J4 ›› 2010, Vol. 37 ›› Issue (6): 999-1004.doi: 10.3969/j.issn.1001-2400.2010.06.004

• Original Articles • Previous Articles     Next Articles

Novel estimation and equalization of time-variant channels for OFDM system

CHEN Dong-hua1;QIU Hong-bing2   

  1. (1. School of Telecommunication Engineering, Xidian Univ., Xi'an  710071, China;
    2. Info. and Communication College, Guilin Univ. of Electronic Tech., Guilin  541004, China)
  • Received:2010-12-13 Online:2010-12-20 Published:2011-01-22
  • Contact: CHEN Dong-hua E-mail:dhchen@hqu.edu.cn

Abstract:

This paper considers the orthogonal frequency division multiplexing (OFDM) transmission operated in highly mobile channels. We propose a time-domain channel estimator based on the Karhunen-Loeve basis expansion model (KL-BEM). The proposed estimator adopts the KL-BEM to approximate the channels' time variation within several OFDM symbols. The KL-BEM coefficients are estimated by using a group of short-time pilot clusters that are inserted between two successive OFDM symbols. The properly designed pilot clusters pattern enables accurate channel estimation and higher throughput. Following this, we adopt an improved linear minimum mean squared error equalizer with sorted serial interference cancellation (LMMSE-SIC) to recover the transmit symbol. Preceding the LMMSE-SIC, time-domain windowing is applied to shorten the intercarrier interference (ICI) support, thereby improving the equalization performance and reducing the complexity as well. Simulation results show that the proposed scheme can significantly improve the system performance in highly mobile channels.

Key words: orthogonal frequency division multiplexing, time varying channel estimation, equalizer, basis expansion model