Masahiro Uno

Introducing dual regularization parameters into regularized channel inversion (RCI)-based vector perturbation for modulo loss reduction

This paper proposes a method to reduce the symbol error caused by modulo operator for regularized channel inversion (RCI)-based vector perturbation (RCI-VP). The proposed method introduces dual regularization parameters of RCI precoding matrix. First regularization parameter a is set so as to maximize the signal to interference plus noise ratio (SINR) value of each stream to find the optimal perturbation vector. The second regularization parameter 0.2 is set so as to minimize the total MSE between the transmitted perturbation vector and received perturbation vector to reduce the symbol error caused by modulo operator. The simulated results confirm that the proposed method can improve the performance of RCI-VP over i.i.d. MIMO channel and frequency-selective fading MIMO channel.

A simple method of transmitting stream adaptation for fixed-matrix-size precoding on MU-MIMO transmission

During the operation of multiuser multiple-input-multiple output (MU-MIMO) algorithms, some antennas of BS or UEs may neither transmit nor receive signals due to the long distance or unexpected damages. Therefore, the BS cannot obtain the channel state information (CSI) from/to these antennas. Without these CSI, the performance of algorithm is degraded. We call this effect as `broken antenna effect (BAE). To cope with this issue, this paper proposes a process to regularize the channel matrix by replacing the missing CSI with zero(s) and setting nonzero values to diagonal elements. Then, it is confirmed through computer simulation that the proposed method can adjust the number of transmitting parallel streams according to the number of `broken antennas, and provide MU-MIMO transmission with maximum diversity order without any change of the precoding process.