Wi-Fi, or local wireless, is undergoing a major advance, to Wi-Fi 6. Cellular, or mobile networking, is getting upgraded to 5G. These improvements will open new opportunities for businesses and new capabilities for users. The upgrades to local and mobile networking will serve us in different ways, as they have been optimized for different use cases.
How are these use cases different? There are three key dimensions that influence wireless design. They are: Physics, Economics, and Human Behavior. Understanding these factors helps us see how we can best leverage the evolving wireless landscape.
The Landscape of the Airwaves
As of early 2019, the new local wireless standard, Wi-Fi 6, is already incorporated into some mobile devices. By the end of this year, it will be built in to nearly all new medium-tier to premium smartphones, tablets, and notebook computers. Similarly, Wi-Fi 6 access points (APs) are expected to become widely available by later this year.
Wi-Fi devices and services are optimized to provide excellent indoor coverage, even in high-density situations with high data rate demands.
5G is also starting to be supported by premium smartphones (the first company to announce a 5G phone in the U.S. rolled it out in an upgrade model above their flagship phone), and its availability in devices will increase over the next 2 years. 5G base stations are also beginning to be deployed, in select markets, and 5G coverage will continue to expand.
Cellular service, in comparison to Wi-Fi, is designed to provide ubiquitous outdoor coverage, including roaming between service providers and across countries.
Given those basics, let’s look at the three factors we’ve identified, to form a holistic understanding of the forces that influence wireless design.
All wireless transmission systems obey the same laws of physics. But when we optimize engineering to take advantage of one aspect of the physics, we often have trade-offs in another area. In wireless networking, there are constant trade-offs between the speed of data transmission, range of the signal, its propagation within buildings, the size of the antennas, and other attributes.
In terms of choosing a radio frequency for transmission, there are two basic parameters: the radio frequency band of the transmission, and the width of that band, generally referred to as the bandwidth, that is available for transmitting data. Generally, lower frequencies are better for propagation: Radio and broadcast TV, which occupy frequencies from about 500 kilohertz (KHz) to about 800 megahertz (MHz), propagate long distances and into homes or buildings. But broadcast’s use of these frequencies means they’re unavailable for modern data networking. Wireless data protocols generally start right above broadcast: the first cellular band was 800 MHz; and Wi-Fi bands starts at 2.4 GHz.
Fortunately, at higher frequencies, there is generally a significantly larger bandwidth available to transmit data. Higher bandwidths allow you to modulate the signal more rapidly, which means you can send data quicker – and as technology advances and the richness of applications increase, faster speed is something that almost everyone needs.
Higher frequencies have the added benefit that they require smaller antennas (because antenna size needs to be on the order of the wavelength, and as frequency increases, wavelength decreases). This is a significant benefit for miniaturization. Therefore, operating at a higher frequency provides two benefits: often more bandwidth available leading to faster transmission speeds, and smaller antennas leading to smaller devices. But there’s a big trade-off: As frequency increases, the capability of the signal to penetrate into buildings decreases. A cellular or mobile signal that comes in clearly outdoors might be weak once you walk indoors.
Today, you may not even notice that you have poor cellular signal strength indoors, since activities like talking on the phone and reading emails do not require significant bandwidth. But your phone notices: if you are indoors, sending a packet to an outdoor cellular base station generally requires far more power (at the expense of battery life) than it does to deliver that exact same packet if you were outdoors with better reception. In addition, your data rate will be significantly lower, too. Deep indoors, mobile devices will not receive close to their rated best speed.
These physics-driven trade-offs were key motivators for the design of Wi-Fi, as well as the economics considerations which we will discuss next.
Despite the fact that 5G and Wi-Fi 6 use similar technology (similar methods for encoding data into radio waves, and even similar ways of scheduling when radios transmit and receive), the economics of building out cellular vs. Wi-Fi systems are dramatically, even diametrically, different from each other. This is because they are designed to serve different use cases – both for the users and the people who run the networks.
On the network operator side, it’s massively expensive to build out a cellular infrastructure to provide ubiquitous coverage. Major US telcos indicate they will be spending tens of billions of dollars in 2019 as they upgrade to 5G. Likewise, the carriers have invested billions in rights to use the licensed frequencies of cellular. That’s the way massive communication infrastructure works, economically: A small or piecemeal approach doesn’t work.
To cover the expenses and generate a profit for the companies who build it, the cellular network is paid for directly, a tiny bit at a time, by its users through monthly and metered subscriptions. And the more they use the infrastructure, the more they can be charged.
Local wireless network equipment, in comparison, is incredibly inexpensive. Since Wi-Fi routers are bought outright and for small amounts, their use doesn’t have to be amortized by the bit. Wi-Fi can also provide great value to users even if it’s only installed in one place; it doesn’t require a large geographical or global footprint to earn its keep. And Wi-Fi also uses unlicensed frequencies which do not require equipment makers to bid for rights to use them.
Additionally, new Wi-Fi 6 devices can still communicate with APs that run older versions of Wi-Fi. Therefore, the installed base of Wi-Fi access points in indoor spaces can continue to be leveraged (of course, new Wi-Fi 6 APs will provide better performance). In developed economies, Wi-Fi is the accepted indoor data transfer system, and it’s still growing: our research shows that the number of public-access Wi-Fi hotspots will grow over 4.5x between 2017 and 2022, to 549 million access points installed. Wi-Fi is cheap, widely deployed indoors, and stable.
Cellular standards are also different from Wi-Fi in that building and selling a cellular product requires a significantly higher license cost for the technology itself. For Wi-Fi, which is based on IEEE standards, the per-device cost for the associated licenses is dramatically lower than for LTE/4G or 5G products. This is why you can buy an entire computer with a Wi-Fi radio built in for $10; but simply adding an LTE modem to an iPad adds more than $100 to the price.
Wi-Fi remains inexpensive because Wi-Fi requires neither the licensing fees of cellular nor the expensive all-or-nothing infrastructure build-out. This low cost is critical, because the Wi-Fi economic model does not include a monthly subscription fee.
The cellular network would never have been possible on Wi-Fi’s economic model. And Wi-Fi wouldn’t have become as pervasive and powerful if it had the cost structure of cellular. The two different economic models have made these two networking systems successful in their two different use cases.
But neither technology nor economics drive how people need to use wireless products. Rather, the technologies, and the business models to support those technologies, were developed to meet the needs of the people who use them. Which brings us to the third major dimension for how wireless products are built.
The biggest reason our two major wireless networking technologies, cellular and Wi-Fi, are different is because people have different needs. The two networks have evolved to serve those needs.
Mobile users – people who are using a portable device while on the move between places – require above all else a consistent, persistent connection. When you’re walking from place to place while making a phone call, or answering email from a bus, or streaming a podcast while driving, you are intolerant of gaps in your network coverage. You’re also unlikely to be consuming a lot of bandwidth; at most you’ll require 2 to 20 Mb/s for HD or 4k video. That’s not insignificant, but that’s the outlier requirement; often the video resolution needed on a mobile device is lower, leading to lower bit rate needs. Furthermore, most everything else people currently do on mobiles consumes far less data. People using their devices while mobile mostly snack on data: They consume (and send) small bits of data, from time to time.
The cellular infrastructure supports this behavior. It is good at keeping a device connected persistently while it moves over a potentially large geographical area. And while it’s reasonably-priced for data snacking or most streaming media, it’s not financially efficient for more intensive or continuous uses of bandwidth.
Nomadic use is different. We are nomadic when we take our device, sit down, open it up, and do work. When we are nomadic, we might be using data intensively – receiving large files, editing them, and then sending them back online. Or taking part in hour-long multi-person video conferencing sessions. In these cases we need a fast, stable connection, and we’re likely using so much data that paying by the bit doesn’t make sense.
The same users, on the same devices, generally use data differently when they are working as mobile users vs nomadic. One major handset manufacturer tells us that 90% of the data their handsets send and receive goes over Wi-Fi – even though their devices spend far more actual time connected to cellular.
Our behavior and use case needs are what drive the business models behind our networks, which of course must incorporate the various trade-offs in physics and economics.
The physics/economics/behavior model also works when looking at other needs and other networks. Physics, economics, and behavior (of not only people but also apps, devices, processes, etc.) can explain why there are unique networks for IoT (e.g., Zigbee, Thread), why a new high-frequency extension of 5G (called millimeter-wave) may serve the needs of fixed wireless access very well, and why a new “lightly-licensed” offshoot of LTE called CBRS may serve the needs of industrial plants.
Our behaviors and use cases, coupled with physics and economics, have defined Wi-Fi and cellular networks to date. The major advances coming with Wi-Fi 6 and 5G cellular will open up ever more opportunities and surprising applications for all of us.
To learn more about the future of Wi-Fi 6 and 5G, sign up for our ‘Wired for Wireless’ virtual event on 4/29.