Network slicing will play key role in 5G networks

With dynamic network slicing, 5G applications will operate on a flexible zone on the network, allowing them to draw varying, adaptable levels of bandwidth and reliability.

Network slicing will play key role in 5G networks
Thinkstock
Current Job Listings

Single wireless networks will be separated into many, layered virtual networks when 5G is launched, experts say. The technique, called network slicing, is of a similar concept to software-defined networking (SDN), found now in some fixed networking, where managers program network behavior.

The way dynamic network slicing will work is that communications specific to a particular 5G application, such as those found in Internet of Things (IoT) sensors or video, will be layered over the top of a common infrastructure, then software will manage the different service types.

That’s unlike how wireless networks function now. Currently, infrastructure is loosely shared among voice and data with little consideration for just how much resource is needed. That throw-it-against-the wall method means assets get wasted.

“Sensors need to transmit small amounts of data sporadically, but frequently,” Dawn Bushaus writes in a TM Forum article. However, “an application, like remote surgery, needs huge amounts of bandwidth, extremely low latency and guarantees of availability and reliability” — thus the opportunity to provide adaptability for what will be a limited resource (like any other spectrum).

How to accomplish network slicing

A slight hiccup is that despite players agreeing that network slicing should play a major role in 5G networks, there isn’t a consensus about how it should be accomplished. And indeed that problem ties in with 5G itself: Despite 2020 being proposed by the International Telecommunication Union (ITU) for the rollout, the new tech has yet to be ratified as a standard, and frequencies and equipment haven’t been decided upon universally — important for globalism.

5G is hoped to provide efficient machine-to-machine communications for robots, sensors, healthcare, smart meters, cameras, cars, drones and so on — all with differing demands.

With network slicing, flexible zones will be created for each element, which will lie dormant when not being used. Examples: One slice for always-connected IoT data that trickles in from a smart meter and a high-throughput slice for augmented reality (AR) data that comes barreling through.

The network operator must be able to orchestrate bandwidth, latency, capacity and reliability separately as distinct offerings, 5G players believe.

“A fundamental rethinking of the mobile network is needed to realize the full potential 5G offers,” telco equipment maker Nokia said in a 2016 white paper on its 5G slicing plans — one of many from vendors. It says its 5G network proposals are based on services, not connectivity, and says connectivity is a misnomer these days.

“This is a paradigm shift from past networks, which were primarily built to offer connectivity,” Nokia says of its proposed slicing.

But, of course, behind the excitement from the vendors, 5G isn’t the only communications network around when it comes to IoT. Others exist now or are in varying forms of development. They include traditional cellular, low-power wide-area-networks (LPWAN), LoRaWAN, RPMA, XBee and Bluetooth Mesh. (I wrote about many of those networks earlier this year.)

Never wait for announced technology — you can be waiting a long time. That's something that I learned in tech many years ago. This may well be a good moment to repeat the mantra.

Join the Network World communities on Facebook and LinkedIn to comment on topics that are top of mind.
Now read: Getting grounded in IoT