SVD-Based Linear Precoding Using Channel Estimation for MIMO-OFDM Systems

Abstract

Orthogonal Frequency Division Multiplexing is an approach that can be used to combat the effect of frequency selective fading channels. Apart from spectral efficiency, reliability over fast fading channels is a primary concern in OFDM. MIMO systems offer a solution to this problem owing to the high capacity and beamforming gain. The integration of MIMO with the traditional OFDM systems allow parallel transmission of data with a significant reduction in inter-carrier interference. The tradeoff to be considered is the increase in complexity of the user equipment. This is due to the advanced modulation and estimation operations performed at the receiver. In this paper, we present a transmitter site channel estimation scenario with the help of pilot symbols. Further, a MIMO-OFDM transceiver is designed which utilizes the information of the channel state at the transmitter to perform Singular Value Decomposition (SVD) based linear precoding on the data. Such a type of precoding scheme provides reduction in bit error rate to an optimum range. The simulation results show that a singular value-based precoding technique performs better than the traditional Alamouti space–time block code. But the requirement of additional time slot for channel estimation is a challenge that accompanies this technique.