Adaptive channel equalization for high-speed train

Author

Liu, Weilin ; Romanens, Hervê

Author_Institution

ASCOM Tech. Ltd., Magenwil, Switzerland

fYear

1994

fDate

8-10 Jun 1994

Firstpage

225

Abstract

The paper deals with the performance of the GSM physical layer for the railway application with a possible train speed up to 500 km/h. An adaptive Viterbi equalizer with two channel tracking approaches has been considered. The first is the conventional least mean square and the second is a simplified Kalman predictor. Simulation results demonstrate the necessity of channel adaption for a high speed train. They also show that for almost comparable complexity, simplified Kalman predictor considerably outperforms the LMS due to the utilization of the channel second-order statistics

Keywords

Kalman filters; adaptive equalisers; cellular radio; computational complexity; digital radio; least mean squares methods; maximum likelihood estimation; minimum shift keying; prediction theory; railways; telecommunication channels; time division multiple access; 900 MHz; GSM physical layer; adaptive Viterbi equalizer; adaptive channel equalization; channel adaption; channel tracking approach; complexity; high-speed train; least mean square; performance; second-order statistics; simplified Kalman predictor; Adaptive equalizers; GSM; Kalman filters; Least squares approximation; Maximum likelihood estimation; Pulse shaping methods; Rail transportation; Resonance light scattering; Statistics; Viterbi algorithm;

fLanguage

English

Publisher

ieee

Conference_Titel

Vehicular Technology Conference, 1994 IEEE 44th

Conference_Location

Stockholm

ISSN

1090-3038

Print_ISBN

0-7803-1927-3

Type

conf

DOI

10.1109/VETEC.1994.345132

Filename

345132