Title :
Minimization of quantization noise amplification in biorthogonal subband coders
Author :
Makur, Anamitra ; Arunkumar, M.
Author_Institution :
Dept. of Electr. & Commun. Eng., Indian Inst. of Sci., Bangalore, India
Abstract :
Quantization noise amplification (QNA) restricts the coding gain (CG) in biorthogonal subband coders. Here, a coloring filter is introduced to color the quantization noise. The optimal coloring filter eliminating/minimizing QNA, with or without order restriction, is found for a given finite-impulse response filter bank (FB). An efficient implementation of the coloring filter is proposed. With the coloring filter, the optimal biorthogonal ideal FB becomes the full whitening coder achieving maximum possible CG. Results verify the CG improvement due to coloring for existing FBs.
Keywords :
FIR filters; channel bank filters; encoding; filtering theory; minimisation; noise; quantisation (signal); biorthogonal subband coders; filter bank; finite-impulse response; minimization; optimal coloring filter bank; quantization noise amplification; subband coding gain; Additive white noise; Band pass filters; Character generation; Colored noise; Filter bank; Iterative algorithms; Linear matrix inequalities; Noise shaping; Passband; Quantization; CG; Quantization noise; subband coding gain;
Journal_Title :
Circuits and Systems I: Regular Papers, IEEE Transactions on
DOI :
10.1109/TCSI.2004.835660