Signals Flash! Good Grief!
Signals Flash! Good Grief!
Proving the Case for 5G SA, bit by bit
08/06/2026 | 18 pages
Signals Research Group (SRG) just completed testing in downtown Saint Paul, MN in which we looked at the side-by-side uplink performance of an NSA-enabled smartphone and an SA-enabled smartphone. This Signals Flash report provides the results and analysis, plus it demonstrates we are alive and well since we won’t have our next Signals Ahead report out until early September.
Key Highlights
Our Thanks. We leveraged test equipment from Accuver Americas (XCAL5 and XCAP) as well as from Keysight Technologies (Umetrix Data). Both companies have been trusted partners for nearly two decades.
Our Methodology. We used two Motorola razr fold 2026 smartphones – one configured tosupport 5G SA (Band n2) and the other one configured to support 5G NSA (Band n2 + LTE Band 66). We did walk testing with both smartphones while the two phones simultaneously uploaded data (UDP full buffer) on the Verizon network (Ericsson infrastructure). We then analyzed the data using our standard methodology and metrics, with analysis looking at performance as a function of RSRP as well as geo binned data.
The Results. The results show the 5G SA smartphone had 2.2x higher spectral efficiency than the 5G NSA smartphone. The higher spectral efficiency was evident with both the 5G radio channel (Band n2 = 10 MHz) and especially the LTE radio channel (Band 66 = 20 MHz).
The Explanation. With 5G SA, the smartphone used uplink MIMO (2 data layers) while with 5G NSA, there was a single data layer on 5G and another data layer on LTE. LTE is also less spectral efficient than 5G. We also limited our analysis to 32T32R radios, although we note both LTE Band 66 and 5G Band n2 used the same massive MIMO radio.