Time-domain ML channel estimation considering antenna switching time for RF-MIMO-OFDM

Abstract

In this paper, we propose a channel estimation method for radio frequency multiple-input multiple-output orthogonal frequency division multiplexing (RF-MIMO-OFDM) systems. In RF-MIMO-OFDM systems, a receiver combines the received signals at the RF front-end. This implies that a receiver can estimate only one of MIMO channels during one OFDM symbol duration. Therefore, the conventional channel estimation method requires a number of pilot OFDM symbols in proportion to the numbers of transmit antennas and receive antennas. If the channels are varying during the pilot transmission, the accuracy of the conventional channel estimation is deteriorated. In this paper, we propose a channel estimation method which can estimate channels with a shorter time period considering the antenna switching time. Our proposed method uses time-domain maximum likelihood estimation and pilot symbols which have a repetitive waveform within one OFDM symbol duration. The proposed method estimates channel impulse responses (CIRs) of multiple channels during one OFDM symbol duration. From simulation results, we show that the proposed estimation method achieves a better estimation performance than the conventional estimation method. Furthermore, the error performance is also improved by the use of the proposed method.