Node Selection for Belief Propagation Based Channel Equalization

Mitsuyoshi HAGIWARA; Toshihiko NISHIMURA; Takeo OHGANE; Yasutaka OGAWA

doi:10.1587/transcom.2016FGP0015

Node Selection for Belief Propagation Based Channel Equalization

Mitsuyoshi HAGIWARA, Toshihiko NISHIMURA, Takeo OHGANE, Yasutaka OGAWA

Full Text Views

0

Share
Cite this

Summary :

Recently, much progress has been made in the study of belief propagation (BP) based signal detection with large-scale factor graphs. When we apply the BP algorithm to equalization in a SISO multipath channel, the corresponding factor graph has many short loops and patterns in an edge connection/strength. Thus, proper convergence may not be achieved. In general, the log-likelihood ratio (LLR) oscillates in ill-converged cases. Therefore, LLR oscillation avoidance is important for BP-based equalization. In this paper, we propose applying node selection (NS) to prevent the LLR from oscillating. The NS extends the loop length virtually by a serial LLR update. Thus, some performance improvement is expected. Simulation results show that the error floor is significantly reduced by NS in the uncoded case and that the NS works very well in the coded case.

Publication: IEICE TRANSACTIONS on Communications Vol.E100-B No.8 pp.1285-1292

Publication Date: 2017/08/01

Publicized: 2017/02/08

Online ISSN: 1745-1345

DOI: 10.1587/transcom.2016FGP0015

Type of Manuscript: Special Section PAPER (Special Section on Radio Access Technologies for 5G Mobile Communications System)

Category: Wireless Communication Technologies

Authors

Mitsuyoshi HAGIWARA
  Hokkaido University
Toshihiko NISHIMURA
  Hokkaido University
Takeo OHGANE
  Hokkaido University
Yasutaka OGAWA
  Hokkaido University

Keyword

belief propagation algorithm, log-likelihood ratio, interference canceller, equalization

Cite this

Copy

Mitsuyoshi HAGIWARA, Toshihiko NISHIMURA, Takeo OHGANE, Yasutaka OGAWA, "Node Selection for Belief Propagation Based Channel Equalization" in IEICE TRANSACTIONS on Communications, vol. E100-B, no. 8, pp. 1285-1292, August 2017, doi: 10.1587/transcom.2016FGP0015.
Abstract: Recently, much progress has been made in the study of belief propagation (BP) based signal detection with large-scale factor graphs. When we apply the BP algorithm to equalization in a SISO multipath channel, the corresponding factor graph has many short loops and patterns in an edge connection/strength. Thus, proper convergence may not be achieved. In general, the log-likelihood ratio (LLR) oscillates in ill-converged cases. Therefore, LLR oscillation avoidance is important for BP-based equalization. In this paper, we propose applying node selection (NS) to prevent the LLR from oscillating. The NS extends the loop length virtually by a serial LLR update. Thus, some performance improvement is expected. Simulation results show that the error floor is significantly reduced by NS in the uncoded case and that the NS works very well in the coded case.
URL: https://globals.ieice.org/en_transactions/communications/10.1587/transcom.2016FGP0015/_p

Copy

@ARTICLE{e100-b_8_1285,
author={Mitsuyoshi HAGIWARA, Toshihiko NISHIMURA, Takeo OHGANE, Yasutaka OGAWA, },
journal={IEICE TRANSACTIONS on Communications},
title={Node Selection for Belief Propagation Based Channel Equalization},
year={2017},
volume={E100-B},
number={8},
pages={1285-1292},
abstract={Recently, much progress has been made in the study of belief propagation (BP) based signal detection with large-scale factor graphs. When we apply the BP algorithm to equalization in a SISO multipath channel, the corresponding factor graph has many short loops and patterns in an edge connection/strength. Thus, proper convergence may not be achieved. In general, the log-likelihood ratio (LLR) oscillates in ill-converged cases. Therefore, LLR oscillation avoidance is important for BP-based equalization. In this paper, we propose applying node selection (NS) to prevent the LLR from oscillating. The NS extends the loop length virtually by a serial LLR update. Thus, some performance improvement is expected. Simulation results show that the error floor is significantly reduced by NS in the uncoded case and that the NS works very well in the coded case.},
keywords={},
doi={10.1587/transcom.2016FGP0015},
ISSN={1745-1345},
month={August},}

Copy

TY - JOUR
TI - Node Selection for Belief Propagation Based Channel Equalization
T2 - IEICE TRANSACTIONS on Communications
SP - 1285
EP - 1292
AU - Mitsuyoshi HAGIWARA
AU - Toshihiko NISHIMURA
AU - Takeo OHGANE
AU - Yasutaka OGAWA
PY - 2017
DO - 10.1587/transcom.2016FGP0015
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E100-B
IS - 8
JA - IEICE TRANSACTIONS on Communications
Y1 - August 2017
AB - Recently, much progress has been made in the study of belief propagation (BP) based signal detection with large-scale factor graphs. When we apply the BP algorithm to equalization in a SISO multipath channel, the corresponding factor graph has many short loops and patterns in an edge connection/strength. Thus, proper convergence may not be achieved. In general, the log-likelihood ratio (LLR) oscillates in ill-converged cases. Therefore, LLR oscillation avoidance is important for BP-based equalization. In this paper, we propose applying node selection (NS) to prevent the LLR from oscillating. The NS extends the loop length virtually by a serial LLR update. Thus, some performance improvement is expected. Simulation results show that the error floor is significantly reduced by NS in the uncoded case and that the NS works very well in the coded case.
ER -