Title :
A Fast Sea Interface Reverberation Suppression Method Based on Unitary Transformation
Author :
Li Yue-qin ; Li Jin-ping ; Han Lei
Author_Institution :
Inf. Coll., Beijing Union Univ., Beijing, China
Abstract :
The low-rank structure reverberation suppression algorithm is a better method to restrain the sea interface reverberation. But the data processing speed of the method is very slowly due to the complexity of singular value decomposition. It can not meet the real-time requirement in actual application. This article present a signal subspace fast approximation method based on unitary transformation which can estimate the reverberation subspace rapidly. This method transform array data matrix to real matrix using the center symmetry of uniform linear array and unitary transformation at first. Then it calculates singular value and eigenvalue decomposition of the real matrix. Signal subspace can be estimated using the method of rational approximation finally. The method´s computation is little comparatively and can improved the operation speed greatly. The conclusions of the experiments show that the algorithm has a higher operation speed meanwhile suppress the reverberation effectively.
Keywords :
acoustic signal detection; acoustic waves; eigenvalues and eigenfunctions; interference suppression; oceanographic techniques; reverberation; singular value decomposition; underwater acoustic communication; underwater sound; array data matrix; data processing speed; eigenvalue decomposition; low-rank structure reverberation suppression algorithm; rational approximation; real matrix; reverberation subspace estimation; sea interface reverberation suppression; signal subspace fast approximation method; singular value decomposition; uniform linear array; unitary transformation; eigenvalue; sea interface reverberation; singular value decomposition; subspace; unitary transformation;
Conference_Titel :
Digital Manufacturing and Automation (ICDMA), 2010 International Conference on
Conference_Location :
ChangSha
Print_ISBN :
978-0-7695-4286-7
DOI :
10.1109/ICDMA.2010.100