Before starting an article about 5G I think it’s extremely important to clearly define what 5G actually is. The term “5G” has been commonly used by the media to mischaracterize many upcoming wireless technologies.
5G is wireless protocol defined by the International Telecommunication Union’s IMT-2020 standard. The most recent iteration of this standard is called 5G NR. Henceforth when referring to “5G” I will be talking about this 5G NR standard.
Now that those boring semantics are out of the way we can get into the meat of what makes 5G interesting, and how it will impact your life. One of the most important distinctions between 5G and the now-commonplace 4G LTE is increased frequency allocations.
An important characteristic that determines a network’s speed in crowded and electronically noisy areas is how much of the electromagnetic spectrum is allotted for it. In general, networks with more allocated spectrum can support more concurrent connections, increasing the network speed for all parties. Major US carriers bid against each other to buy portions of the spectrum from other companies and the US government. Carriers will spend billions of dollars on only a few megahertz.
LTE signals range in frequency from around 700 MHz to 2.5 GHz. Generally, lower-frequency waves have more of an ability to transmit data at long distances and are thus more valuable to carriers because a single cell site can cover more customers. This comes at the tradeoff of speed, however, as higher-frequency networks support higher data transmission rates.
5G NR specifies two network bands creatively named Frequency Range 1 (FR1) and Frequency Range 2 (FR2). FR1 encompasses the current spectrum that we all know and love: frequencies below 6 GHz. This portion of the electromagnetic spectrum includes everything from AM/FM radio all the way up to fast 5 GHz dual-band Wi-Fi.
The new addition, FR2, includes “millimeter wave” frequencies above 24 GHz. This portion of the spectrum is used to transmit data back and forth from satellites, used in radar installations along the coasts, and much more.
These aforementioned millimeter wave FR2 bands allow carriers to take much of the pressure off of the congested FR1 bands. The only problem is that you basically need to be able to see the antennas to get any kind of useable connection, so this is useless for anything but dense urban environments where network congestion is the highest. FR2 bands will also facilitate faster data transfer rates.
This discussion leads me to the main point of the article: 5G will not radically change the types of services available on your phone. You won’t suddenly be able to download multi-gigabyte files in one second. What you can expect is a bump in speed and capability comparable to the shift from 3G to 4G. A significant shift, but certainly not something too world-changing.
The IMT-2020 5G standard demands maximum speeds of 20 Gbps (billion bits per second), or around 20 times faster than “gigabit” (1 Gbps) broadband connections. The LTE standard demands maximum speeds of 300 Mbps, but even if you’re a few hundred meters away from a cell tower you can expect maybe a tenth of that in practice.
As such, 5G’s increased speed and this added speed’s benefits for things like self-driving cars is completely over-hyped. You can’t expect more out of the “mobile-broadband” part of 5G than you can out of traditional residential broadband, which itself isn’t much better than 4G LTE.
5G will still have a major impact on society, although you likely won’t witness it directly. The most exciting part of the 5G standard is its focus on new applications of fast, secure, and reliable communications. Specifically, the 5G NR standard breaks down target use cases into three groups:
Enhanced Mobile Broadband (eMBB): This is the traditional cell service that we’ve been discussing so far.
Massive Machine-Type Communications (mMTC): The name of this segment suggests physically large machines, but think of this more as a network of small, inexpensive connected devices.
Ultra-Reliable and Low-Latency Communications (uRLLC): This segment is comprised of connections demanding unparalleled reliability and response times.
I firmly believe that the true impacts of 5G will be felt most in the unification of these services. These three categories have typically required separate network infrastructure: traditional cell service for eMBB, services such as LoRaWAN and Sigfox for mMTC, and dedicated ethernet/fiber links for uRLLC. 5G represents the grand unification of all these disparate parts into one network structure.
Massive Machine-Type Communications
This segment represents what many traditionally consider the “Internet of Things.” Everything from sensors in tractors to your connected toaster needs to be linked to the internet, and there are currently a variety of ways to do this. Traditional Wi-Fi works great in home settings, but battery-powered devices have many requirements that Wi-Fi and cellular technologies cannot meet (namely power consumption and range).
Technologies called LoRaWAN and Sigfox are looking to fix this with Low-Power Wide-Area Networks (LPWAN). You can think of these networks as similar to traditional cell service just with support for longer range communications at much lower power consumption and data transfer rate than regular cell service. While similar in purpose, however, none of the underlying specifications of these LPWAN networks are shared with 4G standards. This means that companies need to specifically invest in LoRa or Sigfox modems to get their data to the internet. Additionally, network coverage is primarily limited to major US cities.
We now enter the arguably more sexy segment of 5G service, which is enhanced network capabilities for reliability-focused devices. We can first begin by addressing network latency.
Latency is the time it takes for a signal to propagate through a network. This is not synonymous with data transfer rates, although the two are somewhat correlated. Latency in the case of cellular connections is also not caused by physical distance to the transmitter. The electromagnetic waves ferrying cute photos of your dog to the nearest cell tower travel at the speed of light, so even standing a mile away from a tower will only add a few microseconds to the travel time.
Latency in networks is most often measured in milliseconds. For traditional network services distance to the server in question actually does make a noticeable difference, although again much of this delay is attributed to the increased number of computers required to process long distance requests.
Latency in the case of cellular networks is still measured in milliseconds — this most often refers to the time it takes for the data from a phone to propagate through the cell tower’s communications hardware and into the network. 4G achieves a typical latency of 20ms, while uRLLC is spec’d out at less than 1ms.
Combined with increased network reliability 5G suddenly becomes the ideal platform for a host of interesting applications. Take, for example, autonomous vehicles. With 4G you could at best send a packet of data from one car to another in 100ms — around 50ms to get the packet processed and sent to the cell tower and another 50ms to get the packet to the destination car. If this packet of information is relaying a crash up ahead on the road to other cars around it then this latency is much too high.
In contrast, 5G NR will support a direct communication modethat enables these vehicles to share information in only a handful of milliseconds. Engineers can extend this same principle to ensure safety and enhanced features in industrial environments as well.
Bringing it All Together
In summary, 5G is more than just an incremental bump in speed; 5G represents the grand unification and expansion of network infrastructure that will connect our society for years to come. Over the next five years, I encourage you to follow 5G’s adoption and see how it impacts our world.