Alex Miyamoto Mussi

Efficient Lattice Reduction Aided Detectors Under Realistic MIMO Channels

This contribution analyses the performance of efficient multiple-input-multiple-output (MIMO) detectors under correlated channels and imperfect coefficients channel estimation. A number of signal detection principles and techniques, including the minimum mean squared error detector with and without ordered successive interference cancellation; the sphere decoding MIMO detection, as well as promising near-orthogonal transformation techniques combined with these detectors, namely the lattice reduction and the QR decomposition are analysed under the perspective of complexity-performance tradeoff. While in most of available works perfect channel state information and uncorrelated channels have been considered, herein the complexity-performance tradeoff has been analysed and compared with the maximum likelihood (ML) limit under specific but practical scenarios of interest, namely: high spectral efficiency scenario; channel error estimates; channel/antenna correlation; combined channel errors and correlated channels. Under performance-complexity perspective, the optimum ML–MIMO detector is deployed as reference aiming to evaluate the efficiency and performance degradation of those sub-optimal MIMO detectors operating under hostile channel conditions.

SDR Lattice-Reduction-Aided Detector

Communication techniques based on multiple-input multiple-output (MIMO) have been heavily exploited in the past decade, in order to obtain high capacity communication systems, aim to dealing with increasing spectrum scarcity scenarios. In this context, the semidefinite relaxation (SDR) signal detection strategy is promising due to its very near-optimal maximum-likelihood (ML) performance combined with polynomial complexity. However, the SDR detector has proved insufficient for systems with high order modulation, exactly where the recent pre-detection technique by lattice reduction (LR) has demonstrated excellent results in terms of performance-complexity. This paper investigates the performance-complexity tradeoff of the SDR detector under high order modulation in uncorrelated MIMO channels aided by lattice reduction technique applied in the pre-detection stage.

Message passing detection for large-scale MIMO systems: damping factor analysis

A message passing detector based on belief propagation (BP) algorithm for Markov random fields (MRF-BP) and factor graph (FG-BP) graphical models is analysed under different large-scale (LS) multiple-input multiple-output (MIMO) scenarios, including system parameters, such as damping factor (DF), number of users and number of antennas, from N = 20 to 200 antennas. Specifically, the DF variation under different number of antennas configuration and signal-to-noise ratio (SNR) regions is extensively evaluated; bit error rate (BER) performance and computational complexity are assessed over different scenarios. Numerical results lead to a great performance gain with damped MRF-BP approach, overcoming FG-BP scheme in specific scenarios, with no extra computational complexity. Also, message damping (MD) method resulted in faster convergence of MRF-BP algorithm in LS scenarios, evidencing that, besides the performance gain, MD technique can lead to a computational complexity reduction. Specifically under low number of transmit antennas scenarios, the DF value needs to be carefully chosen. Furthermore, based on the proposed analysis, optimal value for the DF is determined considering wide LS antennas scenarios and SNR regions.

Local Search Detection in Multiple Access DS/CDMA Networks

This paper demonstrates the effectiveness of local search algorithm (LS) for multiuser detection (MuD) in multiple access DS/CDMA communication networks, using digital sinal processing platform. Even been the simplest search algorithm found in literature, the 1-LS algorithm applied to MuD problem is capable to achieve good performance-complexity tradeoff, indicating a certain efficiency in approaching the optimal solution (NP complete problem) under severe operation system and channel conditions, i.e., flat Rayleigh channel, medium and high system loading and a wide range for signal-noise ratio.

Bayesian estimators by particle filtering

This paper analyzes the main resampling methods associated to the Bayesian estimation technique namely Particle Filter (PF), trying to confirm by computational simulation the viability and efficiency of PF deployment on the radio mobile channel coefficient estimation. Also, this work establishes which of those resampling methods are more promising in terms of performance versus complexity trade-off.

Analysis of semidefinite relaxation detector in MIMO channel

Communication techniques based on multiple-input multiple-output (MIMO) have been increasingly exploited in order to obtain communication systems with high capacity and throughput, against the scenario of increasing scarcity of spectrum. In this context, the semidefinite relaxation (SDR) based-signal detector has attracted enormous interest. The sub-optimal SDR detection strategy is very promising due to its very near-optimal maximum-likelihood (ML) performance combined with polynomial time complexity cost. This work analyzes the SDR detector under MIMO channels, considering its robustness to imperfect channel coefficients estimation, as well as its performance under antenna selection technique.