Title :
A parallel AR spectral estimation using a new class of filter bank
Author :
Nishikawa, Kiyoshi ; Miyazaki, Kyoko ; Kiya, Hitoshi
Author_Institution :
Nippon Steel Co., Kanagawa, Japan
Abstract :
It is reconfirmed that the aliasing terms cannot be cancelled under the ordinary perfect reconstruction filter bank conditions and therefore the perfect reconstruction filter banks cannot be used for the purpose. To realize parallel spectral estimation, the sufficient filter bank conditions are derived and the general procedure for parallel estimation is shown. Then the proposed method is applied to the Levinson-Durbin algorithm and it is demonstrated that the method can be implemented effectively with the algorithm.<>
Keywords :
antialiasing; digital filters; parallel processing; parameter estimation; sampled data systems; spectral analysis; Levinson-Durbin algorithm; aliasing; autoregressive spectral estimation; parallel estimation; sufficient filter bank conditions;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1993. ICASSP-93., 1993 IEEE International Conference on
Conference_Location :
Minneapolis, MN, USA
Print_ISBN :
0-7803-7402-9
DOI :
10.1109/ICASSP.1993.319640