Recently, a cluster map based blind RBF equalizer (CM-BRE) has been proposed. By utilizing the underlying structure characteristics of RBF equalizer, the CM-BRE can be implemented by the combination of neural-gas algorithm (NGA) with several sorting operations. Although the CM-BRE is able to achieve almost identical performance with the optimal RBF equalizer, the high computational load mainly caused by NGA limits it's application. In this paper, we propose a downsizing method that employs the inter-relation among RBF centers and significantly reduces the NGA's computational load. Furthermore, a method to determine the feedback vector is derived, then CM-BRE is extended to a cluster map based blind RBF decision feedback equalizer (CM-BRDFE). The proposed CM-BRDFE also shows the close performance with the optimal RBF decision feedback equalizer in simulations.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Hai LIN, Katsumi YAMASHITA, "A Cluster Map Based Blind RBF Decision Feedback Equalizer with Reduced Computational Complexity" in IEICE TRANSACTIONS on Fundamentals,
vol. E87-A, no. 10, pp. 2755-2760, October 2004, doi: .
Abstract: Recently, a cluster map based blind RBF equalizer (CM-BRE) has been proposed. By utilizing the underlying structure characteristics of RBF equalizer, the CM-BRE can be implemented by the combination of neural-gas algorithm (NGA) with several sorting operations. Although the CM-BRE is able to achieve almost identical performance with the optimal RBF equalizer, the high computational load mainly caused by NGA limits it's application. In this paper, we propose a downsizing method that employs the inter-relation among RBF centers and significantly reduces the NGA's computational load. Furthermore, a method to determine the feedback vector is derived, then CM-BRE is extended to a cluster map based blind RBF decision feedback equalizer (CM-BRDFE). The proposed CM-BRDFE also shows the close performance with the optimal RBF decision feedback equalizer in simulations.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/e87-a_10_2755/_p
Copy
@ARTICLE{e87-a_10_2755,
author={Hai LIN, Katsumi YAMASHITA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={A Cluster Map Based Blind RBF Decision Feedback Equalizer with Reduced Computational Complexity},
year={2004},
volume={E87-A},
number={10},
pages={2755-2760},
abstract={Recently, a cluster map based blind RBF equalizer (CM-BRE) has been proposed. By utilizing the underlying structure characteristics of RBF equalizer, the CM-BRE can be implemented by the combination of neural-gas algorithm (NGA) with several sorting operations. Although the CM-BRE is able to achieve almost identical performance with the optimal RBF equalizer, the high computational load mainly caused by NGA limits it's application. In this paper, we propose a downsizing method that employs the inter-relation among RBF centers and significantly reduces the NGA's computational load. Furthermore, a method to determine the feedback vector is derived, then CM-BRE is extended to a cluster map based blind RBF decision feedback equalizer (CM-BRDFE). The proposed CM-BRDFE also shows the close performance with the optimal RBF decision feedback equalizer in simulations.},
keywords={},
doi={},
ISSN={},
month={October},}
Copy
TY - JOUR
TI - A Cluster Map Based Blind RBF Decision Feedback Equalizer with Reduced Computational Complexity
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 2755
EP - 2760
AU - Hai LIN
AU - Katsumi YAMASHITA
PY - 2004
DO -
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E87-A
IS - 10
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - October 2004
AB - Recently, a cluster map based blind RBF equalizer (CM-BRE) has been proposed. By utilizing the underlying structure characteristics of RBF equalizer, the CM-BRE can be implemented by the combination of neural-gas algorithm (NGA) with several sorting operations. Although the CM-BRE is able to achieve almost identical performance with the optimal RBF equalizer, the high computational load mainly caused by NGA limits it's application. In this paper, we propose a downsizing method that employs the inter-relation among RBF centers and significantly reduces the NGA's computational load. Furthermore, a method to determine the feedback vector is derived, then CM-BRE is extended to a cluster map based blind RBF decision feedback equalizer (CM-BRDFE). The proposed CM-BRDFE also shows the close performance with the optimal RBF decision feedback equalizer in simulations.
ER -