New wireless connectivity: it’s not just 5G
While the 1990s connectivity revolution was mostly “wired”, the new connectivity is all about high-speed wireless. In this arena, few technologies arouse as much controversy as 5G. For some, it’s analogous to the second coming of Christ, while for others, it’s the work of the Devil (or at least Bill Gates — who allegedly wants to inject it into us as part of the COVID vaccine). The circuit diagrams published to expose this conspiracy appear to be nothing more than guitar pedals, which will hardly slip unnoticed into a hypodermic needle. In more serious circles, 5G has become a proxy for nation-state technological supremacy, with three main suppliers of the technology competing to win the deployments that have now begun (Huawei, Nokia, and Ericsson).
Previously, each generation of mobile technology was enabled by a new use case (a.k.a. the “killer app”):
- 1G: Voice
- 2G: Text
- 3G: Web
- 4G: Video
Each new generation enabled faster speeds and reduced latency. 4G in particular also replaced the “circuit-based” (phone) foundation with a more modern “packet-based” (Internet) approach where everything (including voice & video) is encoded as packets of data. As even faster download speeds and lower latency will do little to further improve the viewing of video on small mobile screens, 5G has been on a quest to identify the next killer app.
New use cases that have been suggested to drive 5G adoption include: telesurgery, connected factory machines, wireless virtual reality, smart cars (Vehicle-to-Vehicle (V2V)), and smart traffic lights (Vehicle-to-Infrastructure (V2I)). Most of these however, can be implemented using simpler, more reliable, and cheaper alternatives (a variant of Occam’s razor). For example, vehicle collision-avoidance (a use case for V2V) can be (and is being) implemented simply using cameras or sonar. Also, anywhere that a machine can be powered using an electrical cable a regular Ethernet cable will provide faster and more reliable data connectivity.
The ultimate use case is the Internet of Things (IoT). Billions of smart devices will need connectivity and 5G has features that have been designed specifically with this in mind. Unlike Tom Cruise, IoT doesn’t care that much about speed. It cares more about range (distance from the device to the access point), low energy consumption (many mobile/remote devices operate on battery power), and high fan-in (how many devices can talk to a shared access point). Many IoT applications aren’t very chatty: they work well with kilobits/second, not the multi-megabits/second that video requires. Above all, IoT requires pervasive coverage. This is a problem with each increase in G: the price of higher speed is reduced range. Each generation required a higher density of access points to reduce the distance to the mobile device. 5G will require twice as many access points to cover a given area than 4G. This is good for the makers of 5G technology but not so much for customers looking to connect devices spread out over sparsely populated areas. So IoT applications have responded by making their devices bulimic: they ingest data as they go and regurgitate it later when they get within range of an access point. Tesla cars sip 4G data when on the move (for navigation) and defer the uploading of much more voluminous video captured on cameras or large software updates when they arrives within range of a home WiFi network.
Nonetheless, 5G enthusiasts are confident that they’ve now found the “one ring to rule them all.” This version handles high speed, urban, rural, licensed, unlicensed, and energy efficient deployments in a highly secure way. Thanks to geopolitical pressure, Huawei has been knocked out of the running in many western markets, so the 5G bonanza has become a two-horse race for Ericsson and Nokia. Despite all this, Wall Street isn’t convinced. The chart below shows both companies trailing badly these past 5 years vs. the overall NASDAQ technology index.
And yet, 5G will succeed. It’s not the second coming, it’s more incremental innovation, using Silicon Valley’s variant of the Olympic motto (citius, altius, fortis): “faster, cheaper, better.” All new phones will come with 5G inside. Instead of a single killer app, several perhaps more modest, applications will emerge:
The first connected device: just like our phones come with a physical (or now increasingly virtual) SIM card, more and more devices will come with “connectivity inside”. This is useful when this is the first device to be deployed in a factory or building. There’s no need to string a new network just to get this one device connected. Once you start having several, a local area network and a gateway will create a more efficient way to connect its friends.
Autonomous mobile machines: drones and delivery vehicles will benefit from “always on” connectivity to help guide them to their destination and to permit remote control in case they get stuck or need help.
Mobile broadband: already with LTE, many consumers found that a cost-effective way to enable broadband internet was via a 4G router. With 5G (at least in urban or densely blanketed areas), it may be even more useful to ditch cable or DSL in favor of 5G backhaul to the Internet.
Alternatives to 5G that are better optimized for IoT have been tried. Sigfox and LoRaWAN are some of the more recent ones. They’ve both struggled to establish significant market share. Meanwhile, the perennial scrappy underdog of wireless networking, WiFi continues to grow and adapt. In its most recent incarnation, WiFi 6E, it gains additional speed, reduced latency, better energy efficiency, and higher fan-in (as well as access to more frequencies). While each generation of mobile technology required licensed spectrum to operate, WiFi operates in the “wild west” of the unlicensed frequency bands (2.4 GHz, 5 GHz, and now, 6 GHz). With LTE (4G), the mobile suppliers decided to offer variants that would operate in the unlicensed frequency bands as well — to better compete with WiFi’s success. Unlicensed spectrum lets anyone (or any company) quickly deploy a wireless network without requiring prior permission (a license) from the regulators. Thus far, however, WiFi continues to rule the unlicensed bands and very few 4G or 5G deployments use these.
What has held Sigfox and LoRaWAN back has always been a “chicken and egg problem”: it only makes sense to deploy lots of access points when you have lots of devices deployed. But customers will only deploy lots of devices when they know they can get connectivity via the widespread availability of access points. There are two new contenders that are solving the problem of ensuring pervasive connectivity in radically new ways: Amazon Sidewalk and SpaceX Starlink.
Sidewalk operates in the 900 MHz spectrum and provides a range of up to 800 m (half a mile). That by itself isn’t novel. Every new Amazon Echo device is a Sidewalk Bridge (access point) that connects it to the customers’ broadband network and shares some of that bandwidth with Sidewalk devices connecting through it. Given the pervasive adoption of these Amazon devices, this means the Sidewalk network will have incredible coverage (at least in places where Amazon customers live). Devices to help track your posessions, like Tile, which today rely on Bluetooth and therefore disappear when beyond range of your phone, if operating on Sidewalk, would remain trackable wherever they go. This could help reduce the theft of bicycles in urban environments, for example.
At the other end of the spectrum is Starlink. Low earth orbit satellites provide high-speed connectivity (10s to 100s of megabits/second) anywhere on the planet at (relatively) modest cost ($99/month) to a small satellite dish. This has always been a huge barrier to the mobile providers: deploying cellular networks in sparsely populated areas remains unprofitable. The low orbit reduces latency (a big issue with geostationary satellites) and also reduces very high fan-in (allowing for higher speeds). For stationary applications, Starlink will offer unmatched high-speed connectivity in rural and outlying areas.
For IoT, connectivity is an enabler. The value ultimately resides in the insights and decisions that are made based on the data that is transmitted. We’ll have many ways to get devices connected. New connectivity technologies that can achieve ubiquitous coverage such as Sidewalk, Starlink, and 5G will provide new options. This new wave of pervasive, wireless connectivity will provide inexpensive connectivity to anyone and anything, anywhere on the planet.
More articles
Return to the main page from which you can find more articles like this.
