Friday, August 4, 2017

Verizon LAA Test shows 953 Mbps 4G Downlink Speed

Why do mobile service providers want to use Licensed Assisted Access (LAA), aggregating mobile spectrum with unlicensed spectrum? Because doing so allows higher speeds.

In what Verizon calls “a U.S. wireless industry first,” Verizon, Ericsson, and Qualcomm Technologies achieved 953 Mbps downlink  speeds on a commercial 4G network in Boca Raton, Fla.

The demonstration used all commercially available Verizon network components including a cell site, hardware, software, and backhaul. Riding on the backbone of Verizon's most reliable network infrastructure, while Ericsson provided an advanced remote radio head in the industry.

The micro Radio 2205 for LAA, designed for unlicensed spectrum use, provides small dimensions, flexible mounting and superior performance, and is a component of the Ericsson Radio System, an end-to-end modular radio network portfolio of hardware and software designed to fit all site types and traffic scenarios as networks grow in scale and complexity on the road to 5G.

Qualcomm Technologies provided a Qualcomm® Snapdragon™ 835 mobile platform test device, with Gigabit LTE capability thanks to the integrated Snapdragon X16 LTE modem.

Inevitably, as near-gigabit speeds are deployed on 4G, some are going to start complaining about “fake 5G.” For human users, for typical applications, it will not matter, no matter what we call it.

There will be some apps that will benefit from 5G lower latency, but most 4G users likely will not much notice the difference between 4G and 5G latency in normal daily use.

The same sorts of complaints were raised when 4G first was introduced. There was the issue of whether WiMAX was really 4G, for example. I used it. So did lots of other people. We just wanted faster speeds. We couldn’t really have cared much which “G” we were using.

No comments:

Many Winners and Losers from Generative AI

Perhaps there is no contradiction between low historical total factor annual productivity gains and high expected generative artificial inte...