Performance Analysis of MIMO OFDM ML Detection in the Presence of Channel Estimation Error

In this paper, the effects of channel estimation errors on the performance of a maximum likelihood detector in a MIMO OFDM system are assessed. The analysis is conducted in the framework of a Long Term Evolution air interface and is based on a robust pilot aided channel estimator. To this end the correlation coefficient between the estimate and the real channel is computed and the impact of interpolation errors on the detector is evaluated as a function of key design parameters such as the pilot spacing, the filter length, and the number of transmit antennas. The effect of partial knowledge of the channel statistics at the receiver side is also studied.