Zigbee / Matter Motion Switch | Project Linus (Blue Series)

Ugh - These are exactly what I am looking for … and they don’t exist :cry:

I’m moving into a new house within 2 months and I need to smart home it up (my current home is Z-Wave, but I am going to switch to Zigbee next place - and a path forward to Matter/CHIP is even better). Meaning buying something around 70 switches. I was going to get the 2-1 switches, but I don’t want to burn money on that since THESE switches are what I really want (with the presence / motion / lux sensor).

I hope these can be developed even without LTT’s support (and quicker than the 2-1 since they are mostly based on it and as such MOST of the R&D is done). I would even volunteer to beta test it.

Hi Everyone

I am a little late to the party and have a few thoughts based on my experiences with a few mmWave sensors this past year. I have a number of projects which you can follow in the HA Communities.
mmWave Presence Detection
Low-Latency DFRobot+PIR

This is a pretty exciting project and there is a lot to discuss, let’s get to it :slight_smile:

mmWave - Where I see us today

As of June 2022 I have tried four (4) different mmWave sensor modules with varying capabilities. They have taught me a few things which have been echo’d earlier in the thread. Please consider me a layman in this area; one with a little experience of the basics.

So what are mmWave basics worth considering?

  • the antenna array which will influence the detection area and impact physical design (size/placement)
  • the PCB/enclosure design which will influence detection direction (going backwards is possible)
  • the MCU, F/W, configuration parameters which will influence the detection trigger time (a lot more on this below)
  • the FTT analysis which will influence featureset/configuration-possibilities

Physical Design

I suspect that with the proposed placement behind the paddle there is sufficient room for most [recessed] antenna design. The biggest contraint is already mentioned, it likely means +/-170° at best!. This needs significant testing in order to understand the relationship between detection distance/sensitity and end-of-range/angle detection/sensitity. mmWave is just a radio wave after all and it drops in strength at the edges of its dispersion pattern. There is a tradeoff in antenna design/gain/distance/angle/etc. I will leave that to the PhD design experts with the understanding that you can’t have everything without paying for it.

What I could envision are two arrays protruding (30°?) from the switch (ala PIR lens) for greater coverage, I am not clear there is a requirement or if there is acceptance factor for this.

PCB Design

This seems like a requirement to dictate to the sensor module manufacturer in order to ensure no rear detection is possible. I have a SeeedStudio 24Ghz mmWave sensor module that does this and requires a metal box to ensure forward facing detection.

MCU, F/W, Cost

This will honestly be the biggest contributor to “mmWave performance.” mmWave in itself is just radar (doppler/FMCW/CW/etc) and the “interpretation” of those radioforms is key. The faster the MCU, the more cost and not necessarily the better performance as firmware influences this. Just like an application on your PC, the greater the workload the slower it runs. This leads to the next section…

FTT Analysis

This is the app the mmWave MCU runs. If I were to categorize these based on my experience from simple to complex;

  • Simple. The DFRobot or equivalent Leapmmw sensors that perform the most simplified movement-based analysis. Human micro-kinesis at best.
    – Pro: fastest response time ~25ms
    – Con: fans, AC’s, airflow wiggling a curtain are all potential triggers.
    – Risk: without a PIR this a challenge based on my limited experience.

  • Medium Complex: This is your SeeedStudio “Human/Heart/Fall” or Aqara FP1 type module that performs “interpretive analysis” on the mmWave FTT target scatterplot.
    – Pro: Can be programmed to recognize “human’ish movement patterns” which may inherently ignore object-based movements.
    – Con: 2022 versions are slow. >1 sec trigger time and as the Aqara demonstrates can get much worse.
    – Risk: you are at the whim of the algorithm for detection resilience.

  • Complex: This is your RFBeeam KLD-2 or KLD-7 type sensor that do all sorts of amazing things.
    – Con: first up, cost!
    – Pro: awesome’sauce :slight_smile:
    – Examples: The KLD-2 detects approaching/receding as well as direction in <800ms and can be configured below 400ms. The KLD-7 is not intended for this application.

  • The Unknown: "what about the [insert hundred other AliExpress modules here]
    – Risk: you need to test it. Nothing can be assumed without validating a comprehensive datasheet. No datasheet, don’t bother.

Requirements & My Vote

Design decisions are needed in order to determine a suitable module for this application. Quite frankly, there is no perfect answer. My thoughts are;

  • #1
    – <100ms response time. There is nothing worse that slow lighting and the number one reason to toss out a sensor.

  • #2
    – A fast sensor that cannot be tuned to a wide variety of installation situations will just as quickly be binned. What you lose in wall/switch placement flexibility you must compensate for with detection parameters. Sensitivity is not enough, without distance at minimum you will quickly accumulate those single star reviews. Direction/angle/speed all add to the cost/detection-performance

  • #3
    – Cannot penetrate walls. This is a dealbreaker. I didn’t believe this at first until I owned a SeeedStudio.
    – Wall contruction must be defined.

My Vote

  • My 2022 Vote
    A non-interpretive-based mmWave sensor module/MCU that is fast and permits at minimum distance and sensitivity adjustments. Combined with a [good] PIR it would eliminate the majority of object-based [non-heat generating] false positives. Having been quoted $15 (China-direct) per sensor module in single digit quantities, I suspect this is closer to $10-12 for the mmWave in quantity. This gives you $5-10 for a good PIR? If the $150 version has anything less than a Panasonic EKMx PaPIR I will be cry /s

  • My 2023 Vote
    An interpretive-based mmWave sensor module/MCU that ditches the PIR for waveform analysis to detect human-type motion in the target detection period [<100ms].

Design Decisions

This thread is long with a lot of great suggestions in a wide variety of areas. Is there a tabulated breakdown of the final direction?

  • Physical design
    – button placement
    – sensor placement
    – artistic look
  • Conceptual Design
    – what is my target response time?
    – what is my target detection area?
    – what are my configuration parameters? We have discussed a few, this should be investigated further; eg channel selection (interference avoidance)
  • Logical Design
    – spectrum (5.8/24/60) - impacts gain/distance/interference. My vote: avoid 5.8, 24 for distance, 60 might be immature this year?

Conclusion & Thoughts

In my journey of deploying no less than seven (7) mmWave sensors in 900sq feet I have to conclude; done right, this will be a game changer. A non-interpretive [DFRobot] sensor is great for rooms where object-based detection is of little concern. Yet I suspect this is a small percentage of deployments. For all the rest, if you don’t chose a human-interpretive or PIR+mmWave combination you will experience a significant slope in excitement-factor and then a mirrored decline. I have had a dual-mmWave+PIR combo in my kitchen for months and it is reliable and what you want to deploy without fast interpretive option. Given the added placement contraints of a wall switch, anything less will be extremely challenging to gain long-term acceptance.


PS. @Eric_Inovelli - thanks for inviting me here…

…if this switch does not include a spotlight that follows you across the room I will be extremely disappointed! :stuck_out_tongue: /s → https://imgur.com/a/F8CMBt8


@crlogic – thanks so much for taking the time out to write up about your experience. Holy crap. This is great stuff and why I’m so thankful for building this as a community. This is way above my pay-grade as a marketing guy.

That said, I’m hoping we can work together (everyone) with the knowledge in this thread to figure out the DIY version vs the Commercial version and come to a consensus. As I’m quickly learning, there are so many options and so many variables to consider, I’m actually very overwhelmed lol.

But… I agree with @crlogic – this will be a game-changer if implemented properly and if anyone can do this, it’s our community!

Let me digest a lot of this and try to synthesize it down. My goal is normally to take information from the really smart people and try to turn it into use-cases that the everyday person can understand. This is often a challenge for me bc it means I also have to understand it haha!


Crlogic’s experiences reinforce a lot of my theoretical concerns. The cheap sensors are going to be overly sensitive to non-human objects and/or require a LOT of processing to do human detection properly, which probably isn’t feasible in a switch which installed in a high-heat environment, has limited airflow, limited physical space, and has cost concerns.

This is probably why most commercial devices in switches have went the dual-sensor route (PIR + mmwave) as crlogic suggested.

@Eric_Inovelli Can you reach back out to your manufacturer and see what the feasibility/cost would be to do a PIR+mmwave sensor? You can use the original aesthetic design for PIR placement and still put the mmwave behind the toggle switch.

1 Like

There is also one avenue that I don’t believe I see covered in the first hundred posts. And that is, customized mmWave MCU firmware. Clearly this impacts cost and time-line. Perhaps it opens the possibility should an ODM have the ability to customize a trigger vs a “hold-high” parameter.

eg. trigger requires approaching direction or a threshold for left/right movement detected versus ‘hold-high’ at a micro-kinesis level.

I do this with two sensors ATM. KLD-2 gives me an approaching trigger (but no receding - because placement). Then a DFRobot gives me sustained hold-high. Call it a “customized logic” option akin to a PIR+mmWave but with a single sensor (if possible).

I have seen a number of sensor ODM’s list custom options available (at unknown price/timeframe) on their websites.

Consider this a tangent but worth the ask.

1 Like

All I want is an option that if motion sensors turn on my lights, to be able to click the switch on and have them stay on manually, until I decide to turn them off. I’ve done workarounds in the past, and I know I can whip some automations up in node red, but for this to be a feature would be amazing.

PS. I’ve never willingly subscribed to a newsletter before, but I’ve subscribed to Inovelli’s! I love the look of the Blue series and I hope to get some soon.


I like where that is headed, but I can see where it might still be problematic. Let’s say I walk into a room and turn a ceiling fan on and then turn around and leave… does the motion sensor still report active motion?

I would expect so. I would expect the same to be true for a mmWave+PIR as well.

This is a good use-case example in order compare different sensor package options.

Eg. a mmWave sensor that has a wide angle but shallow height is a good solution to avoid detecting ceiling fans or low-objects (cats).

[edit] - this also highlights one design crux; until you explicitly define a requirement, people often have their own implicit impressions. In this example, my assumption is that a mmWave+PIR never allows the light to turn off by PIR-alone. If we did that, it is no better than PIR without mmWave.

Can we also implement PIR, temperature and humidity sensors to the switch please :slight_smile:

1 Like

Temp is probably not viable for the same reasons the ecobee switch failed. The switch internals give off enough heat in various load applications that the temp reading would be wildly innacurate (think 90F in a 70F room).

Humidity seems like a really good idea since that could drive bathroom fans (if certified for fans).

1 Like

Possibly, right now we’re getting kind of on the high end for switch costs, so we’ll have to see. I just realized the quote we received was for a 5.8Ghz module, so we’ll probably have a higher cost already if we switch over to 24/60. I’ve heard that 60Ghz modules cost about as much as the switch itself :exploding_head:

This would indicate 5.17cm or ~2" detection accuracy. That seems like enough for people moving around? Maybe not someone typing at a desk? Interesting…

On one side accuracy is good, on the other side seeing a curtain move slighly due to HVAC running might be too accurate…

Every time I’ve seen 5.8Ghz in this thread I’ve just assumed it was 58Ghz. If it’s really 5.8Ghz, will cordless phones cause an interference issue?

Do you actually have a cordless phone? I haven’t had one in over a decade. I’d be more worried about wifi interference.

I don’t even have a landline :joy:
I don’t think the 5ghz channels are wide enough to interfere with 5.8ghz


Right. Unless the mmWave MCU is maxed out, there is little reason why the MCU could not (in theory) support multiple outputs. For example, a mildly complex interpretive, and a minimally interpreted high speed output. OR various other combinations. It might even be possible to have a fast lightly interpreted output that detects only significant motion in the range where a PIR would detect motion. The real question is costs. What would a sensor with customized MCU code actually cost? And that could vary significantly by the pricing model of the mmWave ODM.

Some might be willing to do a larger upfront NRE, with smaller per unit additional costs, while others may insist upon pricing such things entirely by way of increasing per unit costs.

But on the other hand, trying to customize anything here will inherently increase costs. We are already talking about a fairly expensive switch here. Further trying to do something custom here is made more complicated by the fact that Inovelli outsources the low level design (which is a mixed bag, having in in house design team allows for answering questions and exploring possible design space faster, and getting things manufactured can be cheaper when you can hand over a completed design. But outsourcing can take advantage of a team that likely has more experience in areas that are new to you, and avoids the expenses of having to maintain a full design team in house.)

Kickstarter may be nice to raise brand awareness, whatever makes this happen :smile:. I’d love a 10-pack, happy to beta test when the time comes.


He’s using Zwave, not Zigbee, he doesn’t want the additional interference to his WiFi that Zigbee introduces.