Is 5G finally ready for prime time?

The telecom shifts from 2G to 3G and on to 4G were primarily speed boosts. Yes there were (and are) technical issues involved, but if you thought of these generational changes as faster speeds, you got the idea.

This is the “fourth industrial revolution.” a world of unprecedented connectivity where the consumer experience will reach new pinnacles and artificial intelligence and machine learning will transform everyday life.

The 5G ecosystem

See Also

5G Insights October 2017
5G hype pressures telcos to deploy new networks immediately
Keeping Hong Kong's economy competitive with 5G
5G: essential for the Fourth Industrial Revolution

We’ve been writing about 5G for years now, but most industry observers still peg 2020 as the year when we finally get a real-world model we can call “5G”. The reason is simple: 5G isn’t a speed boost, it’s an ecosystem. It’s going to involve far more than the computer/radio receiver-transmitters in our pockets we call mobile phones.

Here’s an example of one part of the 5G ecosystem: a firm has an expensive installation-perhaps a laboratory, or a power plant. The installation has a number of sub-assemblies (perhaps: engines) cooled by water-pumps, and each pump has a monitoring sensor attached. The sensors aren’t monitored by humans, but by other machines-this is M2M or machine-to-machine communication.

Suddenly, one of these sensors pings the central machine to report a problem. The machine checks its database and finds this sensor is approaching its five-year end-of-life cycle. Probably something the tech checking the facility the next day can examine. Change the battery or the sensor.

Then three other sensors on the same sub-assembly also ping the machine. These sensors, according to the database, aren’t near end-of-life. Something’s gone wrong: overheating, incorrect spin speeds, excess moisture perhaps.

This article first appeared in Telecom Asia 5G Insights October 2017 Edition

The key here is that so far, no human has been in the loop. The system has been monitored throughout, and a potential false-positive weeded out. But now that there’s a known situation, the central machine sends an alert to an emergency 24/7 technician, because (you guessed it) all this happens at 04:00 on a Sunday. Bingo: a network of sensors just saved an expensive sub-assembly from failing and possibly taking out the entire installation.

Here’s the evolution path: this scenario is based on a five-year battery life and a sensor that costs $1.0x. When the sensor price drops to $0.5x and battery life extends to ten years, it’s easier to have multiple, redundant sensors. Easier to build sensor-networks to monitor other processes. And smaller firms can more easily deploy sensor-networks to help protect their investments. M2M scenarios like this will be a hallmark of 5G, and they’ll benefit not only from commodity sensors, but from increased bandwidth and enhanced wireless coverage.

The human factor

Here’s another example: Mr Schmidt, a retired Austrian, likes to drive around the region visiting his relatives. He’s elderly and has some medical issues, so he’s informed a multi-national government agency of his condition (assume for the sake of argument that an ironclad EU-wide data privacy policy prevents misuse of this information).

Mr Schmidt is driving in Germany, or Switzerland, when a blizzard strikes. Cars are spinning into ditches all across central Europe. Emergency services are busy dispatching tow trucks and, sometimes, ambulances.

Suddenly, a number of sensors in Schmidt’s car fire off at once: accelerometers, airbag-deployment switches, a rooftop sensor that detects the temperature and moisture of snow. Inside his car, an alarm sounds and a large red button lights up on the dashboard-a computerized voice addresses him, by name, in German, and tells him to press the red button immediately.

A minute passes and there’s no response. So far, as in our power plant example, no humans are involved-this is all M2M. But now a machine informs the local emergency services. They patch into the vehicle’s audio system and call Mr Schmidt by name. No response.

Now the rescue pattern over the region changes. The Schmidt car is where the helicopter goes next-and it follows an automatically dispatched drone that gives first-responders a video overview of the situation.

This too is 5G in action, but it’s a more involved strategy. The make of Mr Schmidt’s car shouldn’t be a factor, nor should the country he’s driving in. Vehicles and emergency services must standardize frequencies to make this scenario work.

Note also that as Schmidt provided information including his medical history and preferred language, those data-parameters must be protected. This level of data privacy isn’t for Facebook or Google to promote products to Mr Schmidt-the information is there so that emergency services can evaluate his situation without his input, and not for commercial purposes.

This level of public service should not be compromised by personal interests, and EU laws are moving in that direction already. What about Asian countries? Japanese drivers in Korea, for example?

Examples like this help people wrap their heads around 5G. Yes, driverless cars are in the mix, along with many other possibilities. Just remember that it’s an ecosystem, with opportunities and pitfalls, but ultimately all 5G developments come down to their human components. Can we get to Mr Schmidt faster, with better information? Can we build a sensor-network that helps a car dealership or pharmacy guard against theft without round-the-clock human security guards?

Pages

About the author

Commentary

5G and data center-friendly network architectures

Matt Walker / MTN Consulting

Webscale and transmission network operators' interests are aligning as the 5G era dawns

Matt Walker / MTN Consulting

Webscale and transmission network operators' interests are aligning as the 5G era dawns

Rémy Pascal / Analysys Mason

The launch of 5G by South Korean operators serves as a first benchmark for other operators around the world