Thanks to multi-user, multi-input, multi-output (MU-MIMO) tech, Signals Research Group (SRG) hit a significant benchmark peak of 5.4 Gbps. The independent 5G NR MU-MIMO study’s results potentially have far-reaching ramifications from a spectral efficiency standpoint, SRG CEO Mike Thelander said in an exclusive interview.
“You can always get a higher and higher data speed if you have more spectrum available. The trick is to get spectral efficiency,” he said of the test, which leveraged Ericsson 16-layer MU-MIMO tech over 100 MHz of C-band spectrum. “In this case you’ve got 5.4 Gbps using 100 MHz of spectrum. If you do the math, it comes down to 54 bits per second per hertz.”
Thelander added that the higher you go into mmWave frequencies, for example, it’s easier to get much higher data speeds. The drawback is “you’re also throwing more spectrum at it,” making efficiency critical.
“So, from an operator’s perspective, the higher the data rate or the more data you can send for a given amount of spectrum, the better you are,” he said. “From a user’s perspective, all you care about is the higher data speeds, but, ultimately, especially with the C-band spectrum, there’s only so much spectrum available, so you’re somewhat limited in that regard to get those higher data speeds.”
Massive MIMO vs. Multi-User MIMO
MU-MIMO is one way to improve spectral efficiency. As it stands, operators already use massive MIMO to optimize mid-band and mmWave 5G throughput and efficiency gains. MU-MIMO is an application of massive MIMO that takes it to another level, allowing operators to reuse the spectrum, 16 times in this case.
According to the SRG study itself, in sharp contrast to massive MIMO, use of MU-MIMO among providers has been extremely limited up to now. Things are changing in that regard, with T-Mobile for example similarly leveraging Ericsson MU-MIMO tech in a recent test to hit a theoretical max speed 5.6 Gbps. Like with the SRG test, eight different devices on the same cell were used (each capable of reaching 700 Mbps). T-Mobile’s test meanwhile ran on 100 MHz of its 2.5 GHz band.
In a separate interview, Dan Warren, director of advanced network research at Samsung R&D UK, agreed MU-MIMO is likely to make more of an impact in the industry moving forward. He called it almost inevitable.
“If you take the trend and the direction that radio wireless network technology has gone over the last few years, there’s a number of comparative parts that give you higher bandwidth, one of which is MIMO and MIMO has been moving up in scale, from kind of 2X2 to 4X4 to now 16X16. And we’ll probably see things rise further than that as well, particularly if you move into higher-frequency bands,” he said.
“There’s a bunch of good reasons why higher-order MIMO will happen. It has technical challenges with it, but they’re technical challenges, which, as the demo demonstrated, we’re actually overcoming.”
MU-MIMO Far from a Silver Bullet
Thelander seconded the notion about the technical challenges, saying MU-MIMO is admittedly not a “panacea.” Ultimately, the goal for providers is to enable high speeds for users in ways that make financial sense. MU-MIMO can achieve that, with some caveats.
“It works really well in some conditions and not as well in others,” he said. “The impact’s not going to be dramatic by any means, but an operator would be able to see it by looking at their [Key Performance Indicators]. It’s not going to be all of a sudden a 2X or 3X increase in consumers’ data speeds, but there will be definite improvements in certain conditions where subscribers will get a higher data speed.”
Even though the consumer gains to be had are far from game-changing relatively speaking, they’re hard to ignore in the context of an ever-present need to improve service. Thelander spoke to how initial speeds at the respective launches of LTE and 5G were very good, only for throughput to decline as more and more subscribers joined. To address the situation, SRG predicts operators with spectrum at 2.5 GHz and higher will start deploying MU-MIMO in the coming months, as it is in everyone’s best interest.
“An operator that maybe has mid-band spectrum but doesn’t have a lot of it, they can offer a 5G service, but you have just a modest amount of users and suddenly your data rate starts to drop and you have to add more cell sites,” he said. “It becomes very economically unattractive, but with MU-MIMO and having higher spectral efficiency you can serve more subscribers simultaneously and give them a good user experience.”
