Analytical Framework for Full-Dimensional Massive MIMO With Ray-Based Channels
- 26 August 2019
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 13 (5), 1181-1195
- https://doi.org/10.1109/jstsp.2019.2937635
Abstract
The performance of baseband beamforming in multi-user multiple-input multiple-output (MU-MIMO) systems has been extensively studied for simplified statistical channel models where no angular parameters are taken into account. In contrast, there is little performance analysis with ray-based models, which are more physically motivated, feature prominently in standardization and have been experimentally validated. Thus, unlike previous studies, we present a mathematical framework to analyze the performance of zero forcing (ZF) and minimum mean-squared error (MMSE) combining. Using a central result for averaging in the angular domain, we derive tight approximations for the uplink signal-to-noise ratio and signal-to-interference-and-noise ratio (SINR) for ZF and MMSE processing, respectively, and the resulting spectral efficiencies. The remarkably simple expressions offer the following insights into the effects of the propagation environment. We demonstrate an improvement in performance when moving from vertical uniform rectangular array (URA) to horizontal URA to uniform linear array (ULA) antenna configurations. There is also a corresponding increase in the robustness of the performance to propagation scenarios. We demonstrate that under specific conditions increasing the angular spread can decrease the SINR for a ULA - an unexpected behavior which we link to the effects of end-fire radiation. Furthermore, our results allow us to investigate the impact of different array configurations and system parameters on the rate of convergence to favorable propagation conditions. Finally, we evaluate the spatial correlation properties intrinsically present in ray-based models, and compare them to the commonly used simple exponential model which yields equal, fixed correlation characteristics for each user.Keywords
Funding Information
- Engineering and Physical Sciences Research Council (EP/P000673/1)
- RAEng/The Leverhulme Trust Senior Research Fellowship (LTSRF1718\14\2)
This publication has 36 references indexed in Scilit:
- On the energy efficiency of massive MIMO with space-constrained 2D antenna arraysPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2016
- Massive MIMO for Maximal Spectral Efficiency: How Many Users and Pilots Should Be Allocated?IEEE Transactions on Wireless Communications, 2015
- Achievable Downlink Rates of MRC and ZF Precoders in Massive MIMO With Uplink and Downlink Pilot ContaminationIEEE Transactions on Communications, 2015
- Asymptotic Orthogonality Analysis of Time-Domain Sparse Massive MIMO ChannelsIEEE Communications Letters, 2015
- Performance Analysis of Massive MIMO for Cell-Boundary UsersIEEE Transactions on Wireless Communications, 2015
- On the matrix inversion approximation based on neumann series in massive MIMO systemsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2015
- Massive MIMO Performance Evaluation Based on Measured Propagation DataIEEE Transactions on Wireless Communications, 2015
- Space-Constrained Massive MIMO: Hitting the Wall of Favorable PropagationIEEE Communications Letters, 2015
- Millimeter Wave Channel Modeling and Cellular Capacity EvaluationIEEE Journal on Selected Areas in Communications, 2014
- Noncooperative Cellular Wireless with Unlimited Numbers of Base Station AntennasIEEE Transactions on Wireless Communications, 2010