摘要
针对超定条件下多运动声源目标二维波达方向角估计中,传统的波束形成法、MUSIC法等不能分辨临近目标、需要设置立体传感器阵列等问题,提出了基于圆形传感器阵列的独立分量分析算法。算法结合完全正交分解和GIVENS旋转实现数据的在线预白化并更新,采用向量来表示传感器和运动目标的方位角,通过结合独立分量分析算法和波束形成法,间接估计出了运动源的二维方向角,最后进行了数值模拟实验并与传统MUSIC方法进行比较。结果表明,该方法能够准确地探测和识别多运动声源目标的数目并估计其二维波达方向,为智能地雷目标探测提供了新方法。
When estimating the direction of arrival （DOA） of multiple sources for an over-determined situa- tion in 3-D space,where the number of sensors exceeds the number of sources,the traditional beam-form- ing method and the MUSIC method can not distingush the adjacent targets,and need to set up a three di- mensional sensor array. In view of this situation,a new method based on the beam-forming of the observa- tion vectors and independent component analysis was proposed. The complete orthogonal decomposition and GIVENS rotation were employed to achieve data online pre-whitening, updating and releasing. The method employs the vector of the sources and the angle （angle between the sources and the vectors of sen- sors） to get DOA,together with the method of independent component analysis. The numerical simulation shows that compared with the MUSIC method9 this method can detect and estimate the multiple moving sound objects in 3-D space more accurately. This provides a new method for smart mines to detect and track targets.
出处
《解放军理工大学学报（自然科学版）》
EI
北大核心
2013年第3期283-287,共5页
Journal of PLA University of Science and Technology(Natural Science Edition)
基金
中国博士后科学基金资助项目(20200471823)
关键词
独立分量分析
波达方向
运动目标
圆型阵列
independent component analysis （ICA）
direction of arrival （DOA）
moving objects
round array
