Recursive Channel Estimation Based on Finite Parameter Model Using Reduced-Complexity Maximum Likelihood Equalizer for OFDM over Doubly-Selective Channels

Kok Ann Donny TEO  Shuichi OHNO  Takao HINAMOTO  

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E88-A    No.11    pp.3076-3084
Publication Date: 2005/11/01
Online ISSN: 
DOI: 10.1093/ietfec/e88-a.11.3076
Print ISSN: 0916-8508
Type of Manuscript: Special Section PAPER (Special Section on Wide Band Systems)
polynomial model,  oversampled basis expansion model,  recursive Kalman,  reduced-complexity ML equalizer,  

Full Text: PDF(677KB)>>
Buy this Article

To take intercarrier interference (ICI) attributed to time variations of the channel into consideration, the time- and frequency-selective (doubly-selective) channel is parameterized by a finite parameter model. By capitalizing on the finite parameter model to approximate the doubly-selective channel, a Kalman filter is developed for channel estimation. The ICI suppressing, reduced-complexity Viterbi-type Maximum Likelihood (RML) equalizer is incorporated into the Kalman filter for recursive channel tracking and equalization to improve the system performance. An enhancement in the channel tracking ability is validated by theoretical analysis, and a significant improvement in BER performance using the channel estimates obtained by the recursive channel estimation method is verified by Monte-Carlo simulations.