Wi-Fi versus 5G? Nope, it’s both

Article By : Brian Santo

The question comes up again and again: With 5G, what will you need Wi-Fi for? Everything you need Wi-Fi for now, actually.

Will 5G replace Wi-Fi? Once you get access to wireless connectivity that is faster, more robust, and has greater capacity than Wi-Fi, why would you need that and Wi-Fi? Why wouldn’t 5G replace Wi-Fi? Because it's theoretically possible, people have been asking this question over and over for at least a couple of years.

We asked the folks at Wi-Fi specialist Quantenna about 5G replacing Wi-Fi, and at first they didn’t even understand the question. After discussing it a bit, the reason for the misunderstanding became clear: it makes no sense. If it were to happen, wireless carriers will have to want to make it happen, and they have no reason whatsoever to want it to happen. They apparently aren't even considering the possibility, hence the initial confusion.

The question is really about residential service. Yes, there are public hotspots and, yes, enterprises rely on Wi-Fi too. But the residential market is where Wi-Fi is most widely used for distributing broadband bandwidth among a growing number of devices.

And in the home, “operators are doubling down on Wi-Fi,” said James Chen, Quantenna’s senior director of product marketing. “It’s the logical thing to do.” He reeled off a list of reasons why.

There are thousands of products with Wi-Fi already built in, and thousands more coming. “What happens to all these tablets that have nothing but Wi-Fi, or Wi-Fi and LTE,” Chen asked. They will have to be supported for a very long time. Perhaps some people will get 5G broadband, but the signal is more likely going to go to a gateway and get distributed via Wi-Fi from there.

Say carriers were to decide to replace Wi-Fi, anyway? There are 90 million to 100 million homes that would require femtocells. Even if operators wanted to do it – even if consumers demanded it (for whatever reason) – it would take years to deploy that many devices. And consumers would continue using legacy devices, so 5G and Wi-Fi would still have to co-exist for many years after that, Chen explained.

There is no compelling technological reason to replace Wi-Fi. It is getting faster, more robust, and capable of supporting an increasing number of Wi-Fi devices simultaneously, with repeaters becoming more common and mesh networking on the way in the forthcoming IEEE 802.11ax generation of Wi-Fi technology, which is being branded as Max Wi-Fi.

5G signals are a shared resource. If you have a hundred homes served by a single base station, that 1 Gbps gets distributed, averaging out to 10 Mbps to each home. “But if people want gigabit service, they want a full gigabit,” Chen observed. Gigabit service is still best provided by cable or DSL, which will get distributed throughout the home by Wi-Fi.

Those conditions likewise apply in the enterprise market, which is unlikely to give up Wi-Fi any time soon. How about urban access, in areas served by public hotspots? “You can do that with LTE now,” Chen observed.

The implication being that if wireless was going to replace Wi-Fi in urban environments, it should already be happening. It’s not, and there’s no reason to expect that a transition from LTE to 5G would change that.

The one place where 5G might end up getting used in lieu of Wi-Fi would be in rural areas where wireline connectivity is inadequate or lacking, but that’s speculative. “We haven’t heard of 5G replacing Wi-Fi anywhere,” Chen said. “In our minds, 5G will never replace Wi-Fi.”

So what’s in store for Wi-Fi? Chen said Wi-Fi equipment developers are in the process of meeting three challenges: more devices in more places, most requiring greater speed, with some requiring significantly lower latency.

“If you ask Alexa or Google Home a question, the answer has to come back in well under a second. People lose interest, or they ask again,” Chen explained. “Even in conversation, people expect no more than about 100 milliseconds of delay. If your assistant makes you wait 1 or 2 seconds? That’s too much.”

He said 802.11ax will address all three issues. It will be faster, and will support mesh networking. It will also support more channels to handle more devices concurrently. Quantenna, for example, is offering 8×8 antennas (8 receive and 8 transmit). Max Wi-Fi will adopt the OFDMA (orthogonal frequency division multiple access) modulation scheme to support traffic scheduling, which will be kind of handy in residential environments but immensely valuable in situations where thousands of users are packed together, arenas and stadia, for example. There are also provisions for spatial reuse.

Source: Broadcom

There is still more work to be done to improve Wi-Fi to complement other connectivity options that include 5G, Chen said. There are several local- and personal area network (LAN, PAN) communications options, and they will have to be made to work with each other.

Wi-Fi operates in the 2.4 GHz and 5 GHz channels, but Bluetooth 5, Zigbee, and Thread – three low-power technologies – are all operating in the same 2.4 GHz band. Wi-Fi is more powerful and can overwhelm the others, but they’re going to have to co-exist. (By the way, Max Wi-Fi has a low-power profile as well, and might end up competing with the other three in some applications).

Chen said he’s seeing gateways specified with all four radios: Wi-Fi, Zigbee, Thread, and Bluetooth. It’s up to infrastructure companies, such as Quantenna, to sort out the coexistence issues.

“If you build something with, say, both 2.4 Bluetooth and 2.4 Wi-Fi, you want the respective chipsets in one box and you want them to talk to each other and get out of each other’s way," Chen said. “Say you’ve got Quantenna Wi-Fi, but the Bluetooth is from Broadcom. A company like Arris or Technicolor will come to us and tell us to work with each other. We’ll do that and go to Arris and tell them we’ve solved it. Because co-existence is at the chip level, you need the chip guys to work it out.”

Brian Santo has been writing about science and technology for over 30 years, covering cable networks, broadband, wireless, the Internet of things, T&M, semiconductors, consumer electronics, and more.

Related articles:

Leave a comment