Summary

SRG just completed its 60th 5G benchmark study. For this endeavor we collaborated with Accuver Americas to conduct a benchmark study of MU-MIMO.  The study included SU-MIMO with sub-band beamforming and MU-MIMO with zero forcing, and how these two new features compare with legacy implementations of SU-MIMO and MU-MIMO.  We tested at T-Mobile’s HQ where Nokia is the infrastructure supplier.

Key Highlights

Our Thanks.  We did this study with the support of Accuver Americas.  Accuver provided us with its XCAL5 drive test software and XCAP post-processing software.  We used XCAL5 to collect the chipset diagnostic messages on all smartphones and we used XCAP for the post processing of the data and subsequent analysis.  Accuver has been trusted partners with us for nearly two decades.  SRG is responsible for all analysis and commentary provided in this report.

The Background.  Nokia announced MU-MIMO with zero forcing at MWC 2026.  When planning for our R17 UL Tx switch study, T-Mobile indicated it had installed MU-MIMO with zero forcing at a cell site near their HQ.  Since we had the opportunity to test the feature on the same trip, we gladly accepted the opportunity.  T-Mobile provided some critical logistical and engineering support, but SRG controlled the test plan and the subsequent analysis.   We had no interaction with Nokia until after we had completed the testing.  This study was solely funded by our Signals Ahead subscribers.

The Methodology.  We used four Samsung Galaxy S25 Ultra smartphones that we positioned at various outdoor locations around the T-Mobile parking lot in front of their buildings – the cell site was across the street.  Testing occurred during the maintenance window but other commercial traffic was present in the network.  In addition to collecting results with multiple stationary positions, we did some mobility tests in which one smartphone moved around the test area, periodically stopping at various locations, including immediately next to another smartphone or along a directional line between another smartphone and the cell site.

The Analysis.  For each placement of smartphones, we tested four network configurations:  SU-MIMO (legacy), SU-MIMO (sub-band beamforming), MU-MIMO (legacy) and MU-MIMO with zero forcing.  We analyzed RB allocations, MIMO layers, MCS, and throughput / spectral efficiency.  The report provides comparative results for each configuration when all smartphones were receiving full-buffer data, while we also show time series results for all metrics across the entire test as we started / stopped the data transfers to each smartphone.  We also include distances to the serving cell and the angular displacements between smartphones relative to the serving cell.

The Results.  We observed meaningful double-digit gains in total throughput / spectral efficiency on a percentage basis between SU-MIMO with sub-band beamforming and SU-MIMO (legacy).  MU-MIMO with zero forcing allowed MIMO pairing to occur where it wasn’t possible with MU-MIMO (legacy), resulting in triple digit gains on a percentage basis with SU-MIMO (legacy) and high double-digit gains compared with MU-MIMO (legacy).

What Happens Next?  The gains from SU-MIMO with sub-band beamforming versus SU-MIMO (legacy) more than justify the upgrade, meaning MU-MIMO with zero forcing is icing on the cake.  We see no reason why an operator with Nokia infrastructure wouldn’t use the enhancements, even if they were MU-MIMO skeptics.  Given so many variables, it is impossible to compare these results with other MU-MIMO benchmark studies that we have done in the past.