EI / SCOPUS / CSCD 收录

中文核心期刊

依赖于频率变化模型的相干信号子空间聚焦处理方法

Coherent signal-subspace focusing to frequency dependent-modeling for direction of arrival estimation of wide-band sources

  • 摘要: 提出了一种依赖于频率变化模型的相干信号子空间聚焦处理新方法。相干信号子空间(CSS)宽带处理方法的核心思想是通过聚焦变换将宽带信号各频率分量的信号子空间映射成为一个信号子空间。而依赖于频率变化模型(FDM)的宽带处理方法则是通过引入一组关于频率的基函数,建立阵列信号依赖于频率变化的模型,从而直接进行宽带处理。本文把CSS的聚焦思想应用到FDM,提出了一种宽带目标方位估计方法,并构造了对角聚焦矩阵。新方法采用了FDM,增大了相关矩阵的维数,有效地反映了噪声子空间,因此比CSS具有更好的性能;采用聚焦,大大减小了FDM方法庞大的运算量。计算机仿真结果验证了新方法的有效性。

     

    Abstract: A new method of direction of arrival(DOA) estimation of wide-band sources by coherent signal-subspace focusing to frequency-dependent modeling (CSSFDM) is proposed. The coherent signal-subspace(CSS)2 was proposed by H.Wang and M.Kaveh in 1985. Its key point is the focusing which maps the signal subspaces of the narrowband components of the wide band sources onto a single signal subspace. Y.Grenier set up the frequency-dependent modeling(FDM)13 of array signals by introducing the basic function into array signals to process directly the wide band signals in 1994. We present a new DOA estimation method that combines the coherent signal-subspace focusing and the frequency-dependent modeling, and construct its diagonal focusing matrix. Because of extending the order of correlation matrix of array signals and presenting effectively noise subspace by FDM, the method shows the better performance than CSS. By means of the focusing of CSS, it reduces obviously computational complexity of the spatial spectrum of FDM. The effectiveness of the new method is confirmed by computer simulation.

     

/

返回文章
返回