Why don't IoT just use Wifi?

Why don’t all IoT just use Wifi rather than these devices using Zigbee, Z this, and X that? My Nest Home Doorbell, Google Hub, Google Home, Google Home mini, and even my Apply Watch has wifi built in. That would seem to solve the compatibility issue.


Not sure entirely. One difference is that unlike Wi-Fi - ZWave, Zigbee and others are mesh networks that do not require any kind of central communications device, like WI-Fi requires a central router. There may be other reasons. But it sure would be great if they would standardize on something no matter what it is.

But getting companies to agree on standards is not a technological issue, more of a political / competitive issue.


The large companies want you in their eco system so they make it so their stuff works best with their stuff. The smaller companies try to make the stuff convenient so they use standards that are simple like wifi. @dking is right that they could make everything work with everything else on 1 standard but every large company wants to do it their way so that becomes the issue. Also part of the problem with everything being wifi is that is a lot of traffic that your wifi router will have to deal with and can slow or even stall out your router when too much stuff starts interacting at the same time. That’s why a lot of serious IoT people will build a separate network just for IoT stuff. Though our technology is getting a lot more capable of handing a lot more stuff like Mesh WiFi it can still cause a lot of issues when you have a lot of IoT. Just my 2 cents.

Two reasons may be range and power consumption. As mentioned, Zigbee, for example is a mesh system so all devices are connected to each other, increasing a device’s coverage, although all the commercial implementations still need to connect back to a hub eventually.

Zigbee and Z-Wave have low bandwidth - basically just control signals. A video doorbell that connects via one of these protocols would have to fall back to Wifi to transmit video.


It depends on what it is. If it’s a doorbell or thermostat it is wifi. If it’s something which requires very low bandwidth and low power then wifi is horrible. Wifi in general has very poor signal characteristics and it’s power consumption is through the roof. For example, the common “vision” of iot is something small which can run on a coil cell for x years. You’d never do that on wifi.
If you go beyond “home grade” wireless signals and look at something like LoRa or sigfox it makes more sense. LoRa for example, don’t quote me on these numbers, can do something like 15km line of sight to a device powered by a battery and lasting in the order of several years. You might be reporting data once a day or once an hour but that’s what all this stuff is about.
The other thing is yes, mesh. Mesh is a very very cool and powerful idea it’s just that iot isn’t yet at a point where you look at and say that makes sense. Mesh networks with Self healing multi pathing between many battery powered devices. No wiring, no setup really, turn it on and go. That’s the idea anyway.


Mesh networking is also used by some IoT setups which use cellular comms to get back to a control server, to cope with areas where no cellular service is available. A swarm of devices will pass packets of information between them until they reach one of them which has cellular reception, which will then forward the data to the central server.

Power is the big problem, months or years of power on Zigbee/Zwave or a few hours or days on Wi-Fi.

Given the lack of security in many IoT devices, so that they can act as a bridgehead into your network, having them on your standard Wi-Fi is never a good idea. When I set up my home network, the IoT was banned to its own VLAN, which doesn’t have any access to the local networks.


I have always been fond of the saying “The ‘S’ in IoT stands for Security…”


If WiFi HaLow (802.11ah, low frequency for longer range and lower power) ever catches on then many more devices will be able to use WiFi instead of the other radio systems. Of course it will mean you need a new router with a transceiver that supports HaLow.

I wonder if that will actually ever catch on based on the fact that it’s part of the WiFi alliance and probably closed source. A lot of the iot wireless stuff is developed by small devs on platforms such a particle.io and by definition that’s all cheap and open source (although I’m not sure if particles mesh protocol is open source). Anyway, the point being that I think wifi has a pretty stuffy and crappy reputation.

There is no controlling source code from the WiFi alliance for any of the versions of WiFi so none of the versions are open source or closed source. They are just technical standards with no licensing required unless you want to use the trademarked logos. This is why there is a huge ecosystem of devices that use WiFi. Some other IoT connection methods are proprietary and require licensing but the most successful ones are open at the application level like WiFi.

Bluetooth = 802.15.1
ZigBee = IEEE 802.15.4 (also used by ISA100.11a, WirelessHART, MiWi, 6LoWPAN, Thread and SNAP)

WiFi 1 = IEEE 802.11
WiFi 1 = IEEE 802.11b
WiFi 2 = IEEE 802.11a
WiFi 3 = IEEE 802.11g
WiFi 4 = IEEE 802.11n
WiFi 5 = IEEE 802.11ac
WiFi 6 = IEEE 802.11ax
Wi-Fi HaLow = IEEE 802.11ah

Wired Ethernet = 802.3


Partly correct, the security protocols are proprietary and you have to sign up to them.

That was the problem with Wi-Fi 6, it was developed behind closed doors by the same Wi-Fi Alliance muppets that brought us WEP. The security was not scrutinized by security experts until the first products were released… Then a couple of days later, the security experts were lampooning the specification for its flaws…


There are mesh routers out there for Wifi. Couldn’t wifi be used as a mesh and improve coverage for IoT?

Wifi uses a lot of power. I think Zigbee and Z-Wave are more akin to Bluetooth in power consumption. Also IP is probably too heavy a protocol for some devices, such as sensors.


I have some ZWave devices. I prefer them over wifi because my zwave hub is my alarm system. It stays with the house. If all my stuff was wifi, anytime the wifi changes it will require reconfiguring the device.


How much is the fee and who paid for them to be used in every FLOSS operating system on the planet?

Dragonblood attack

In April 2019, serious design flaws in WPA3 were found which allow attackers to perform downgrade attacks and side-channel attacks, enabling brute-forcing the passphrase, as well as launching denial-of-service attacks on Wi-Fi base stations.

From Wikipedia.

The actual WPAn encryption is generally performed in hardware on the wireless adapter/chip and can be patched with firmware updates.

AFAIK, the OSS generally just needs a driver that can communicate with the adapter, it doesn’t have to worry about the signal, security protocols etc.

Edit: Certainly the fixes for things like KRACKER were new firmware from the different manufacturers.

The answer is in the question.
Using existing standards would always be cheaper and more convenient for manufacturing and users, but you would then allow the users to use someone elses tech which cannot be tolerated.

You will probably find something written into the Apple law book that covers it, though it may be written in Ferengi. Something about the rules of acquisition I think.


Ah, yes I was aware that manufacturers of WiFi chipsets have to license the patents for the embeddable algorithms used for security. IIRC, the WiFi alliance has bundled the various manufacturer patents into a single portfolio for chipset makers to license at a flat rate vs. the normal process of each manufacturer having to negotiate a license with each company that has an applicable patent.

Of course the issue I was addressing was the post making it seem there was some thorny licensing issue for hobbyists with using any old or future versions of WiFi (it will be many decades before hobbyists can afford to contract a semiconductor fab to produce their own WiFi chipsets)

As someone just techy enough to be dangerous, I’d love to adopt IoT devices into my home…especially as I get older and mobile challenged. I refuse to do so until I can get all the devices I want and know they will work together. Why any company would assume customers will understand that you need this system for these devices but that system for those…it’s beyond stupid reasoning!

So I, and MANY like me, spend nothing and wait until the companies quit trying to saddle me into a system that doesn’t work across the board!