Skip to Main Content
This paper presented a semi-blind channel estimation algorithm for MIMO-OFDM system which had 2 transmitting and 2 receiving antennas. This scheme used least square to estimate the initial channel with the received training symbol in the frequency domain. But the precondition is assumed that the channel characteristic is identical between the two symbol periods. However, the channel characteristic is always changing in fading channel, which make the above channel estimation is not veracious. So the vector Kalman filter was used to adjust the initial estimated channel error. It is shown that the performance of system is improved highly than before. On the other hand, the complexity of this method is lower than the least square algorithm in timing domain.