MEMS矢量传感器方位估计误差分析

Analysis of DOA Estimation Error for MEMS Vector Sensor

为提高MEMS仿生矢量传感器的定位精度,对其误差来源进行分析。对于单个矢量传感器,误差来源于其自身灵敏度不一致及流噪声的影响。提出通过分段插值来绘制线性曲线、选用差分测量结构和调整两路输出信号的大小使灵敏度一致;对于流噪声产生的误差,提出一种加聚氨酯导流罩的新结构,可以使水听器抑制流噪声的能力较之前提高20 dB。对阵列而言,选取不同阵元数进行仿真对比分析,得出随着阵元数的增加,水声传感器阵列DOA估计的准确度会随之增高,分辨力也随之增强。通过对影响MEMS矢量传感器方位估计的误差来源进行分析,进一步提出解决方案,为提高其定位精度提供了有力的科学支撑,从而为MEMS矢量传感器的工程化应用奠定了坚实的基础。

In order to improve the positioning accuracy of MEMS bionic vector sensor,the error source is analysed. For single vector sensor,its own sensitivity inconsistency and flow noise lead to the errors. Piecewise interpolation to draw a linear curve is proposed,selecting the differential measurement structure and adjusted the two output signals are proposed to resolve the problem in the former,then a new structure of polyurethane dome is designed to inhibit the flow noise. Ability to inhibit flow noise of the hydrophone is increased 20 dB. For hydrophone array,the different numbers of hydrophone is selected to analyse comparison simulation,the result shows that the increase of the number of array elements,underwater acoustic sensor array DOA estimation accuracy can be increased,the resolution is also improved. By analyzing DOA estimation error sources,then the solutions are put forward. To provide powerful scientific support so as to improve the positioning accuracy,it laid the foundation for MEMS vector sensor engineering application.