Title :
Soft-decoding vector quantizer using reliability information from turbo-codes
Author_Institution :
Dept. of Inf. Eng., Chinese Univ. of Hong Kong, Shatin, Hong Kong
fDate :
7/1/1999 12:00:00 AM
Abstract :
The optimum soft-decoding vector quantizer using the reliability information from turbo-codes is derived for combined source-channel coding. The encoder and decoder of the quantizer are optimized iteratively. For a four-dimensional vector quantizer having a rate of 1 bit/sample transmitted through a noisy channel, the soft-decoding channel-optimized quantizer can achieve about 3-3.7 dB performance improvement over conventional source-optimized quantizer.
Keywords :
combined source-channel coding; decoding; least mean squares methods; optimisation; turbo codes; vector quantisation; Gauss-Markov source; MMSE estimator; channel-optimized quantizer; combined source-channel coding; four-dimensional vector quantizer; noisy channel; optimum soft-decoding; performance improvement; reliability information; soft-decoding vector quantizer; source-optimized quantizer; turbo-codes; Concatenated codes; Councils; Distortion; Gaussian processes; Iterative algorithms; Iterative decoding; Noise robustness; Probability; Turbo codes; Vector quantization;
Journal_Title :
Communications Letters, IEEE
DOI :
10.1109/4234.775256
