improved.