Julie Jacobson has a really excellent technical article on the recent wave of Wi-Fi IOT devices and how the protocol does and doesn’t fit the use case. Well worth reading.
First, the over-the-air power: Tesla (the man, not the company) would be proud.
Second, the wireless power transmitters obviously won’t be cheap.
Third, all new WiFi devices. Window sensors, door sensors, motion sensors…
Fourth, an infrastructure not designed for this. Again, all new devices.
Fifth, battery backup for wireless power transmitters! Else a power outage kills it all.
As the article points out - all to avoid the hassle of a $50-100 zigbee/zwave hub?
I couldnt have said this better, it’s really sad that the market is being flooded with WiFi home automation devices, but maybe it is because zwave and Zigbee devices are still too difficult for the average consumer to use well and WiFi is more approachable.
I think 2021 will be the year 2.4 ghz WiFi gets a bad reputation cause too many users will have too many devices connected to a single WiFi access point or cheap WiFi devices spam WiFi networks till they are unusable, like the chrome-cast.
But the fact remains: Wi-Fi was never meant to be a home-automation"thing."
For starters, the typical consumer-grade router can only support about 30 Wi-Fi devices, which isn’t much if you count all the smart switches, sensors, light bulbs, cameras and other IoT fare that would comprise a smart-home ecosystem. More advanced routers can support 250+ devices (with VLANs), but consumers wouldn’t know this; nor would they be inclined to splurge on such a thing.
So at some point, they add one too many sensors to the Wi-Fi network and the ecosystem crumbles. Only then do they call in the pros … after dozens of phone calls to every single device manufacturer, every single app developer, every retailer, every department of every TV and Internet service provider, and every neighbor, friend and relative fails them.
The 1st law of smart-home states: The likelihood of ecosystem failure is directly proportional to the number of things that touch the home network.
This is a pretty tough one. People on here that are enthusiasts know why wifi is a bad idea at large scale. But this is all about getting consumers to do mass adoption. A bunch of my friends and relatives have growing collections of IOT devices around their houses. And I can think of ONE house that has a real hub with zwave/zigbee. My parents don’t care about all of the stuff folks on here do. They want the locks, thermostats, lights, and Alexas in both of their houses to “just work”. And, for the most part they do.
It’s really hard for consumer electronics companies to do the right thing when they can do the VERY profitable thing.
Yep. That’s why the article mentions that Leviton is now providing Wi-Fi devices because that’s what customers were demanding.
It will definitely be interesting to see how this plays out. On one hand, every consumer knows what WiFi is and companies are begrudgingly meeting their demand. On the other hand, very big companies are also implementing IoT protocols like zigbee(Amazon, Hue, Samsung etc) and z-wave (Ring (Amazon), Samsung, etc). Ultimately I think the the success of zigbee and/or z-wave is going to boil down to the success of the big companies (Amazon and Samsung) Trojan horsing those protocols I to their more traditional devices like speakers, routers, etc. and then leveraging their “works with…” programs so the consumer doesn’t know what protocol they are using as long as it works.
CE Pro has an undeniable bias in regards to this topic.
They are dealer integrator professionals (Consumer Electronics Professionals) and certainly realize that WiFi based devices are easier for DIY - And that cuts into the revenue and growth opportunities for the businesses run by their readers.
WiFi IoT would be my very last resort… the percentage of me ever doing that can’t be calculated
No video doorbell or outdoor cameras? WiFi is way better for video than the “lightweight” protocols like zigbee or zwave. I just wouldn’t use it for a light switch.
Cameras/Video would be the exception obviously.
And probably voice assistants.
You can see why non-technical people get confused about why everything doesn’t “just work.”
As one of my college professors used to say “obvious isn’t always so obvious.” LOL!
Good article. Thanks for sharing.
I think there are some important use cases that solutions other than WiFi have a problem with:
- Connectivity to/from cloud
- Vendor access to push updates
- Plug and play with consumer hardware
- Relatively Stable network connectivity
If I buy a device, i want it to work out of the box.
If it doesn’t work, i want to be able to call the vendor and get it resolved… Chasing down hub vendor, voice provider, router provider, etc. to find someone that can fix it is too much.
If i am the vendor customer service, i want to be able to see the device, maybe access diagnostics, and maybe push updated firmware. Having to work with hub vendor, router vendor, etc. is too much trouble.I
Every vendor having separated WiFi bridges is the road we are on now to address these issues while using wireless protocols that are better suited to the devices.
I don’t think these problems can really be solved with non Wi-Fi IoT networks that support multiple devices from multiple vendors (zigbee, zwave, etc.) without new protocols that can bridge the divisions we have today between different types of networks.
Interesting thing to me is that you already see threads on reddit with people asking why their wifi sucks or what can they do to help their wifi issues since they got all these outlets. People are starting to see the problem with IP based IOT on consumer level wifi. Even prosumer class stuff has a limit as well. To the point semi knowledgeable people talking about setting up separate wifi just for iot…thats crazy talk.
I dont think z-wave/zigbee either are perfect. At least they are better designed for this environment. I still havent understood why the aversion to ‘another hub’. But maybe I’m the weird one
As I’ve mentioned before, I already do this. But it’s to protect my non-IOT critical services from DoS flooding on the IOT network, either malicious or accidental.
Thanks for sharing this article… really good read. I can relate because I have to confess that I even fell prey to the ease of WiFi when I saw a pack of 3 Tuya smart plugs go on sale one day for under $7 each, so I picked up a pack and started sprinkling them all over my house for Christmas lighting automations. They worked like a charm and were drop dead easy to setup. This is why people love WiFi iOT devices. I predict that routers will get cheaper and smarter so people will migrate in this direction since most don’t have the patience to deal with ZigBee and Zwave and a hub like ST.
Amazon is trying to get ahead of this issue by putting a zigbee coordinator in some of their voice assistants as a stealth hub, then suggesting zigbee devices that will have the same “simple set up“ as a Wi-Fi device work. Amazon can automatically discover them and add them to the network. Not only does this work quite well, they are also adding simple local voice processing for the IOT zigbee devices that will work even when the cloud is not available.
It’s all pretty cool, and they understand as well as anyone else the issues with a mass market device for a non-technical audience.
We’ll see how it goes.
Yeah, I can’t see how Amazon and Google haven’t jammed Zigbee/Zwave radios into most of their stuff. They could just leave it dormant in there until they’re ready to become full-blown HA solutions. Then, one software push and it’s live.
Especially Amazon. Their entire business model is usurping other profitable businesses.
Google bought Revolv a few years back. They had a z/z/wifi/bluetooth stack in their hub. I’ve been waiting for them to shrink this and start jamming it into every Nest/Chrome/Google Home device.
Amazon’s entire retail business model is removing transactional pain points for their customers. (Quick delivery, free shipping, price matching, easy returns, an understandable reviews system, easy open packaging, etc. We forget how unusual all of these things were until Amazon popularized them.)
They won’t add zwave until “Alexa, discover my devices” works with it. Zigbee Home Automation has that: zwave currently doesn’t.
The z wave license and certification requirements also probably don’t help with adoption from the likes of Amazon and Google.
Zigbee is everywhere of Smart home appliance