Adaptive Multiuser Turbo Receiver Structure for Oversaturated MIMO Communication System

Abstract

In this paper we show that an adaptive turbo (iterative) multiuser receiver employing imperfect decision feedback compared to perfect decision feedback has a signal to interference ratio (SINR) gain of over 1 dB at bit error rate (BER) of 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-2</sup> for an oversaturated multiuser multicellular wireless communication system. A (turbo) iterative multiuser receiver system consists of a bank of adaptive linear minimum mean square error (MMSE) channel estimators that retrieves user signatures followed by non-linear MMSE detectors and dedicated log maximum a-posteriori (MAP) decoders to decode user information in an uplink multiuser communication system. We find that estimating filter coefficients every turbo iteration (successive training) assuming perfect feedback does not provide any SINR gain over non successive training based adaptive multiuser receiver. We show that there is a gain of over 1.5 dB in using non-successive training over successive training at BER of 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-2</sup> for an oversaturated multiple input multiple output (MIMO) communication system