Podcasts > Ep. 194 - Convert WiFi Signals into a Sensor Network
Ep. 194
Convert WiFi Signals into a Sensor Network
Tony Nicolaidis, Chief Commercial Officer, Origin AI
Monday, November 27, 2023

In this episode, we had Tony Nicolaidis, who served as the Chief Commercial Officer of Origin AI. Origin AI utilized Wi-Fi sensing along with AI analytics to empower businesses in monitoring presence, motion, and specific events like falls using existing Wi-Fi signals. 

During our discussion, we explored how Wi-Fi signal sensing eliminates the necessity for specialized sensors, particularly in use cases like security systems and fall detection. We also delved into the potential business models for telecommunication providers when Wi-Fi signals had evolved into a valuable data source.

Key Discussion Points:

  • What enables Wi-Fi sensing technology to detect subtle actions, and what's the detection setup?
  • In what ways does Origin AI's approach support end-user privacy and bandwidth
  • How is Origin AI prepared for ongoing algorithm and data processing advancements?

You can find him on:

Website: https://www.originwirelessai.com/

Linkedin: https://www.linkedin.com/in/tony-nicolaidis-9b5b236/


Erik: Tony, welcome on the podcast today.

Tony: Thank you very much, Erik. It's great to be here.

Erik: Yeah, I know. I'm really looking for this conversation. When I first saw your colleague reached out to me, I thought, okay, Wi-Fi. Is this recovering IoT? Is this something interesting? I don't really want to have a podcast on how to hook Wi-Fi up in a house. Then I looked into it, and I thought this is super interesting. So I'm really looking forward to understanding what you're doing, what it's used for, and also how you do it to some extent.

But before we get into that, Tony, share a little bit about your background. I think you've got also a quite an interesting background, something like 20, 25 odd years. Black and Decker, actually. So really working on the industry before you got into this tech company.

Tony: Erik, thank you very much. Again, great to be here. Yeah, I've been around a little bit through the hardware industry, but now I've gotten into SaaS. I spent a long time over 20 years at Black & Decker, a great company, and did a lot of hardware development there for the first 20 — well, it was actually 26 years. The first 20 years, I did a lot of hardware development in product. Then in the last six years there, we started to develop SaaS types of products. We did a lot of initiatives in terms of getting into the technical aspect of construction. That's where I began to really love higher tech value propositions and product solutions. Then after leaving Stanley Black & Decker, I was the chief revenue officer in another small company that called on contractors again, especially residential contractors.

Then now I've got the great opportunity and the honor of being the Chief Commercial Officer at Origin, where my responsibilities are basically anything related to customers, revenue, partnerships, and that kind of thing.

Erik: Got you. What was it that drew you to Origin? You've had a very successful corporate career. I guess there's a thousand problems that you could be throwing your energy at. What was it about this company, this problem, that attracted you?

Tony: Origin was, it was an interesting thing. In the end, a good friend of mine that I worked with for a long time at Stanley Black & Decker, we were talking on the phone. He said, "Hey, I'm a part of the startup." He and I had actually, when we were at Black & Decker, had looked at the startup from a technological partnership standpoint. But he ended up going from Stanley Black & Decker to Origin. Long story short, you fast forward a few years. He said, "Hey, you want to come over here and take a look at this." He's the CEO now, Spencer Maid. I went over. It was 10 minutes of talking to him, seeing the technology in the demo house. I said, wow. I've got to be a part of this. Spencer called me. In the end, we ended up partnering up together again. He's the CEO, and I support him as the Chief Commercial Officer.

Erik: Cool. Okay. Well, let's get into it. Most people, when they hear Wi-Fi, they're going to think, okay, I can connect my device. Then some people might be thinking, okay, maybe I can triangulate a device, for example, that's getting a signal from multiple routers.

Tony: Right.

Erik: But you guys are doing something else. Can you share what is the problem here? What's the landscape?

Tony: Yes, Origin does something very interesting. It was a technology that was 10 years in development before we started actually commercializing it. It was a technology that was developed by Dr. Ray Liu, a University of Maryland professor. Dr. Ray Liu is now our Chief Technology Officer, Chairman of our Board. He is still very, very heavily involved with Origin. Dr. Ray Liu is a world-renowned Digital Signal Processing Engineer, the outgoing President of IEEE. If you know anything about digital signal processing, Dr. Ray Liu will be most definitely in the conversation.

Dr. Ray Liu figured out that Wi-Fi and Wi-Fi disruption, the Wi-Fi waves disruption that happened in a house, for example, we're building. The Wi-Fi is out there. It's like a water wave. As you move, and I move, and my hands move, and I breathe and walk around, those Wi-Fi waves are disrupted.

Imagine Origin being a technology on top of your ubiquitous Wi-Fi that you have in your house. We look at those disruptions in the Wi-Fi signal. Not just look at them, but we can give you extreme detailed context about that disruption in terms of you're walking, or a person has sleeping, or a person has fallen, or that kind of thing. We know if a person should be there, shouldn't be there. And so it's really a technology that's built on top of your regular Wi-Fi that gives you tremendous and clear and definitive context about what's going on in that house or in that building.

Erik: When I saw that, I thought that's incredible. Because I can imagine an empty room and the Wi-Fi can say, okay, this room is stable. Nothing's going on. It's empty. Then I can imagine somebody's in the room — a cat, a person, whatever — and it's moving. You can say, okay, there's a disruption. But that's like an A/B. Is there a disruption or not? That's kind of the presence topic. But then you're talking also about, can you tell if somebody falls — the specific action of a mass moving down suddenly or even if somebody's breathing? I mean, that is such a subtle action. Help us understand a little bit. What does this look like? Is it one Wi-Fi router? Are they set up around the room? What does the actual setup look like in order to sense it?

Tony: Excellent question. That's what we call the topology. There are a variety of different topologies that we can execute depending on the use cases. We have multiple customers that are out there. We're downloaded on a router. Let's say that we've partnered with a customer. For example, one of our biggest customers and our first customer really was Verizon here in the U.S. We're deployed on Verizon routers across the U.S.

Now, when that router has our technology embedded in it, basically, that router pings all the other IoT devices that are on the Wi-Fi network. Imagine you can have Alexa, Google Home, Sonos, your TV, your smart appliances. The average U.S. home has 17 smart devices in it. So we already have a topology that exists naturally in the house that consumers already own. That router now creates a zone from itself to each of those IoT devices. Remember, Wi-Fi can go through walls. It can go across big structure as long as there's Wi-Fi there. It creates these zones between the router and those devices. So there's many zones created to give you very clear coverage in the home or building. Then at that point, if somebody is within that zone, that router is talking to that IoT device back and forth and back and forth and seeing the disruptions through our tech off that Wi-Fi signal between the router and that IoT device. Then from there, our AI goes to work and tells you what's going on.

Erik: Got you. Okay. So it's not just your equipment, but it's your equipment communicating to all the other connected devices. Then I can imagine if it's like a presence use case, then a lot of different topographies could work. You just know there's something being disrupted. If it's maybe a breathing use case, then you might need a more precise topography, right? Because there's this subtle action that's happening in one part of the room. To what extent do you deploy where you say, "We're just going to deploy the situation. We don't know what the topography looks like in this particular person's house, but we know it's going to work"? Then what are the use cases where you say, "We're going to have to make some recommendations about them setting up a specific topography to make it work for a particular use case"?

Tony: The most prevalent topology is the router and all the IoT devices that are in there. That will be able to do what we call 'macro motion' — walking, jumping, running, moving our hands about — all the way to what we call 'micro motion' and breathing. That topography, again, as long as there's Wi-Fi and that router is connecting and pinging all these IoT devices, when you walk into a room and, let's say, you sit down and you're watching TV, and all of a sudden, you're falling asleep, we won't lose you. We know you're there. We know there's a person there. That's the most prevalent topology. That's a topology that exists. That's the easiest topology. Because, basically, as long as you have a router from a Verizon, you're there. All the way to more complex topologies like we're doing in the future in our health tech space, where, let's say falls, for example, where you will need other sensors besides the IoT devices that will allow you to make sure that we catch the fall. Even other topologies such as sleep insights, that we're going to be working on that are going to be coming in the future, you will need sensors on either side of the bed to be able to monitor your sleep.

So it depends on the topology. It depends on the use case. But the prevalent topologies that we're going to market with right now and scaling are very simple topologies that can be executed that are not going to weigh heavily on the customer in the house. We're going to get to these more complicated topologies over time, but we're scaling on very simple topologies as we're getting the business off the ground.

Erik: Got you. Well, let's walk through a few of those and give people a really concrete understanding of what you're doing today. What are the 2, 3, or 4 things that you're focused on today?

Tony: Today we're focused on three areas, three verticals. We're focused on what we call security. It's like the home security and being able to make sure that we know if a person should be there or shouldn't be there. In the security industry today, the biggest issue that security companies have — ADT, SimpliSafe, Vivint, all these guys, and even in Europe — is false alarms. When a consumer buys a security product and their phone is blowing up because of false alarms, they tend to turn everything off. They tend to churn the customers a lot.

So how are we going to solve this false alarm problems? We're going to solve it in two ways. Again, this is a topology that you can have a router with just the IoT devices in the house. A very simple topology where we'll be able to know that a person is in there. Okay. Let's say that you leave, and the security system is armed. We'll know. Partnering up with a Verizon, Verizon knows the phones and the devices that should be in the house. They know them. So they know if somebody walks in and it is a phone that's not known, and we know there's nobody in in the house. We know that the security system has been armed. We know that there's an intruder. Definitively, wWe know that it's an intruder, right? So that will absolutely trigger an alarm.

Another big issue is pets. Pets walk around and set off alarms. Well, can you distinguish human and non-human? So we'll know it's a human or it's a non-human, whether it's a pet, or an oscillating fan, or Roomba vac, or whatever. All those false alarms that are triggered, we will be able to solve those.

Then another big issue with false alarms is disarming. You get in the house. You have your groceries in your hand. You have all these things in your hand. You're trying to disarm the alarm. If you don't get there in 30 seconds, the alarm goes off. What if you're carrying your phone? The system knows. Hey, it's Erik that walked in. I see his phone. I know it's Erik. It'll auto disarm. Then when you leave, it'll auto arm. So you have auto arm, auto disarm. You have pet filtering or non-human filtering. That will significantly eliminate the biggest issue that's happening in the security industry today globally in a massive industry. That will make our technology very, very disruptive. That's security, number one.

Number two, we have what we call smart spaces or IoT. For example, energy management. For the longest time, they have used infrared sensors all over your office. They will use those to make sure you're in your office or not in your office. As you remember, sometimes you're sitting in your office and your lights turn off, and then you got to wave your arms to get your lights back on. You don't have to do that anymore. Our technology will know that there is a person definitively in there, and that the light should be on, and that the temperature should be set to a certain point. Same in your house. Today you got to go through and set up the schedules by day, by night, by weekday, by weekend. No more. All it's going to do is say: I see Erik. He wants it at 70 degrees. It's going to stay at 70 degrees. So that's the other thing we call smart spaces. That's the use case.

Then last but not the least, internet service providers like the Verizons of the world. They want to increase their value, broaden their value to their customers. They're using our technology to offer other services using our technology. That's a broad scope of what we do. Then we're going to be getting into health tech, like we said, fall detection and other things, as we move into the future, into 2024 and 2025.

Erik: Got you. Yeah, the health one is a bit fishy here also because of the regulatory issues and so forth, right?

Tony: We've already started. We can do it. But as we're scaling, we're going from startup to scale up. We're going to get after these first initial verticals, really get in there, and health will come for sure.

Erik: For the security one, I can imagine that also working in a lot of b2b environments but with somewhat different environments. So it's not one family that has a set of mobile phones. Maybe it's a museum. It's a warehouse. Are you looking at those use cases, or you think right now it's better suited to the home where you have more of a controlled environment?

Tony: Security, we want to start in the residential, in the home. Like I said, it's a massive, massive market. We're going to be launching with the largest European alarm company, number two in the world, called Verisure. We're going to be launching with them here coming up this year. We're going to be launching with another massive, great partner, Alarm.com. So we're going to be really becoming prevalent in the security space. There's a significant opportunity there that I think partnering up with the security companies in the space that we can make, that we can take that.

Erik: Yeah, that makes sense. Focus also is important. Talk to me a bit about how you work then. So you mentioned Verizon. You mentioned a couple of partners here. It sounds like your path to market is not by saying, "Here's our product. Buy it," to the consumer. But you're really collaborating with companies that have an existing solution in the market. What does that configuration tend to look like?

Tony: By configuration, you mean the business model?

Erik: The business. Exactly. So how do you—

Tony: We are primarily b2c. We will partner with these companies that are in these verticals. How we typically go to market is, they buy our technology. They embed or put our technology into their router or into whatever they have in the home. Then they go sell that to their marketplace and to their customers. We partner with them and sell along with them if they need, obviously. From there, we'll develop a business model or a rev share model with those companies as we go along and get through the development process. We partner with them. We develop with them. They launch, and then we share in the revenue. That happens afterwards.

Erik: Got you. Yeah, because it sounds just on the use cases you are explaining, it sounds like some of them like the security one might be really well-defined from day one. Like, we know what we're doing. We know the value, and we already know what we're charging people. We're charging them $25 a month. You can say, okay, we're going to take $2 a month or whatever that might be off of the top.

Some of them like the one with ISPs sound like there's maybe a lot of innovation that's going to happen. We know this is interesting. We don't know exactly how we're going to use it, but we'll roll it out. Then we'll start seeing what use cases we can enable around this. So it makes sense. I guess you're going to learn as you go, and then establish business models. That makes sense.

Tony: You got it. Some ISPs already have a very clear idea about what they want to do. Some ISPs want to get into security. So we just apply the security use case to the ISP, and they're off and running. We go.

Erik, we want to make sure when we give a customer playbook for every customer, what they're looking for, we have a very deep discovery process for every customer, as you would guess. We try to really understand what their use cases are. With this technology, it's new to them. Sometimes we've got to guide them and give them options. We usually land in a use case they can go sell their customers. They can monetize. Then obviously, we monetize too.

Erik: Got it. We didn't get too much, or I think we're not going to go too deep into the tech stack. But it sounds like the analysis here is pretty heavy lifting, right? Because you're doing some kind of a probabilistic analysis of waves and Wi-Fi from 17 different sources and then determining, did somebody fall and so forth? So it sounds like a pretty heavy analysis. Is that happening on the Edge, or is it being sent to the Cloud and then sent back with some kind of conclusion? What does that architecture usually look like?

Tony: We are the only company in our space that does everything on the Edge. We've been thoroughly tested by that. What do I mean by that? On the Edge, obviously, there's privacy elements there that are very good and privacy advantages there. But also, it does not affect the Wi-Fi bandwidth in the house or the building.

What do I mean by that? Verizon here in the U.S., we've also won the Deutsche Telekom business in Europe — two massive telecoms on either side of the pond. They thoroughly tested us to make sure that Origin does not affect my bandwidth so that I can give my customer the maximum bandwidth they needed to live their lives. And so being completely on the Edge allows us to have that functionality. It's been proven that we don't affect any of the bandwidth. Then like I said, privacy and everything else comes along with being on the Edge. We do everything completely 100% on the Edge.

Erik: Okay. Yeah, that's an interesting point. I hadn't thought about that. But I guess if you had this solution and it just had to send all the data you are collecting to the cloud to process and send it back, then somebody would be a bit frustrated with their streaming, right?

Tony: Yeah, we're taking up bandwidth to get it to the Cloud back and forth however many times that happens. So it makes sense. Exactly.

Erik: Okay. Great. Then communication to the customer, this is a bit of a new way of doing things. How do you tend to communicate? Do you just take a black box approach and say, "Here's the new solution. It works"? Or do you try to really actively communicate to the end customer, the consumer, about here's this new technology? Do you put an Origin AI, 'powered by Origin AI' in the box, or do you try to keep it under the hood and just say, "Hey, this is the magic that makes this thing work?"

Tony: Yes, so that's something we're always evolving with. Thus far, when you have the muscle of a Verizon behind you and they get behind the product, you don't have to do very much. They're very well-known. It would be the same with Deutsche Telekom in Europe. It would be the same with any security company that we partner with, Verisure in Europe. You don't need to do anything there.

In terms of the 'powered by Origin' or 'Origin Inside' kind of thing, we certainly would love to get to a place like that. But as long as we're on their website, as long as we have a joint formal announcement, we're partnered together, that we can go leverage within our content, we're fine with that. We just let them market the way they market to their customers. That's it.

Erik: Well, it sounds like it's working. You have a fantastic set of very large partners. That's a luxury to have, right? I guess another path that you could take, and maybe you would be taking if these partners hadn't come on board, is to say work with startups who are trying to bring a new value proposition to market, a new product to market. Is that something that you would consider, or are you really focused on saying we need to capture the top 10 players in the market and just focus our energies on these larger partners?

Tony: It's all about focus. You get those top 10 players in each vertical, and you're going to be doing really well. Again, resources in a startup are precious. So you don't want to go after the long tail. Because in the end, you're going to devote — working with a customer in the long tail is just as hard as working with the top 10 customer. It's obvious we're going to work with the top 10 customers in each of those verticals, and that'll get us where we need to be in. The tail will come over time.

Erik: Yeah, it makes sense. Maybe sometimes even more challenging, right? Because they're also figuring out their businesses as you go.

Tony: Exactly. Some of them are smaller, and they take longer. Exactly. Maybe it's even a bigger lift for the long tail versus the top 10. Absolutely.

Erik: In terms of the benefits, let's say, your partner, the b in the middle of the b2b to c are receiving. Do you see this primarily as we have a new value proposition to bring to the market, or are there operational benefits in terms of some kind of cost savings? Are they restructuring how they do delivery?

I guess, I could imagine, for example, an ISP that's now providing security can basically do a security system with no additional hardware. It's like you already have all the hardware in the place. So you're eliminating the whole bomb cost of any hardware sensors that a traditional security provider might need to have installed in the house. That could be a restructuring and a big cost savings there. What's your sense of how they're making a business case around the use of Origin AI?

Tony: You just nailed it. They will position themselves versus these other security companies and do exactly what you just said, kind of dematerialize the whole thing. They make it as simple as possible. They positioned it in a way that makes it very easy and extremely advantageous for their customers to take it. They have that in their messaging that they put in the marketplace.

Erik: Got you. Okay.

Tony: Like for example, in security, false alarms, in reduction of false alarms. In the security industry, they can absolutely market that because they know that's an issue with their customers. They know it's a cause of churn. They can say, you know what? They can market less false alarms. And all of a sudden, consumers will realize that. Then it's not just to their customer benefit. It also reduces their churn. So it's an operational benefit. It goes both ways, really. I mean, they maximize financially from the whole thing.

Erik: Yeah, it makes sense. There's a value proposition side on the architecture of the hardware. That impacts the cost structure. But then also, there's a new value being provided to the customer that really wasn't possible. It wasn't possible without a lot of extra headache and hardware previously.

Tony: Right.

Erik: Let's talk a little bit about what you see in the future. So this is, I'd say, still a relatively young technology. You're now scaling up some of your larger partners. If you look out 12 months, 24 months, what is exciting for you right now in the business?

Tony: Well, when you look out 12 months and you look at our verticals, it's really partnering with these top companies in those verticals, seeing those top companies and their customers, see the advantage of what Wi-Fi sensing can do and what Origin can do for them, and then start penetrating other verticals deeper like health.

The thing we get excited about is positively impacting millions of people in their homes in a positive way, and making their lives easier than it is today. Punching in a security code or having false alarms. You're in the office, and your phone is blowing up because there's somebody in your house, when it's really your cat moving from one room to another. And health. Being able to take care of our elderly parents. They can still, with dignity, live at home. I think we can significantly positively impact the lives of millions of people around the world. At Origins, I think we're going to change the world. I know it's very dramatic. But that's what we say. We're going to change the world. We're going to make the world a better place. That's really what — we really believe it, all of us across the whole group.

Erik: We need solutions like this. I'm sitting here in Shanghai, in China. You probably read the reports. China has got a big aging population.

Tony: Oh, yeah.

Erik: They don't have the luxury. In the U.S., I think we're spending something like $10,000 per person per year on health care. And so they don't have the luxury to do that in China. That would be pretty much the entire GDP of the country. And so, what do you do? Then you need solutions that allow people to live at home safely, and you need to do that cost effectively, right? And so, certainly the world needs new solutions.

If we look at the technology and how it's developing, is it a technology where you say it's basically you've gone through this 10-year development cycle, it's basically ready to go today, it's mature today, and now the goal is to figure out the partnership structure and figure out the use cases? Or, do you think that every 12 months, there's going to be maybe significant advances in the algorithm, for example, or in other ways that you process data, for example, that are going to open up unlocked doors? Is it more about getting the commercial structure in place, or is it also about making these continuous technology advances to unlock new use cases?

Tony: Yeah, I'll bring them to our focus again. You brought it up. I brought it up. We're focused on scaling now. We're ready to go. The technology is ready to go. As long as you have Wi-Fi in the house, in the building, we can work over that Wi-Fi and do what we do. We are ready to go and scale and work with our partners.

Now, we have a very stout product road, Erik. Over time, I think we're going to get better, and we're going to have new use cases that open up. Those will come. But right now, Erik, we're focused on really driving the technology — it is mature enough to drive. It has matured — and working with our partners across the globe.

Erik: Great. Well, Tony, it sounds like you're going to be busy. Chief Commercial Officer, right? The weight is on your shoulders then. Usually, the CTO is the one carrying a lot of the stress. But I think in this case, I'm sure the company is looking at you to pull it forward. It sounds like you have a great product, so I'm sure you'll be very successful.

Last question, Tony. What is the best way? If somebody's listening and they're interested in learning more, what's the best way for them to reach out to your team?

Tony: Go to our website originwirelessAI. Or, search Wi-Fi sensing, and you'll get to Origin AI. You'll get to our website. You can ask for a demo or ask anything, any question. Put any or all the information, we're going to get back with you very, very quickly.

Erik: Awesome. Tony, thanks a lot for joining us on the podcast today.

Tony: Erik, it was my pleasure. Appreciate you having me.

Contact us

Let's talk!
* Required
* Required
* Required
* Invalid email address
By submitting this form, you agree that Asia Growth Partners may contact you with insights and marketing messaging.
No thanks, I don't want to receive any marketing emails from Asia Growth Partners.

Thank you for your message!
We will contact you soon.