An Inflection Point in Telecom: The 5G Vector

There is an increasing convergence of services offered by mobile, cable and over-the-top (OTT) players. Each industry has its own strengths: mobility from the wireless players, video delivery from cable operators and the speed of service introduction from OTTs.

This situation is not new. However, with LTE and cloud computing here now, and 5G not far away, the architectural enablers to blur boundaries and further increase hyper-competition are increasing. We are at an inflection point, and one of the looming disruptors is 5G.

The 5G plan
On the radio side, there are a few more steps in 4G LTE coming, such as increasing data rates with carrier aggregation, better support for devices (LTE-M), support for unlicensed spectrum (LTE-U) and push-to-talk like services (LTE-D). They extend LTE's lifespan and improve yet again in 5G.

Work on 5G is progressing well. It's anticipated that the first products will start showing up in 2018 at the South Korean Winter Olympics, followed by a commercial system at the Japan Summer Olympics in 2020. Of interest, of course, is the date for US deployments, for which there is no obvious event driver. However, the timing of release of spectrum by the FCC after 3.5GHz and 600MHz will be a key indicator.

The phasing in standards, while not strictly tied to these events, has a tight relationship in time. We expect 3GPP R14 in about 2018 to define a Phase 1 consisting of a 4G core and 5G access network. The arrival of 3GPP R15 in about 2020 will define a 5G core and 5G access network. This phasing concept has not officially been ratified, but is a candidate proposal gaining some strength.

Why is 5G exciting? It will provide, as all generations do, two things: The first is an expansion of eternal needs; faster, cheaper networks. In the access network, the rule of thumb is 10x versus 4G, which, in this case, translates to peak rates of 10 Gbit/s and 100 Mbit/s at the cell edge. The second are enablers for new services. These primarily come from a somewhat infamous 1ms latency target and high reliability.

With these characteristics, we'll see cellular used for things it hasn't or couldn't be used for before, like tactile Internet, autonomous cars and industrial control. Think of 5G as taking us beyond human connectivity to connecting all things -- in other words, a "programmable world."

The access network will use spectrum both below 6GHz, up to and perhaps even beyond 100GHz. In the cmWave and mmWave realm, we expect new radio technologies to support the complex propagation characteristics in those frequencies, working seamlessly with LTE for wide area networking. 5G in these frequencies is a dense collection of small cells.

In the core network, the cloud architectures being implemented today are an assumed baseline for 5G, but will need to evolve to support things like low latency with distributed gateways, dual connectivity (between LTE and 5G access nodes), C/U-plane separation and a host of other cost and implementation improvements.

From this description, you can see that 5G will disrupt current networks by introducing small cells on a larger scale than in 4G to date, with multiple cooperative air interface technologies, and presumably distributed clouds, for those cases where low latency services are needed. There are of course linear improvements as well, but disruption is also clear.

Disruptive vector
So why is all of this an inflection point? And a vector? It's an insanely mathematical perspective.

It's an inflection point because 5G somewhat brings with it the merits of the three industries noted initially: high data rates for video distribution, service agility through clouds, and support of low latency, reliable communications for critical services. It's the perfect storm for wireless operators to better compete with cable companies and OTTs. Conversely, it could also be a tool used by those same or new competitors to compete with wireless operators. Imagine new wireless entrants using 5G core network components combined with WiFi and 5G access nodes in shared or unlicensed spectrum to expand their reach.

The key for CTOs, in all industries, is to find the most direct line or vector that maximizes investments made today to support the future. The ideal is a straight line versus one burdened with significant changes in direction over the years. The coming of 5G brings new challenges. For example, with 5G distributed clouds and small cells, is that something that should be considered or optimized for now in network plans?

— Michael Murphy, CTO North America, Nokia Networks

TV Monitor 5/20/2015 | 6:13:34 PM
US 5G service starts at above 24 Ghz I can't believe the CTO of none other than Nokia Networks write a clueless article like this one. Totally speechless.

The FCC already made its intentions known that it plans to allocate a freuqency above 24 Ghz for US 5G services, and the US 5G service will start above 24 Ghz(more like 28 Ghz LMDS band) whether Nokia likes it or not. I understand Nokia and Ericsson are completely unprepared for 28 Ghz LMDS band 5G services and would prefer to see a band below 6 Ghz for low-band 5G services, but that doesn't change the fact that the 3.5 Ghz CBRS band is unsuitable for 5G.

So it is time for Nokia to put a hold on that futuristic 70 Ghz band work and redirect all its efforts and resources on conquering 28 Ghz band, because Samsung stands to take the whole 5G pie and eat it alone unless Nokia gets its acts together. While Nokia's 70 Ghz effort might prove to be useful in 5.5G or 6G, Nokia might not be around to enjoy the fruit of its work on 70 Ghz if Nokia doesn't get acts together on more immediate 28 Ghz band. Think that can't happen? Just remember what happened to Nokia's handset business.

Just remember that Samsung will demonstrate its 28 Ghz 5G service in front of world audiences in Februery of 2018. Once the world press gets to see the 28 Ghz 5G modem equipped Galaxy S8 and LG G6 delivers 4K live streams during the 2018 Olympics, it's all over for low-band 5G afforts by Nokia-Alcatel, Huawei, and Ericsson.
Sign In