MEMS矢量水听器灵敏度自动测试系统设计

  目的  受制作工艺的限制，水听器在不同频点上的灵敏度存在起伏。由于采集设备只能通过设定某个固定的参考灵敏度进行声电转化，这将导致宽带测量信号与实际信号存在偏差，需要通过幅度修正来提高测量信号的准确性。为此，提出一种基于滤波器设计的幅度修正方法，将幅度修正转化为FIR滤波器的设计问题。  方法  首先，利用二阶锥规划设计出满足幅度修正特性的滤波器系数；然后，对测量信号进行时域滤波，从而实现幅度修正。  结果  处理结果表明，修正信号的相对误差随着滤波器长度的增加而减小。当滤波器长度为257时，修正信号的相对误差仅为0.3%。  结论  该方法克服了傅里叶变换分块处理需要时间积累的缺陷，仅通过时域滤波的方式即可获取连续准确的修正信号，具有实时性好、相对误差小、运算量低等优势。

Design of automatic testing system for sensitivity of MEMS vector hydrophone

  Objectives  Due to the limitation of the existing fabrication technology, the sensitivity of hydrophones at different frequency is not exactly the same. Most of the acquisition equipment can only convert acoustic waves to electrical signals by setting up a fixed reference sensitivity, which will lead to the deviation between measured signal and actual signal, so it is necessary to improve the accuracy of the measured signal by amplitude correction. An amplitude correction method of signals measured by hydrophone based on filter design was proposed in this paper to convert the amplitude correction into Finite Impulse Response(FIR)filter design problem.  Methods  Firstly, the filter coefficients that meet the amplitude correction characteristics were designed by the second-order cone programming method. Then, the measured signal is filtered in time domain to achieve amplitude correction.  Results  The processing results show that the relative error of corrected signals decreases with the length of the filter. When the length of the filter is 257, the relative error of the corrected signal is only 0.3%.  Conclusions  The method overcomes the defect of time accumulation required by block processing of Fast Fourier Transform (FFT), and obtains continuous and accurate corrected signals only by using time-domain filter, which has the advantages of good real-time capability, small relative error and low computational cost.