Determined semi-blind algorithm for channel estimation in space-time coded MIMO-OFDM systems

An Orthogonal frequency division multiplexing (OFDM) system with a Multiple input multiple output (MIMO) configuration is considered, which uses Space-time coding. A blind method is proposed for estimating the MIMO-OFDM system's channel responses, which uses the special processing after Space-Time encoding on part of sub-carriers to acquire the channels' estimates. The method is different with the blind estimators based on the second statistic information, which need a lot of receiver data blocks to achieve the channel estimation. Our solution is a determined algorithm of channel estimation, which can achieve good estimation performance for the higher signal noise ratio values with only two OFDM blocks. The algorithm is robust to the different kinds of modulation schemes. In the presence of noise, the algorithm has very low complexity-only single eigenvalue decomposition is needed, no matter how many OFDM blocks are used. One of the most desirable merits of this approach is its data efficiency-it can be implemented using two data blocks to achieve reasonably accurate estimate, and this property is independent of the decoder or finite alphabet property. Finally, we demonstrate validity of the algorithm by some simulations.