A closed form solution to semi-blind joint symbol and channel estimation in MIMO-OFDM systems

Due to the scarcity of the electromagnetic spectrum, multidimensional signaling schemes that take into account several signal dimensions such as space, time, frequency and constellation, are good candidates for increasing the data rate and/or improving the link reliability in future communication systems. Recently a new space-time-frequency diversity based MIMO-OFDM system has been proposed where transmit signal design combines frequency-domain Vandermonde spreading with a time-varying linear constellation precoding, while the received signal is formulated as a nested parallel factor (PARAFAC) model. A joint channel estimation and symbol decoding process has been developed for this system based on the alternating least squares (ALS) algorithm. In this paper, we propose a low-complexity blind receiver based on the least squares Khatri-Rao factorization (LS-KRF) for joint channel estimation and symbol decoding. Our proposed LS-KRF receiver is a closed-form solution which provides the same performance as that of the ALS solution while being less complex since no iteration is needed. Simulation results are included to verify the benefits of the proposed receiver.