The Road to WiFi 6E Part 4: Device Density

Published on May 9, 2022

This blog post is the fourth in a series of articles we’ll be presenting over the course of 2022 about WiFi 6 (and 6E) – the current and future top-of-the-line WiFi standards championed by the WiFi Alliance. We’ll be discussing the history of the evolving standard of WiFi, new features introduced in WiFi 6 and 6E, applications beyond the commonly known consumer use cases, and much more. Additionally, we’ll be releasing a companion series of video interviews with Ezurio (formerly Laird Connectivity) experts, which you can find here.

The Internet of Everything – Explosive Growth in WiFi Deployments

For many years, the world has moved steadfastly towards creating connectivity in every kind of device. The vision of the Internet of Things has resulted in wireless connectivity in connected medical monitors, robotics, vehicle telematics, access control, wireless sensors, and everything under the sun. This is enabled by adding wireless hardware to an increasing number of classes of devices – and those devices all need service from the wireless gateways that serve them, sometimes in incredibly congested environments.

In our last post, we briefly looked at the connected hospital room as an example. Just a single room in an ER bay can have dozens of devices, a challenge to service all by itself. But when you consider that hospitals are full of floors of rooms much like this, the scale of the challenge begins to take focus. In factories, healthcare facilities, sports arenas, travel terminals and more, the density of WiFi devices continues to increase at an exponential rate.

WiFi 6 and 6E have this problem in mind as they harmonize previous standards, introduce new means of organizing spectrum temporally and physically, and adopt new models for how to categorize and group client devices. And as we’ll explain, many of the mechanisms we’ve discussed in our previous blog posts on link rates and low latency are instrumental to supporting an incredibly RF-dense environment, via MU-MIMO, spatial streams, OFDMA, and (new to this series) BSS Coloring.

Using the Space with Spatial Streams, OFDMA, and BSS Coloring

For an example of a crowded RF environment, let’s use the example of the office floor to visualize device density. The ordinary office is filled end to end with devices that utilize WiFi: laptops, scanners and printers, room displays, personal devices, and more, all packed densely into cubicles which are themselves designed specifically TO be high-density. The more devices enter the workplace, the more infrastructure has to keep up. And because the actual real-world space is so constrained, older implementations of WiFi would explicitly involve a lot of RF chatter talking over each other.

Part 1: Optimizing Physical Space

As we discussed in our blog post on WiFi 6 link rates, one of the ways that the infrastructure can minimize RF congestion is by utilizing Multi-User Multiple-In Multiple-Out (MU-MIMO) spatial streams to focus RF activity in the physical direction of the intended device(s) when it’s their turn to communicate. This is achieved by using two antennas and creating an intentional interference pattern to focus signal toward the intended device or groups of devices.

If we start with specifically creating targeted areas for RF signal, we already eliminate a huge amount of wasted energy, signal sent in every direction except where it’s needed. This allows an AP to actually use less power to achieve a strong RF link with a device. In WiFi 6, as many as 8 spatial streams are possible, which allows access points flexibility to target many devices nearby with precision.

In addition the controlled directionality of the beam forming creates a cleaner RF environment, and uses less power: a beamformed transmission within a spatial stream achieves a better link with less power. As we’ll see down the line, that’s also helpful for not talking over traffic on nearby access points, which may be using the same channels and which may pick up these signals as interference.

Part 2: Optimizing RF Space

Now that our access points have reduced the physical area of their communications, we already have a more efficient footprint for WiFi in the space we’re operating in. Neighboring access points have a lower concentration of competing RF activity, and the whole RF space is cleaner. Between multiple access points, this prevents the whole area from being saturated with unintended signal.

But now, let’s look at the RF frequencies in use within that physical space. WiFi has long used channels to allow simultaneous, non-interfering communications to take place in the same area. By operating on different frequencies, access points avoid each other’s traffic and ensure a cleaner link.

In our previous post on low latency in WiFi 6, we discussed how access points can utilize a new feature, Orthogonal Frequency Division Multiple-Access (OFDMA), to allow multiple clients to talk to the access point at the same time. The short version is: each chunk of time in a WiFi link (a frame) can be used to send and receive multiple messages across different channels (frequencies) to and from different devices. Rather than all devices waiting in line for their turn to send or receive, they do so at once, but in non-interfering channel spaces, which means they all enjoy much more immediacy to their communications.

If you can imagine multiple access points all doing this at once with their connected client devices, you can see how WiFi 6 and 6E will begin to further resolve the device density problem. We’re now sending targeted spatial streams from multiple access points, AND those streams are capable of supporting multiple devices in every single frame of WiFi traffic. It’s a target within a target, on some level, for the access point to talk to an individual device in the direction it’s located and at a specifically band of RF space. We’re getting to a high degree of precision in how the client and access point talk with one another, and the results are clarity for the surrounding devices. No longer is every device shouting in all directions and waiting for its turn to do so. We’re able to achieve a previously-impossible kind of coordination of devices on the network.

This results in a greater number of devices being able to operate within the same close proximity than previous generations of WiFi, not enabled with OFDMA, could hope to achieve.

Part 3: Categorizing the Space with BSS Coloring

The previous two mechanisms we have discussed before, in earlier in the blog series, but now we move on to something new in WiFi 6 / 6E, which is an added layer of intelligence around which AP the client means to communicate with, or is assigned to.

To start: A Basic Service Set (BSS) is the basic network topology where devices talk to an access point (AP). That sort of cluster of devices around an AP is the fundamental unit of a WiFi network. When multiple APs are located within RF range of each other, they try to avoid interference by communicating with devices on a different (ideally nonoverlapping) WiFi channel. Let’s use 2.4 GHz WiFi as an easy example: There are three nonoverlapping 20 MHz channels available in the FCC in the 2.4 GHz range: Channel 1, Channel 6, and Channel 11. If you’re attempting to cover a floor plan with access points, you will assign APs to non-overlapping channels so that they don’t interfere with each other. However, as soon as you have more than 3 access points, you’ll need to reuse channels.

As soon as that happens, we encounter problems, just based on the nature of how WiFi traffic works. If a device hears another device talking on the same channel at the same time, it must stop and listen until it’s safe to communicate. That means that devices on either of the Channel 1 APs are bound to encounter collisions and spend more time than is ideal waiting for their turn to broadcast.

These two Channel 1 APs are what is called “Overlapping Basic Service Sets” for this reason. BSS coloring attempts to resolve this conflict by allowing the AP and devices to differentiate between AP’s on the same channel by including a prepend to their broadcast identifier in the form of a color. It’s a binary value added to the BSS that figuratively corresponds to a color, helping to provide a clearer picture of how things work. If a device hears a broadcast on its channel, but the broadcast is of a different color, the device can determine that it’s for another service set and go ahead and broadcast without waiting. In doing this, we can categorize devices into association or affinity with a given access point with a high degree of likelihood that they will not interfere with each other, due to proximity. Devices will talk to their nearest AP without fear of interrupting another service set. And as we’ll explore in a later post, configurable Target Wait Time (TWT) further decreases the chance of interference between overlapping BSSs, further facilitating high client density.

This example is simple enough considering four access points, with a handful of devices and on three overlapping channels. But it can become incredibly complex and nuanced for building WiFi topologies across 5 GHz or 6 GHz channels, covering large areas and serving a massive number of devices. APs can serve up available channels in many ways, providing lots of 20 MHz channels or group them together into 40 MHz, 80 MHz, or 160 MHz-wide channels. With 63 available BSS colors and 88 non-overlapping 20 MHz channels, there are over 5500 unique combinations of channel and BSS color alone. This means that BSS coloring gives an exponentially greater way to prevent collocated service sets from slowing each other down. What this all produces is an exponential increase in flexibility, with large channels for data-heavy applications, small channels for ordinary traffic, and tying clients to APs to prevent interference even if other APs are using some of the same channels nearby.

The result of BSS coloring is that more devices can be placed within a smaller area and exist without being impacted by closely co-located AP’s utilizing the same spectrum. Providing a net device density increase when combined with OFDMA.

Putting it All Together

Between spatial streams, OFDMA, and BSS coloring, we arrive at a network built of very targeted access points, talking in narrow areas of physical space, with carefully managed sub-channel frequencies to keep them distinct, and with an organizing schema to keep them talking just to their nearest access point without shouting over traffic. It’s a huge jump in intelligence available to the infrastructure and in keeping the RF space overall as clean as possible. And via these features, we arrive at a WiFi implementation that can handle an overwhelmingly higher number of devices with much more care than was ever possible before.

As we discussed, there are many applications which are served by this, applications with lots and lots of devices collocated in physical space:

Smart Factories and Facilities: These are notoriously difficult RF environments with lots of RF-reflective surfaces and open floor plans which lend to traffic spillover from different areas. But they’re also typically very large, meaning that RF spillover may arrive very quietly at the unintended access point. Using spatial streams to target devices reduces ambient RF, and BSS coloring can allow an AP to disregard interfering traffic that may come from far away and very faintly, posing minimal risk to the AP and its devices.
Medical and Healthcare: As previously noted, hospital rooms can contain dozens of devices, and there are many rooms per floor and sometimes many floors of rooms, meaning RF traffic can overflow from different rooms or different floors very easily. It’s especially helpful in an environment like this to use BSS coloring, since it’s not just the lateral floor plan that is dense with devices, but also the vertical space between rooms. Interference can arrive through walls, floors, and ceilings. BSS coloring takes a very complicated issue of closely-packed clients and access points in three dimensions and helps separate their traffic.
Outdoor venues: Large outdoor venues have a similar issue, and in a way are a worst-case challenge for WiFi installations: Nearly every attendee is likely to arrive with a cell phone, and be packed into the smallest possible space. It’s hard to imagine a higher-density environment than a stadium for wireless devices. Those 5500 combinations of channel and BSS color, in addition to using sub-channels in OFDMA, provides a huge configurability scheme that ensures access points and their devices do not interfere with each other.

Conclusion

Increasing the possible density of WiFi clients around an access point and the density of access points in a deployment is a mission-critical upgrade for WiFi. As a technology, WiFi has proven to be as essential as expected. And these improvements are a huge leap forward to support the growing number of WiFi devices in the future.

As the applications for WiFi continue to expand the number of points of use within a location is getting higher and client density is becoming an increasing concern, especially when looking at connectivity, performance and latency. WiFi 6 has addressed this issue by providing a more efficient version of WiFi to enable the increase is client density expected in the future.

All of these WiFi 6 features translate to a better user experience. And in practice, the only thing worse for user experience than poor connection is poor power usage. Battery-powered devices are a hugely-growing portion of the WiFi ecosystem, and poor management of the limited power available means a device that’s constantly in need of charge. The addition of Enhanced Low Power support gives devices and access points the opportunity to set sleep and wake times, which has advantages to saving power on the client as well as reducing congestion at the infrastructure level.

Subscribe to our blog series to get updates on our next post and many more as we continue our 2022 series on WiFi 6 and 6E.

Learn more on these topics:

WiFi 6E