A FAST LEAST-SQUARE DECONVOLUTION ALGORITHM FOR VOCAL-TRACT CROSS-SECTION ESTIMATION

被引：4
|
作者：
HU, YH
MILENKOVIC, PH
机构：
[1] Department of Electrical and Computer Engineering, University of Wisconsin-Madison
来源：
IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING | 1990年 / 38卷 / 06期
基金：
美国国家卫生研究院; 美国国家科学基金会;
关键词：
D O I：
10.1109/29.56053
中图分类号：
O42 [声学];
学科分类号：
070206 ; 082403 ;
摘要：
An efficient computation algorithm, based on unitary Jacobi-type rotations, is developed in this paper for fast least-square deconvolution. Specifically, this algorithm is able to solve a positive definite covariance matrix with a special Toeplitz structure in 0(N2) operations instead of 0(N3) operations. Taking advantage of the inherent structure of the underlying deconvolution problem, this algorithm is guaranteed to be numerically stable since only unitary transformation is used. It has been implemented on an experimental human vocal tract cross section estimation system. Significant gain in processing speed has been observed. © 1990 IEEE
引用
收藏
页码：921 / 924
页数：4
相关论文
共 34 条
1234