January 12, 2022
Signals Research Group (SRG) completed a new study on T-Mobile’s 5G New Radio (NR) network in the D.C.-Maryland suburbs, where the SRG team examined a range of technologies using at least one handset with pre-commercial software. Ericsson was the network vendor in the area they tested.
A lot of technologies are involved here: Standalone 5G, Non-Standalone 5G, carrier aggregation, Time Division Duplex (TDD) and Frequency Division Duplex (FDD).
The upshot shows what a lot of folks may already suspect: T-Mobile’s 600 MHz spectrum is good, and 2.5 GHz is good, but when you put the two together, that’s “super good.” (Granted, that’s not the most techie of descriptions, but Mike Thelander, CEO and founder at Signals Research Group, took time to confirm this most basic translation for us.)
T-Mobile has been blasting its own horn about various tech-related accomplishments for a while now. It’s the first operator in the U.S. to roll out a national SA network. It’s first with a sizable mid-band spectrum play for 5G, in the form of 2.5 GHz spectrum that it acquired from Sprint. President of Technology Neville Ray, who frequently talks about the “spectrum layer cake” for 5G, promises to lead in the Voice over 5G, or VoNR, front as well.
So it’s of interest that SRG hit the roads around Washington, D.C., to measure what T-Mobile is putting out there, particularly when it comes to 5G NR FDD-TDD Carrier Aggregation (CA).
In its study, SRG noted that T-Mobile provided the smartphones and some insight on the cell cluster where they tested, but T-Mobile otherwise “had no direct involvement in our study,” which was conducted independently by SRG mainly for the benefit of its Signals Ahead subscribers.
Specifically, the SRG study focused on 5G NR FDD-TDD CA and how the use of 600 MHz as the primary cell results in better performance of 5G NR in 2.5 GHz spectrum, including higher data speeds close to the cell site and more importantly, higher data speeds and extended 2.5 GHz coverage at the edge of the cell.
With the TDD-TDD CA configuration using 100 MHz and 20 MHz NSA channels, SRG recorded peak throughput exceeding 1.7 Gbps with 85% of that due to 5G NR. The throughput was lower in FDD-TDD CA mode, but that was due in part to less bandwidth in the FDD channel.
SRG used two Samsung Galaxy S21 smartphones. One of the handsets had a pre-commercial software release, which was intended to enable some of the carrier aggregation functionality they wanted to test, according to the report. However, by the second day, it appeared that both smartphones supported most, if not all, of the functionality they set out to test, the analysts said.
Using “hidden menu” commands, “we were also able to force the smartphones into various modes of operation, including SA or NSA, Band n41 only, etc. This ability provided an additional dimension to our test methodology,” wrote SRG. The research firm collaborated with test and measurement companies Accuver Americas and Spirent.
T-Mobile launched 5G NR in the Standalone version in 2020, but the operator is still in the process of enabling that functionality in its 2.5 GHz spectrum on a nationwide basis, SRG noted. Plus, it still has LTE in its 2.5 GHz spectrum that it inherited from Sprint, “so they’re somewhat limited in what they can flip over to 5G,” Thelander said.
Old FDD/TDD debates
FDD/TDD existed with LTE, and for a number of years WiMAX and LTE battled it out, although LTE ultimately won out. Clearwire, where a lot of T-Mobile’s 2.5 GHz spectrum can be traced back to, was behind WiMAX, which was seen as the technology of choice for those with TDD spectrum. LTE fell into the FDD camp.
The SRG report noted that back in the days of WiMAX, there was a corollary debate between TDD (WiMAX) and FDD. The debate was that TDD delivered higher capacity, but suffered with coverage, while FDD achieved better coverage with the tradeoff being lower capacity.
The largely unused uplink channel with FDD factored heavily into the argument, but lost in the debate was the “impracticality of deploying WiMAX in paired spectrum (FDD) or 3G in unpaired spectrum (TDD),” the SRG report said.
“5G NR FDD-TDD CA demonstrates that you can have your cake and eat it too,” SRG said. “With FDD-TDD CA, the primary cell, or another carrier, uses a low-band FDD channel, and the secondary cell uses a mid-band TDD channel. In the case of T-Mobile, the operator is leveraging Band n71 (600 MHz) for the P cell (primary cell), which is used for the uplink control and data channels of the P cell and the S cell (secondary cell). The uplink is always the limiting factor with respect to coverage. Higher frequencies have inherent coverage challenges compared with lower frequencies.”
In addition, TDD is coverage limited compared with FDD because there’s less time to transmit the same amount of data, meaning higher transmit power is required in the uplink direction to deliver the same amount of data. “Net-Net: there is a triple whammy of reasons why FDD-TDD (low-band/mid-band) delivers much better coverage than TDD mid-band only,” SRG concluded.
Operators have deployed TDD/TDD in other parts of the world, so T-Mobile is not first in that regard. “Where I think they are first is doing FDD/TDD” in 5G, Thelander told Fierce. That existed to some extent with LTE, but “it becomes more interesting with 5G,” where there’s a much bigger focus on mid-band spectrum. “Having FDD/TDD is going to be very important for those operators,” he said.
It’s worth noting that AT&T and Verizon are using dynamic spectrum sharing (DSS), “so they’ve got to work through that.” But they will benefit from a similar scenario where mid-band is combined with lower-band spectrum.
T-Mobile also combined Band 41 + Band 41 (2.5 GHz) for the TDD/TDD aggregation. That’s OK, “but your primary carrier is 100 megahertz wide, and in T-Mobile’s case, the secondary carrier is only 20 megahertz,” so “you don’t gain a lot, necessarily.”
“It’s a technology achievement, but at the end of the day, it doesn’t do a huge amount to the performance of the network and the user experience, but it’s nice to have,” he said. “To me, the real exciting thing is the FDD/TDD side of the house.”
Thelander said he expects the benefits of FDD/TDD to become more evident in in-building situations, where the signal is more likely to be blocked or attenuated, and in general, in areas where the cell density isn’t as great.
If an operator has a dense network and a lot of capacity on the downlink, “you’d want TDD. If you want to cover a broad area, you want FDD. With FDD-TDD carrier aggregation, it gives you the best of both worlds,” he said.