A Novel Approach for Blind Estimation of a MIMO Channel
Including Phase Unwrapping Ambiguity Elimination

Xia Liu¹, Hui Miao², and Xiaodi Huang^2*

¹Kordia Solutions Pty Ltd.
Banyo, Australia, QLD 4014
leo.liu@kordia.com.au

²School of Computing and Mathematics
Chales Sturt University
Albury, Australia, NSW 2640
mzzhmh@gmail.com, xhuang@csu.edu.au

Abstract

In this paper, a novel method for blind estimation of a multiple input multiple output (MIMO) channel is described. Opposite to the conventional blind estimation algorithms that use singular value decomposition (SVD) or eigenvalue decomposition (EVD), it employs a simple coding scheme and a square root algorithm to recover the channel state information (CSI). The block coding scheme accompanying the proposed estimation approach requires a block encoder only at the transmitter. The proposed block coding scheme is spectral efficient as it offers the full coding rate when the numbers of transmitting and receiving antennas are equal. When only two transmitting and two receiving antennas are used, the coding scheme reduces the noise power to half of the original noise power. The phase ambiguity accompanying the blind approach to channel estimation is eliminated with only one symbol pilot sequence.

Keywords: MIMO system, Channel Estimation, Phase Ambiguity Elimination, Singular Value Decomposition, Eigenvalue Decomposition.

*Corresponding Author: Web: http://csusap.csu.edu.au/~xhuang/, Tel: +61-02-60519652

IT Convergence Practice (INPRA), Vol. 1, No. 2, pp. 20-33, June 2013 [pdf]