The Cranky Admin
Hyperlocality and 5G: Filling the Need for Speed
Building the next generation of mobile networks.
We broke the Internet exactly how it was always predicted: we filled up the tubes with cats. When it comes to wired bandwidth problems, this can be solved by simply installing a bigger pipe. Mobile networks aren't quite as simple, something 5G aims to address.
Mobile networks rely on radio signals. In an effort to build networks with the longest possible range and the fewest dead zones, the first mobile networks were built using low-ish frequency radio signals, generally below 2Ghz.
The lower the frequency of the radio signal, the longer the effective distance. Back in the days of voice-only cellular networks, an 800Mhz signal was great for rural areas. Stand up a single tower, cover a huge swath of the countryside!
Mobile network design is a little more complicated in practice than simply worrying about coverage. Neighboring cells shouldn't be exactly the same frequency, so within your allotted spectrum you usually have a few frequency blocks you repeat in a pattern design so that no one cell tower interferes with another.
Each tower also represents a fixed amount of bandwidth. HSPA+, for example, could theoretically handle up to 42.2 Mbits/sec. That sounds great, until you realize that theoretical capacity has to be shared by all users.
Additionally, each cellular device regularly emits beacons looking for service, neighboring devices, or advertising itself. Get enough devices in a small enough space and they'll occupy the entire spectrum, doing nothing more than twittering on about their own existence.
WiFi and mobile connectivity in stadiums and conference centers is fun. For dystopian values of fun.
None of this stopped us from moving from mobile voice to mobile data. Cell phones became smartphones and smartphones became us. They are the digital embodiment of ourselves. Our gateway to the sum total of human knowledge, right there in our pockets.
Opening a Door
The mobile revolution wasn't one in which we replaced our desktops and notebooks with iPads and Android phones so much as we unlocked an entire universe of possible computer applications that would have seemed absurd two decades ago. This had a knock-on effect where even the most mundane technologies received convenience upgrades made possible only because of both the miniaturization of compute capacity and the ubiquity of mobile connectivity.
In 1985, paying for dinner with a credit card usually involved carbon paper and a machine that went "ka-thunk." In 1995, it involved standing at the till while the card machine squealed loudly and connected with a modem. In 2005, paying was still done at the till, but over always-on broadband and it took a third of the time. Today, the waiter brings a cellular-enabled handheld reader to your table and you absentmindedly punch numbers in without a break in conversation.
The cabbie that drove you to the restaurant doesn't need a full-blown mobile card reader; he can take payment by scanning the chip on your card using the NFC reader on his phone, or using a magnetic swipe device connected via Bluetooth. On your way back, maybe you go through the drive-through of the local coffee shop and pay simply by tapping your smartphone on the payment rectangle hanging out the window.
Coping
Millions of people doing these seemingly mundane things on a daily basis takes it toll. The existing mobile networks are strained to collapsing, and the full force of the Internet of Things (IoT) revolution -- the computer-to-computer bit where human interaction isn't really involved -- has barely even begun.
In light of this, the next generation of mobile networks -- 5G -- are being designed to capitalize on the fiber rollouts the public cloud forced. 5G networks will not ask a neighborhood in an urban environment (or hundreds of square kilometers in a rural environment) to all connect to the same cell tower. Instead, 5G will rely on lots and lots of really small cells connected to any broadband source they can find.
Yes, the large 5G towers will still exist, and they'll use frequencies that propagate appropriately for their environment. Lower frequencies like 2-5Ghz will be used to maintain connectivity for those outdoors. There are also plans to use ultra-short distance frequencies in the 20 GHz and 50 GHz range that are only useful if you're almost standing right next to the base station.
These high frequency, ultra-short range cell nodes would be put in businesses, malls, stadiums and so on. Places where high bandwidth consumers (people) are relatively stationary. They would provide high bandwidth connectivity without burdening the primary towers and without leaking signal for miles, allowing for incredible cell site density. Add in the proposed device-to-device capabilities and the 5G network promises to be not only capable of high speeds, but incredibly resilient.
This is 5G. It is the exact opposite of a single large tower on the hillside beaming connectivity for hundreds of kilometers in every direction. 5G will still have that capability, but the focus will be on hyperlocality: on meeting the our insatiable need for speed as close to where we are as possible.
About the Author
Trevor Pott is a full-time nerd from Edmonton, Alberta, Canada. He splits his time between systems administration, technology writing, and consulting. As a consultant he helps Silicon Valley startups better understand systems administrators and how to sell to them.