Linear Precoder Design with Spatial Multiplexing for MIMO System

Abstract

This paper presents linear precoder design with spatial multiplexing for multiple-input multiple-output (MIMO) system with perfect knowledge of channel state information at transmitter (CSIT) restoring the low complexity at both transmitter and receiver, subject to a total transmit power constraint. Assuming Rayleigh fading channel and using a maximum-likelihood detector at receiver, we show that the proposed precoder is a function of noise variance and the eigenvectors of the estimated channel matrix. The singular value decomposition (SVD) method and standard water filling algorithm are used for designing the precoder and beam power allocation respectively. Simulation results are provided to illustrate the significance performance gain achieved by using the proposed precoder. The precoder shows the bit error rate (BER) of 10^-5.5 at 2dB signal to noise ratio (SNR). The precoder may be promising in wireless data transmission application especially for worldwide interoperability for microwave access (WIMAX).