Damping Factor Analysis on Large Scale MIMO Detector Based on Message Passing

Alex Miyamoto Mussi, Taufik Abr˜ao

DOI: 10.14209/sbrt.2016.207

Evento: XXXIV Simpósio Brasileiro de Telecomunicações (SBrT2016)

Keywords: Low complexity MIMO detector message passing damping messages Markov random fields graphical models

Abstract

A message passing detector based on belief propagation (BP) algorithm for Markov random field (MRF) graphical model, named MRF-BP, is analysed under large scale MIMO scenarios. The contribution of this work consists of the analysis of message damping method, applied to such MRF-BP detector. A damping factor variation under different number of antennas configuration and SNR regions is considered; BER performance and computational complexity are evaluated over different scenarios. Numerical Results lead to a great performance gain with message damping with no extra computational complexity, although with low number of antennas the damping factor value needs carefully be chosen. Besides, based on the proposed analysis an optimal value for the damping factor is evaluated for different number of antennas scenarios.

Download