A new two-dimensional parallel block adaptive filter with reduced computational complexity

Author

Kinjo, Shigenori ; Oshiro, Masafumi ; Ochi, Hiroshi

Author_Institution

Dept. of Electr. & Electron. Eng., Ryukyus Univ., Okinawa, Japan

Volume

3

fYear

1997

Firstpage

2305

Abstract

Two-dimensional (2-D) adaptive digital filters (ADFs) for 2-D signal processing have become a fascinating area of the adaptive signal processing. However, conventional 2-D FIR ADFs require a lot of computations. For example, the TDLMS requires 2N² multiplications per pixel. We propose a new 2-D adaptive filter using FFTs. The proposed adaptive filter carries out fast convolution using the overlap-save method, and has a parallel structure. Thus, we can reduce the computational complexity to O(log₂N) per pixel

Keywords

FIR filters; adaptive filters; adaptive signal processing; computational complexity; convolution; fast Fourier transforms; two-dimensional digital filters; 2D adaptive digital filters; 2D adaptive filter; 2D signal processing; FFT; adaptive signal processing; convolution; overlap-save method; parallel structure; reduced computational complexity; two-dimensional parallel block adaptive filter; Adaptive filters; Adaptive signal processing; Computational complexity; Convolution; Cost function; Digital filters; Finite impulse response filter; Signal processing algorithms; System identification; Two dimensional displays;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1997. ICASSP-97., 1997 IEEE International Conference on

ISSN

1520-6149

Print_ISBN

0-8186-7919-0

Type

conf

DOI

10.1109/ICASSP.1997.599513

Filename

599513