基于声矢量传感器阵列的DOA估计
发布时间:2018-06-16 05:05
本文选题:声矢量传感器 + Unitary-MUSIC算法 ; 参考:《哈尔滨工业大学》2014年硕士论文
【摘要】:声矢量传感器是由声压传感器与质点振速传感器两部分构成的,在空域环境中具有同步采集声场中的声压信息(标量)与振速信息(矢量)的特性。其增加的数据维度等同于更多的数据快拍,较之声压传感器会有更高的检测灵敏度。但是现有的大量声矢量阵信号处理技术都是将矢量传感器的振速信息作为额外的独立阵元来处理,而没有充分利用远场环境下声压与振速的相关性特点。事实上,在各项同性的噪声场中,声压及振速是非相关的;利用这一特性,即可在信号波达方向角估计过程有效降低高斯白噪声所产生的影响及干扰。本文在回顾声矢量传感器的基本原理、声压振速联合信息处理的物理基础以及子空间类算法的声矢量阵信号处理技术的基础上,采用声压振速联合信息对矢量传感器阵列信号处理进行算法上的改进,为解决在低信噪比环境下高效地定位目标方位提供了一种可行的解决思路。本文引入了基于声压振速联系信息的P-V互协方差矩阵来处理接收数据信号。这种新型的协方差矩阵将振速信息合成到某一观测方向上,避免了接收数据矩阵的维度增加,因此降低了声矢量阵信号处理的计算复杂度。在接收信号矩阵的后续处理中,本文结合现有的Unitary-MUSIC算法和Root-MUSIC算法提出了一种基于声矢量传感器阵P-V互协方差矩阵的Unitary-Root MUSIC改进算法。该算法不仅在原有的算法基础上提升了信号的检测性能,更进一步降低了计算复杂度。文中也从不同信噪比条件下的均方根误差、检测概率、空间功率谱、仿真耗时等多个角度对新提出的算法进行了仿真,并通过与现有算法的对比,验证了其优秀的检测性能。
[Abstract]:Acoustic vector sensor is composed of sound pressure sensor and particle vibration velocity sensor. It has the characteristics of collecting sound pressure information (scalar) and vibration velocity information (vector) synchronously in the spatial environment. The increased data dimension is equivalent to more data shot and has higher detection sensitivity than sound pressure sensor. However, a large number of existing acoustic vector array signal processing techniques are based on the vector sensor's vibration velocity information as additional independent elements, without fully utilizing the correlation between sound pressure and vibration velocity in far-field environment. As a matter of fact, the sound pressure and velocity are non-correlated in various homogeneous noise fields, which can effectively reduce the influence and interference caused by the white Gao Si noise in the estimation of the DOA of the signal. On the basis of reviewing the basic principle of acoustic vector sensor, the physical foundation of combined information processing of acoustic pressure and vibration velocity and the technology of acoustic vector array signal processing based on subspace algorithm, The algorithm of vector sensor array signal processing is improved by using the combined information of sound pressure and vibration velocity, which provides a feasible solution for efficiently locating target azimuth in low signal-to-noise ratio (SNR) environment. In this paper, P-V cross covariance matrix based on acoustic pressure and velocity contact information is introduced to process received data signals. The new covariance matrix synthesizes the vibration velocity information to a certain observation direction, avoids the dimension increase of the received data matrix, thus reduces the computational complexity of the acoustic vector array signal processing. In the following processing of the received signal matrix, an improved Unitary-Root music algorithm based on P-V cross-covariance matrix of acoustic vector sensor array is proposed, which combines the existing Unitary-MUSIC algorithm and Root-MUSIC algorithm. This algorithm not only improves the performance of signal detection, but also reduces the computational complexity. This paper also simulates the proposed algorithm from different SNR conditions, such as root mean square error, detection probability, spatial power spectrum, time consuming and so on, and verifies its excellent detection performance by comparing with existing algorithms.
【学位授予单位】:哈尔滨工业大学
【学位级别】:硕士
【学位授予年份】:2014
【分类号】:TP212;TN911.7
【参考文献】
相关期刊论文 前1条
1 江南,黄建国,冯西安,管静;矢量传感器阵列的空间谱估计及定向性能分析[J];昆明理工大学学报(理工版);2003年02期
,本文编号:2025449
本文链接:https://www.wllwen.com/kejilunwen/wltx/2025449.html