Parameter estimation and equalization Techniques for communication channels with multipath and multiple frequency offsets

Sajid Ahmed*, S. Lambotharan, A. Jakobsson, J. A. Chambers

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

29 Scopus citations

Abstract

We consider estimation of frequency offset (FO) and equalization of a wireless communication channel, within a general framework which allows for different frequency offsets for various multipaths. Such a scenario may arise due to different Doppler shifts associated with various multipaths, or in situations where multiple basestations are used to transmit identical information. For this general framework, we propose an approximative maximum-likelihood estimator exploiting the correlation property of the transmitted pilot signal. We further show that the conventional minimum mean-square error equalizer is computationally cumbersome, as the effective channel-convolution matrix changes deterministically between symbols, due to the multiple FOs. Exploiting the structural property of these variations, we propose a computationally efficient recursive algorithm for the equalizer design. Simulation results show that the proposed estimator is statistically efficient, as the mean-square estimation error attains the Cramér-Rao lower bound. Further, we show via extensive simulations that our proposed scheme significantly outperforms equalizers not employing FO estimation.

Original languageEnglish (US)
Pages (from-to)219-223
Number of pages5
JournalIEEE Transactions on Communications
Volume53
Issue number2
DOIs
StatePublished - Feb 1 2005

Keywords

  • Frequency offset (FO) estimation
  • Minimum mean-square error (MMSE) equalizer and complexity
  • Wireless multipath channels

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'Parameter estimation and equalization Techniques for communication channels with multipath and multiple frequency offsets'. Together they form a unique fingerprint.

Cite this