Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

On the block size of trellis quantizers

Eriksson, Tomas LU ; Hellerbrand, S ; Anderson, John B LU and Novak, Mirek LU (2005) Proceedings. DCC 2005. Data Compression Conference p.457-457
Abstract
Summary form only given. In this paper, we examine the effect of block size on the performance of trellis based quantization. In particular, the Viterbi and tailbiting BCJR algorithms are compared. It is shown that for short blocks of data, the T-BCJR algorithm achieves a superior performance over the Viterbi algorithm (VA). One approach is to use the maximum a posteriori probability (MAP) heuristic and the T-BCJR algorithm. If the MAP-encoder does not produce a tailbiting state sequence, the path is modified for a number of stages at the beginning and end of the block such that it tailbites. The enclosed figure shows MSE as a function of block size and sample position, respectively, for a rate R=1 bit per sample, 32-state trellis... (More)
Summary form only given. In this paper, we examine the effect of block size on the performance of trellis based quantization. In particular, the Viterbi and tailbiting BCJR algorithms are compared. It is shown that for short blocks of data, the T-BCJR algorithm achieves a superior performance over the Viterbi algorithm (VA). One approach is to use the maximum a posteriori probability (MAP) heuristic and the T-BCJR algorithm. If the MAP-encoder does not produce a tailbiting state sequence, the path is modified for a number of stages at the beginning and end of the block such that it tailbites. The enclosed figure shows MSE as a function of block size and sample position, respectively, for a rate R=1 bit per sample, 32-state trellis quantizer and an IID Gaussian source. The effects of start-up are clearly visible. For the T-BCJR algorithm, the distortion is evenly distributed across the whole block. The performance decrease for short blocks stems from the increase in the number of tailbiting violations for short blocks and the suboptimal modification to ensure tailbiting. The results presented here hold for a large class of trellis constructions, such as TCQ (Less)
Please use this url to cite or link to this publication:
author
; ; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
trellis based quantization, tailbiting BCJR algorithm, Viterbi algorithm, trellis quantizer block size effects, T-BCJR, MAP-encoder, maximum a posteriori probability heuristic, tailbiting state sequence, MSE, IID Gaussian source, start-up distortion effects, TCQ, sample position effects, tailbiting violations
host publication
Proceedings. DCC 2005. Data Compression Conference
pages
457 - 457
publisher
IEEE - Institute of Electrical and Electronics Engineers Inc.
conference name
Proceedings. DCC 2005. Data Compression Conference
conference location
Snowbird, UT, United States
conference dates
2005-03-29 - 2005-03-31
external identifiers
  • wos:000229070000054
  • scopus:26944440089
ISBN
0-7695-2309-9
DOI
10.1109/DCC.2005.62
language
English
LU publication?
yes
id
71b15d30-ebce-411d-916d-95c50e25bd80 (old id 615929)
date added to LUP
2016-04-04 10:32:42
date last changed
2022-01-29 20:26:13
@inproceedings{71b15d30-ebce-411d-916d-95c50e25bd80,
  abstract     = {{Summary form only given. In this paper, we examine the effect of block size on the performance of trellis based quantization. In particular, the Viterbi and tailbiting BCJR algorithms are compared. It is shown that for short blocks of data, the T-BCJR algorithm achieves a superior performance over the Viterbi algorithm (VA). One approach is to use the maximum a posteriori probability (MAP) heuristic and the T-BCJR algorithm. If the MAP-encoder does not produce a tailbiting state sequence, the path is modified for a number of stages at the beginning and end of the block such that it tailbites. The enclosed figure shows MSE as a function of block size and sample position, respectively, for a rate R=1 bit per sample, 32-state trellis quantizer and an IID Gaussian source. The effects of start-up are clearly visible. For the T-BCJR algorithm, the distortion is evenly distributed across the whole block. The performance decrease for short blocks stems from the increase in the number of tailbiting violations for short blocks and the suboptimal modification to ensure tailbiting. The results presented here hold for a large class of trellis constructions, such as TCQ}},
  author       = {{Eriksson, Tomas and Hellerbrand, S and Anderson, John B and Novak, Mirek}},
  booktitle    = {{Proceedings. DCC 2005. Data Compression Conference}},
  isbn         = {{0-7695-2309-9}},
  keywords     = {{trellis based quantization; tailbiting BCJR algorithm; Viterbi algorithm; trellis quantizer block size effects; T-BCJR; MAP-encoder; maximum a posteriori probability heuristic; tailbiting state sequence; MSE; IID Gaussian source; start-up distortion effects; TCQ; sample position effects; tailbiting violations}},
  language     = {{eng}},
  pages        = {{457--457}},
  publisher    = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
  title        = {{On the block size of trellis quantizers}},
  url          = {{https://lup.lub.lu.se/search/files/5564084/635634.pdf}},
  doi          = {{10.1109/DCC.2005.62}},
  year         = {{2005}},
}