A novel channel estimation technique for MIMO-OFDM systems for Frequency Selective Ricean channel

The MIMO system convincingly promises a better bit rate in comparison with a SISO system. The system gets even efficient when OFDM is implemented with MIMO to obtain better data rates even in hostile channel environment. Research establishes OFDM as a potential candidate for use in MIMO systems for the reasons like inherent FFT, waveform adaptation, advanced antenna techniques, multiple access, interoperability, etc. Since, channel estimation is an integral part of OFDM system; it stands to demand a technique that can offer accurate channel characterization for effective equalization to follow. In addition, there exists a tradeoff between accuracy of estimation and complexity of the technique. In this paper, a combined channel estimation technique, which combines the virtues of Least Squares and LMMSE estimators with SVD in MIMO-OFDM systems. This combined strategy increases the estimate accuracy without much increase in complexity or computation considering a Ricean Frequency Selective channel environment. The proposed technique is explicated with necessary mathematics and the theory is well substantiated with simulations. The technique is fitted into the OFDM framework and the simulation results indicate a good estimate of the channel. The theory and results are presented and discussed in the paper.