A Tensor-Based Subspace Method for Blind Estimation of MIMO Channels

In this paper, we introduce a tensor-based subspace method for solving the blind channel estimation problem in a multiple-input multiple-output (MIMO) system. The current subspace methods of blind channel estimation require stacking the multidimensional measurement data into one highly structured vector and estimate the signal subspace via a singular value decomposition (SVD) of the correlation matrix of the measurement data. In contrast to this, we propose a 3-way measurement tensor to exploit the structure inherent in the measurement data and introduce a Higher-Order SVD (HOSVD) to obtain the signal subspace. This tensor-based subspace estimate is an improved estimate of the signal subspace, thereby leading to an improved estimate of the system channels. Numerical simulations demonstrate that the proposed method outperforms the current subspace based blind channel estimation methods in terms of the channel estimation accuracy. Furthermore, we show that the accuracy of the estimations is significantly improved by employing overlapping observed data windows at the receiver.