It’s accurate to characterize mobile 5G as simultaneously everything, and nothing.

Ask any wireless-side technologist how they explain 5G to their bosses. A bemused look of exasperation tends to happen, followed by some variation of this: “I tell them, it’s easy! 5G is everything! Let the imagination run wild!”

Let’s start with what 5G isn’t. It isn’t 5 GHz (GigaHertz, which is a frequency/spectrum thing.) It isn’t 5 Gbps (Gigabits per second, which is a speed/throughput thing.) No, the “G” in “5G” stands for “generation.” Fifth generation. Of what? Of mobile stuff.

The main reason why 5G keeps showing up, as “opportunity” and “threat,” is its architecture: Highly distributed, small cells, running a mix of licensed and unlicensed spectrum. If that smells a lot like HFC (hybrid fiber/coax) to you, take the rest of the day off! Because you’re right.

Depending on who’s talking, here’s a sampling of what 5G will do for our mobile lives: 
  • Retool the airwaves into Gigabit spigots. 
  • Serve as connective tissue for the Internet of Things. 
  • Sensor-reading, machine-to-machine, low power WANs, sure, why not? 
People who consider 5G a threat worry that it could replace the “last mile,” as in, the part of the plant that connects homes to headends. Allow us to quote Balan Nair, CTO of Liberty Global, about that potential: “Total bullshit.”

People who see 5G as an opportunity are the same people who point out that every single wireless thing in the world needs a wire, some of the time. Plus, 5G is coming out of the mobile industry, which is already a bestie in the frenemy category, thanks to those established relationships with operators for cellular backhaul. The point: All those little 5G cells will need backhaul, too. 

One way to look at how 5G might intersect with the wired world is to pick a place, and compare the blueprint of its cellular network, with a blueprint of an HFC network in that same area. What would it take for a mobile carrier to light up a 5G small cell overlay? 
Step 1: Determine the radius of the 5G radio / the small cell. How small is small? Millimeter wave radios, because they live way up in the Extremely High Frequency (EHF) zone, don’t go very far. Estimates vary (predictably), but somewhere between 100 and 300 meters is conceivable. 

Step 2: Figure out the how many small cells will be needed, based on how many people/devices are already inside that footprint. Let’s say it’s Denver. Let’s also say that mobile carriers here service the city’s 650,000-ish inhabitants with 100 macro cells (which is a total guess.) 
The point is that “densifying” that mobile footprint for 5G will likely involve an order of magnitude increase in small cells. Like if it’s 100 macro cells now, for 3G-4G, ball parking 1,000 small cells for 5G is, as they say, close enough for horseshoes. 
When might you see 5G, like in your hand, working?  Standards bodies -- plural -- are working on 5G, from the traditional mobile side, and the WiFi side. If the standards finalize by mid-2018, plus another 1.5 year-ish margin for silicon development and manufacturing, 5G looks like a 2020 thing. So we have three years to see which of its many tangents win, lose, or draw.​