Title :
Designing the fast filter bank with a minimum complexity criterion
Author :
Wei, Lee Jun ; Ching, Lim Yong
Author_Institution :
Dept. of Electr. & Comput. Eng., Nat. Univ. of Singapore, Singapore
Abstract :
The fast filter bank (FFB) is an extremely efficient single-rate filter bank due to the sparse population of non-zero coefficients in the subfilters. Its subfilters are designed and operated based on the frequency response masking (FRM) technique. The node-modulated fast filter bank (nm-FFB) was proposed as a modification to the fast filter bank to further reduce its complexity when processing real-input, real output signals. An optimal implementation of the nm-FFB can be achieved by shifting the terminal stage at which node-modulation is performed. Such an implementation is extended to the complex signal case and found to provide a lower complexity when compared to the original FFB implementation.
Keywords :
computational complexity; filtering theory; frequency response; hearing; fast filter bank; frequency response masking; minimum complexity criterion; node-modulated fast filter bank; single-rate filter bank; Attenuation; Design optimization; Filter bank; Fixed-point arithmetic; Frequency response; Linear programming; Quantization; Radar applications; Radar signal processing; Signal processing;
Conference_Titel :
Signal Processing and Its Applications, 2003. Proceedings. Seventh International Symposium on
Print_ISBN :
0-7803-7946-2
DOI :
10.1109/ISSPA.2003.1224868