Improved MIMO precoding for decentralized receivers resembling concepts from lattice reduction

Publisher Website
Google Scholar
Abstract
This paper presents an improved version of pre- coding, i.e., nonlinear preequalization, for situations with one central transmitter (e.g., base station) and a number of dis- tributed receivers (e.g., mobile terminals). The new method is based on Tomlinson-Harashima precoding, applied to multiple- input/multiple-output (MIMO) channels, and inspired by the concept of lattice-reduction-aided detection in MIMO commu- nication systems. We show that using the proposed technique, a considerable degree of diversity can be gained over linear preequalization, as well as conventional precoding. I. INTRODUCTION Recently it has been shown that Tomlinson-Harashima pre- coding (14), (12), originally proposed for the equalization of intersymbol-interference channels, is also suited for the suppression/cancellation of multiuser interference (3), (8), (1). In particular, in broadcast scenarios (downlink situation), where no joint processing of the receive signals is possible, "MIMO precoding" is attractive since it can be viewed as the counterpart to successive cancellation which is applicable in uplink scenarios only. Besides this, Yao and Wornell (16) proposed a novel scheme for improved detection of signals transmitted using multiple antenna systems. Using this scheme based on lattice (basis) reduction, significant gains over linear equalization and even the BLAST approach are possible. In particular, full diversity reception is achieved. In this paper, we present an improved precoding scheme for broadcast scenarios using the ideas from lattice-reduction- aided equalization strategies. For this, the lattice-theoretic decomposition of the channel matrix has to be modified suitably. The operation of this type of precoding is explained and results from numerical simulations are given to illustrate the performance gains obtained using the new method. The paper is organized as follows: In Section II the channel model is given and in Section III precoding for broadcast scenarios is briefly reviewed. The new strategy is presented in Section IV and simulation results are shown in Section V. Concluding, in Section VI some possible extensions are briefly addressed.

This publication has 11 references indexed in Scilit: