← All talks

LOLGrid: Don’t bring your own network, one already exists - Brent Shaw

BSides Cape Town38:58506 viewsPublished 2025-03Watch on YouTube ↗
About this talk
A fun look at network over powerline systems. The use and abuse of these solutions, what kind of attacks are possible, what are the limits, and how can you defend yourself? Sometimes things just need to be looked at differently. Ethernet over power lines (IEEE 1901, Powerline, PowerPlug, you might or might not know it under different names) has been around for quite some time. That being said, while maybe might have heard of it, it seems to be a technology that never took off. The technology allows for Ethernet traffic to be routed over alternating current (AC) power lines. This can help avoid clumsy cabling in small flats, or remove the need for long cable runs, providing a neat alternative to running new Ethernet cables. It can also be viewed as an interesting attack vector. While work has been done to stop different power line devices intercepting other devices traffic, allowing multiple networks to be hosted together, this leads to new problems when it comes to securing your own infrastructure. The talk will do into some background on the tech (very briefly) to explain its intended purpose and use cases. We will then look at how it can be used as part of attacks/engagements. We will then look at how one can defend against this. We will looks at some interesting things I discovered along the way. ========================================================= About the Speaker - Brent Shaw I am a Cyber Security Researcher working for Nedbank. I'm a big fan of learning and pulling things apart. I have previously given talks on at BSides: "Securing the Industrial Internet of Things", "Hearing the Internet Background Radiation" and "Hashing the $#!+ out of firmware". ========================================================= Thanks to our AV Sponsor Tenable for making these recordings possible.
Show transcript [en]

Cool. So, welcome to my talk. Um, hopefully uh this will be of interest, maybe some entertainment. I'm actually not sure what you're going to take away from it. Um, but that's why we're here. Maybe learn something new. So, let's get started. So, a little bit about me. Um, I'm cyber security researcher for NetBank. Um, uh, did a masters in computer science. Um, I like pulling things apart. I like learning. Um, sometimes I dive into a project that I'm actually really not equipped to do. Um, this is a little bit of that and you'll see a little bit what I'm talking about, but that's kind of why I want to talk about it because no one else is going to try

something. A lot of people don't like to try things that don't work out. So, what what to expect? Okay, it's me talking for like a long time. Um, that's probably not what's going to happen. It's going to look a little more like this. Um, basically I just want to go through a little bit of background on what power line communication is. Uh, go through some uses, real world use cases. Maybe you've played with this before. We'll find out. And then look at really what I'm focused on, which is is it a good attack vector? because that's really where I saw this going was it's a very obscure weird thing and I thought maybe you can use it as an attack

vector. So power line communication or PLC um has anyone heard of this before? Okay, it's the other PLC. It is not the PLC's I spoke about last time. Um it just so happens that for some reason they like them to go with acronyms that are reused and used elsewhere. very annoying. In this particular case, when we talk about power line communication, we're talking about networking using electrical wiring as the physical medium. So, instead of running Ethernet cables all over, use your house wiring. Okay. Interesting idea. Has anyone actually seen these adapters before? Yeah, few people. Has anyone played with them before? Has anyone had success with them? Way fewer hands. Okay. And that is what my experience was going into this.

I had never played with them. Okay, they look like this. They're pretty terrible. Um, I don't actually know who used them. They were like they came out. I remember seeing them around and the idea was you basically plug it into your power outlet somewhere, plug an Ethernet cable in, plug another one in somewhere, who knows where. Um, and take the Ethernet out. Yeah. and the only people I ever knew that bought them swore they don't work. So, the correct way to do this is to dive into it. Now, it's very complicated. It basically uses carrier frequencies to transmit data over your home wiring. I put a lot of slides in on this and then decided no one wanted to

really hear about that. Um, if you do want to hear about it and the really nitty-gritty stuff, we'll probably come find me afterwards. we'll we'll have a chat because there's some really fun things that you can actually do with this and there's some fun things people are doing with it. Um but normally very expensive things that I don't get to play with. Um so what what is power line communication really aimed at? Now when you try go look this up if you go on to wiki try actually find any kind of information it's kind of weird. It's used for home networking. They say it's used for IoT. Okay. Um, it's uh used for broadband over power lines. Now, this I have seen.

I've seen people try and implement this. I've never seen it truly work. Okay. And basically trying to deliver internet to your house over your existing power line. It's a pretty cool idea. Um, but I'm sure we can see why that might not work. Um, then everyone's favorite, smart grid or smart grid city because that can only be cool. Um, so I went hunting and there's a bunch of working groups and advisory groups that have started to put together use cases for these things. When I say started in like 2000 um when they really thought that a lot of things would be connected to the internet. Not sure why Ness thought the toilet should be online, but this is where I guess IoT

comes in. Yeah. Now this stuff actually dates back really far. Okay. And when I was there really far, back in like 1975, X10 came out. Okay. X10 was a power line communication protocol aimed at bringing automation to appliances, homes, stuff like that. And in a way, that's a great idea. You know, pre the advent of Wi-Fi, if your microwave and fridge could talk to each other over your power line, that would be cool. I don't know what they'd say, but it was a cool idea back then and it kind of sat around for a really long time. Home Plug Power Line Alliance was only formed in 2000. Yeah. And this is what most people think of when they

think of power line communication. They think of these funny looking little devices with an Ethernet port underneath. And they don't they don't work too well. Um, anyone that's played with them generally says that they're terrible. Yeah. Turns out there's actually quite a few iterations of this stuff. Um, starting in 2000, they started throwing things together and they've actually thrown together a pretty crazy spec. Um, has anyone here ever read like industrial control system specs? Okay. Like, and they're they're normally quite mad. They're like a few hundred pages long. If you've ever like gone and read like an RFC, got, you know, the original specs on stuff, they throw things together. This is ridiculous. It's like 800 pages. They've

catered for stuff no one ever asked for. If you really want to spend a weekend and never get that time back, go read it. It's It's interesting. Yeah. Basically, they've reimplemented Wi-Fi on power lights. Yeah. It's a lot of the same stuff. They've just thought about it differently and in a very backwards kind of way. Why did they do this? I don't know. Um, it seems like they really thought that AV was going to be huge. Connecting your, I don't know, like Blu-ray player and Xbox to your TV over power lines might have been a good idea. Um, the truth is Wi-Fi kind of killed these things. Yeah. And that's realistically where we're going to this.

Um along the way, the reason I got interested I was digging around at liquidation store as one does and um I found bins of these things because no one wants them. Yeah. And they're generally pretty cheap and terrible. Um so I picked up quite a few of them. Yeah. This gets to why should you care? Yeah. And realistically, I think you should care because this is dead technology and that's the technology you should care about the most because no one else cares. Yeah, no one's looking at this stuff. I tried to look. It seems very weird, but like everyone seems to have given up, which they should have. Wi-Fi is better. Wi-Fi is faster. It's a lot

simpler. In a lot of cases, this stuff doesn't work as people has have kind of found out. But that doesn't make it not cool. Yeah. There's a lot of work that's been done on this technology to secure it. Yeah. Stuff's got AES encryption running on your like power, you know, in your house. That's pretty cool. Um, but no one's really looked at how to secure against it. Ever heard of an evil twin multiplug attack? No. No. No, no, you haven't. Um, cuz I don't think anyone would ever try. But it's but we should think about it. That's really what I'm kind of trying to get through with this talk. Um, with Ethernet, we purposefully lay down

copper cabling. We run it all over the show to do exactly what we wanted it to do. With Wi-Fi, we put down access points. They broadcast through the air. It's great. Um, with PLC, with powerline coms, we kind of just use what was already there. That's the problem. This network exists. It's in all your homes. It's in all your businesses. It's everywhere. The network is there. No one is setting it up. No one is setting up access points. No one is setting up anything. If they want to attack you, in a way, they can. They've got infrastructure already. Yeah, that's kind of the premise of the talk. Um we're going to go through some possible scenarios. When I say possible

scenarios, I mean very remotely accessible scenarios. Um I mean yeah, network implants, remote access, harvesting data in Xville. Yeah, these things are possible. Are they likely? Probably not. But I kind of don't care. Um so in this first instance, I'm going to look at a few different things. Whether it's two businesses next to each other, you trying to steal the free internet from your res's common room, or possibly trying to gain access to a secure area from an adjacent public access area. Yeah. In a lot of these situations, you could run an Ethernet cable. You could simply set up a Wi-Fi hotspot in the secure area and sit in the public access area and access

it. You could use a torch. Like really there's a lot of different ways you could get data from that secure area to the public area. But PLC when you actually start to use look at how power lines factor into this it's a little bit different. Those areas are connected if they have power running to them which they probably do because it's everywhere uh when Eskim puts it there. Um but that's the great thing is it actually doesn't matter. Eskim doesn't need to put it there. the lines are there. As long as your devices have electricity, you can utilize this network. So, this is the sort of um thing we're going to look at possibly how to harvest data with this. Yeah, in

some cases, uh maybe it's not a very realistic attack, but maybe someone sets up a remote camera. You got to get the data from that camera to where you want it to be. Yeah, maybe that's what Wi-Fi run an Ethernet cable, go in and occasionally fetch that information. Yeah, getting the data is one step. Getting the data out. Very different. Yeah. And that's always the problem with expo. Maybe you've gone and stolen a whole bunch of data. It's, you know, probably going to be noticed when you try and send it somewhere. Yeah. But who's looking? Yeah. If you've got a nice network already installed, maybe not. Yeah, this is the generic scenario I'm going to talk about just because

it's a very kind of conceptually simple thing to think about. Um, and yeah, I get that maybe this isn't super realistic. Um, but there's actually quite a lot of cases where this is something you might want to do. Okay, whether it's a coffee shop next to a high security company sharing a wall. Yeah, that could be something you're interested in. So, how much setup do we need? Very little. Yeah, you would need to plug one of these in in your high security area and you'd need to plug one of these in in your public access area. Yeah, that's a ton of setup. Um, and it's that easy. plug them in and start using them. They

do have different mo modes. Broadcast paired. We'll get a little into that. It's uh the spec is ridiculous. Um I don't know why. Um no one I don't think was ever running these things in broadcast mode. Um it just makes a mess. It's noisy. Uh my experience is doesn't really work properly. Yeah. But how do we get these things working? Plug one in. Plug another in. You basically click the little pair buttons and these things now identify as what they call a logical network. Yeah. Now the cool thing is you can stick a ton of these things with different logical networks within the same room within the same house, same building. You can run a whole bunch of

them. Yeah. They actually will all talk on the same line. Much like Wi-Fi accesses the same air, you know, uses the same space. Uh they kind of do the same thing. And that's kind of cool. Um, TPLink has very nicely provided the most worthless web interface on the face of the earth. Um, there's like a really long number underneath these things and you can manually type it in to pair them or you can click the pair button. Um, so good job TPLink. Um, very luckily um, because this is like a group of companies that put this all together. Um, they're all the same. Like the colors change a little bit, but I mean, Mercury and TPLink,

they've Wow. Um, very, very simple to use. Unfortunately, that's where the kind of fun and games stop. These things as far as setup and playing with them, they're kind of dumb. Yeah. And Yep. We're we're going to talk a little bit about testing because along the way and as a few people have mentioned by showing their hands. Yeah. Different people have experienced very different things when they've tried to play with this. Yeah. It's a product. You buy it, you use it. It's great. Very simple. You plug it in. Uh I just don't believe that. Um, and I didn't believe this until I bought it and plugged it in. And, um, it really works. Uh, I was

quite surprised. It said it pretty much hit the speeds it said it would. I plugged one in in one room, went to another room, plugged one in, just worked. Um, and I was actually a little bit confused because I really thought it wouldn't. Um, and uh, I spoke to quite a lot of people around this and there are a lot of people that have very valid concerns of this. Um, they say, well, you want to use it as an attack vector. You know, it's not going to work. People have UPS's, it won't go past a UPS. People have inverters that'll get in the way. People have other devices that make noise. Uh, I'll be honest, I don't know.

Uh, so I thought, time for testing. The most basic way I figured I could test this. Just plug it into a multiplug. Plug two in next to each other. See what happens. What happens? It kind of does what it says it does. Get about 90 megabit a second throughput. No problem. It just works. I didn't expect anything less than that. So, I thought this is where the rigorous academic testing should start. And uh where this really comes down to is everyone said that these things really don't work very far. Yeah, everything online says they work 300 m. Um, now there's a trick to that. 300 m is not 300 m. Yeah, when your Wi-Fi says it can reach 300 m, they

normally mean line of sight. Yeah, this is not line of sight. It's something very different. 300 m is 300 m in a different way. Okay, so if we imagine I don't know how well that shows up on the screen. There we go. We've got a nice long cable, 6 m long. Yeah, we can make the same 6 m long cable a little shorter. Those two ends are only 2 m apart. Yeah, but we're now talking about a 6 m cable. Okay, that's our 300 meter problem. Yeah, in a normal room like this, a plug on that side of the room, plug on this side of the room, not very far apart. Okay, but the cable taking

the power back to the DB possibly those plugs could be on different circuits. They could be on different phases. There can be so many issues. Okay, very quickly, two plugs that are 2 m apart, and I've experienced this in a few different offices. I tried to plug these things in. Um, from one desk to another, you might get nothing because the cabling between those two desks, desks 3 meters apart, the cabling could be hundreds of meters long on different phases, on different segments, who knows? Could be very complicated. But I also had fantastic success going very long distances uh because the cable runs must have been pretty short. So, it's a bit of a tricky

one. Uh, we can do some very nice things. Um, but for the basic testing, I think it's winning. Uh, for the more academic testing, um, I didn't feel like building a building. Um, so I borrowed as many extension cables as I could. I got up to about, um, about in the end about 200 mters. I borrowed a a reel of unused house wire. Um, it's not in this picture for safety concerns. Please do not do this. Um, but this was quite annoying. Um, because it actually worked. It worked with almost no noticeable impact on the signal at all. And that was actually very very annoying. Um, I was really hoping like four net like, you know, four extension

rules would just foil this thing. Uh, apparently not. So, we needed to go to more more extreme lengths to try and stop it working. UPS inverter. Yeah, easy one. Plug in the UPS. It works. Very annoying. Inverter. Inverters are big. So, I found one in a box. Um, inverter in a box. Works. No impact in speed. Very annoying again. Yeah. Now I must note here when I say inverter uh it's running in kind of uh pass through line interactive standby mode. Um obviously if your one plug is on Eskim side and the other's on the inverter side and Eskim stops working it's not going to work. Okay. Unless you're selling back to Eskim, in which

case probably there are tons of these things and unfortunately I did not get to do the full rigorous testing on all of them. But they all work. And when I say they all work, they all work when plugged in at the same time. In fact, they all work when I chain one into the other and send the data back and forth through an extension cable, which is a dumb thing to do. Um, I will have that. Uh, if you actually have a look, we're down to about 20 megabits a second. Um, going back and forth through an extension cable, looping through multiple different home plugs. Yeah, I really thought that that might be the limit. It works. It's just

slower. Okay. And that's pretty robust actually in my opinion. I really thought these things would fall over long before that. What about transformers? Yeah. Well, turns out transformers are not the kind of panacea that people maybe thought they were. Yeah, I had high hopes. Um, so got myself a little isolating transformer to isolate me from Eskim. Did some testing and it works. I really thought this would be the one. Yeah. Now, I should note transformer plus a lot of cabling actually did eventually kill it. Yeah. Enough cabling it does stop working. But it took a lot more to stop this than I thought. Okay. So, have we learned anything? Maybe. Now, how do you defend against

this? Well, I can tell you naturally if you're talking about office buildings, they're probably just fine because the cable runs are so ridiculously long that these things probably won't work. Yeah. Put them out. So, you know, that's fun. Jamming does work. Yeah. I set up a bunch of these things and I basically set up a Python script to just bury a ton of network traffic into one of them. And yeah, you can basically just congest a line enough for this to really become a problem. Yeah, especially if you've got one device that's faster than the rest of them. Uh because I bought uh kind of leftover ones that I could get my hands on. Um I

only had one AV1000 device. All the others were AV600. If you let the AV 1000 do most of the work, uh you do start to see degradation on the other lines. Okay. But it took quite a lot. Yeah. And this is really where this starts to come to. Filtering does work. If I put this through um certain um power um power filters and power smoothers, you will eventually take out the carrier signal. You won't get this through. You can filter it out, but that's not the easiest way to do things, and it's certainly not very easy to do for a whole building because it's only going to filter from one side of the filter to

the other. Everything sitting on the same side of the filter will still talk to each other. Yeah, this is quite a problem because when you think of jamming, when you think of attacks on Wi-Fi, you always everything kind of coexists in the same medium. This is a bit more complicated, but a lot of fun. Yeah. What about detection? Um, well, the specs public. You can go look up how these things work. It is entirely possible to detect. It's a bit tricky. Um, I had a good go and I did manage to see a few different carrier frequencies. Um, but h I seem to be missing a picture. That's annoying. Um, it's a bit trickier. This is not how you want to

really go about detecting this. Okay, an oscilloscope trying to monitor your home power, not a great way to do things. Um, but there's a nicer way to do things and that is provided for us. They've built it already. In theory, we could just use these things to monitor themselves. Yeah. Now, unfortunately, unless they're in broadcast mode, uh, you can't use one to find others. Yeah. Once they've got an encrypted stream up and running, it's a bit tricky. You can do a few funny things to um mess with it. Um I started modifying some firmware in the hope that we could take one of these things and turn it into a monitoring device. Rather, I there's,

you know, it's going to be impossible for me to detect the traffic. Traffic's all encrypted. Everything on the power is encrypted. I'm never going to be able to see what's getting Xfilled, but maybe I can get an alert to say someone is doing something. Yeah. Can't, you know, you can't always detect what is going over a Wi-Fi network, but you can tell when someone spins one up. That's what we're after here. It would be really nice to eventually get some kind of detection method for if somebody starts to use your building wiring as a new Xville channel. Yeah. How did that go? Well, uh, not too well. Um I have managed to kill uh the only good one um

that I actually had. Um so uh firmware mod did not go as planned. Um so there's more work to come there. Uh but what else can we do? Well, it depends on the attack. Obviously if somebody is trying to gain access to your network, what do you do? Monitor your own network. Okay. No point in monitoring the power if they're trying to actually access your network. Much simpler. Monitor your own network. Now, once these things are up and running, chances are if they're not moving traffic and nothing is plugged in on the other side, your network's probably not going to see anything. Oh

dear. Yeah.

Nice. Cool. Well, while they're figuring things out, anyone got any questions so far?

So, the encryption makes it hard to obviously read the traffic. Um, they break it up into multiple different channels. Um, I think it's like 4,000 and something channels that sit on these things. Yeah. And when they set up a channel, obviously, it's just a little carrier frequency on there. Yeah. Now, also because it's running at 240 volts, you don't have a lot of equipment that you can natively kind of play on to try and do that detection. Like your laptop or your PC has no way of um actually, you know, it's got no analog to digital over its power source. You need something specifically designed to pull that information out. And that's really it is there's no dev kits kind of for

this. I mean, I'm sure there are. I'm sure they're really expensive. Um but and and yeah um I'm sure in the smart smart grid world there's a lot of interesting things going on. I started reading into it. It looks fun. You know people are doing these kind of like boutique housing estates where all the houses are connected over power lines and you can distribute um security cameras. People install cameras at their houses but when they plug it in doesn't connect to their network. cameras just plug into the power and your housing estate can pull all those central CCTVs together because they only know the encryption key that it's on that network. Great idea. Yeah. Don't have to

run cables, don't have to use 3G, don't have to use Wi-Fi. Um but it it's a bit of a odd use case, should we say. Cool. Um so depending on the attack um obviously can monitor the network and in some cases you'll catch these things getting plugged in. Yeah. Obviously depending on how they're set up depending on what is on the other side of the connection. If you just plug this into your network with nothing on the other side it's not doing anything. It's not receiving traffic. It's not going to generate probably any alerts. Yeah. At least in a lot of my testing, a lot of these things don't even produce MAC addresses. They act like a wire. Yeah.

And that's kind of annoying. Annoying for detection, but obviously eventually someone has to plug in on the other side and if they want to do anything on your network, you're going to detect it. Yeah. That's really where a lot of the um uh defense comes in here. The next thing is someone still needs to plug one of these in in your secure area. Yeah. Just because this plug here is connected to the whole building does not make it possible for me to just access stuff. Okay. A lot of people think that maybe I just plug this in and off I go. Yeah. Yes. But I still need one on the other side. So obviously your

physical security stopping people plugging things in going to play a massive role. Yeah. And that's really going to come down to your first line of defense. If someone was able to walk in, plug in a Wi-Fi access point, yeah, they can do this, too. Someone was able to walk in, put in a network implant, they can do this, too. Yeah. So, really, it comes down to the same problems you've always had. You just now need to look at other issues. You're not going to detect the Xfill. Yeah. In this particular case, the Xfill will be the part you can't see, but you might see the network access. you might see the actual network setup in this case where they're trying

to plug it in. So, what's next? Well, my goals were to build a really nice little solution. Um, and the goal was to present it here. Um, that's the bit of more than I can chew part of the talk. Yeah. Now, the biggest thing here is, uh, I'm not an electrical engineer. Um, I have a background in like audio technology. Yeah, a lot of this actually crosses over quite nicely with that, but in terms of the voltage side, in terms of actually playing with 220 volt stuff, uh I've I've blown up a few things. Um so I'm not going to claim to be an expert on any of this. Yeah, but I think um I don't know how much time I'm a

little short on time, but that's actually great because I'd much rather people just have a discussion and ask some questions. So, thanks a lot. Hope that was uh enjoyable. Hope you maybe have a new idea on how to use some uh very dead, very pointless technology um to do other pointless things. Cool. Thank [Applause] you. Any random questions?

So they've actually got a whole bunch of stuff in their spec. Um they've got an entire key exchange protocol. Um so they can they they use um I forget its name now. They've basically built their own kind of version of like an arc key exchange to do the transition between these things. Okay. You can obviously if you're watching the power line, you will watch all of that happen. That was what I was really hoping to detect initially and use that to sense the setup, you know, detect what's out there. Um, then I kind of killed my device. Um, which, um, these ones are very nice. They're very nice if you're trying to accomplish a goal. They're very dumb. Um, these do

not have web interfaces. They don't have anything on them. They're actually pretty terrible from that respect. But yeah, the there is a key exchange. Um, it's all visible on the wire. Um, when these things operate in broadcast mode and when they exchange information between networks because they actually do network data exchange between logical networks, all of that is actually in the plane. It's not encrypted at all. Um, very annoying because obviously they need to tell each other what names there and other funny things. Yeah. Now, the annoying thing, I say annoying because they only talk to each other if you tell them to. You have to know the networks there. So, once you've named the network

and everything, you can then start to interact with it. But before that, it's basically designed to be a plugandplay. Click the button, you've got a wire. Um, that's I say bit obscure, but I hope that helps. Cool.

Yeah. So, you can have um it doesn't just have to be pointto-point. You could have a number of these plugged into let's say a house um and then instead of having Wi-Fi or Ethernet runaround, you could just use these and you can actually have a number of them bound to a single logical network that you then pull out of. Cool.

That is exactly the kind of use case I was aiming for. I decided to leave the word airgapped out of it because it's a bit tricky. If it was truly airgapped and it had its own generator, its own power supply, it's not tied to the grid. That's it. All of your air gap stuff is powered. They've got power lines running in. If you can sit outside, you can get inside. And that's why I just said adjacent room. Thank you for saying air gap so that I didn't have to. Um, sorry. Yeah, it's a this. It's more that it's a it's an obscure thing that uh is kind of fun to think about. Like I say, is it a

real attack? You know, if someone goes and finds a plug outside this building, you know, unplugs a light, you know, plugs in one of those adapters, light to powerpoint, plugs this in, are they going to get anywhere? Probably not. Okay, your wire length is a real problem in this. Um, and I really cannot um state that enough. Uh, you do not realize how much house wire sits between plug points until you try and use these things in any kind of meaningful way. Yeah. Any other

questions? So, they're about that big. Um, you can obviously if you made your own one, this is made with cost cutting in mind and usability. Obviously, depending what your goal is. If you're looking to build a device that purely is used for Xfill, okay, there's a lot you can cut off here. You don't need your your you know your RJ45 jack. You can actually do away with a lot of this stuff because this is a lot of this is here to provide general connectivity to people with VCRs or something. Um so you can cut off a lot of this and just ignore it um and only kind of dial it down to only the parts that are relevant

to you. Yeah. The nice thing is obviously they sell these things in pairs. You obviously get them with two sides, but they also sell them where one side is actually a Wi-Fi hotspot. So like you you plug one in and connect it to your target and you sit in a coffee shop or somewhere chilled. You plug the other into the wall and you work on your laptop. Don't even need a network port. So guess what? You can use a Mac. Um so like I joke, but that's this is a legit problem. Uh network jacks are going away. So these type of attacks are becoming a bit weirder by the day. Yeah. But yeah, so as far as how much can you

cut this down, um you can see that's what the commercially available one is. Depending on how much money you want to throw at it, a lot. You can make this way smaller. Um your problem comes down to the transformers. You're physically creating uh you know your AC to put on the wire. You need transformers to do that. They're big and bulky. Um, you can probably buy some really expensive ones that are smaller or possibly do it um with transistors depending on how much you really want to generate a smooth wave. Um, but yeah, you can cut it down quite a lot. Um, I haven't seen any much smaller than this. Um, I tried to put

together a mockup of one where I basically put this inside a surge artor um with a small two port switch and um basically the idea being people plug it in and you put it between your Xerox and the wall. You put it between someone's PC, you know, while someone's not looking, you quickly unplug and plug in. Who's going to question that? Looks legit. Yeah, but it's got this in it. And now everything going through that network cable could be also going through the power lines. Maybe not. Um, is it a feasible attack? Probably not. But yeah, does that help? I've got a few devices. If you are keen on seeing more of this, you can come chat to me afterwards. Um, I

got a few different devices you can check out. And yeah, cool. I don't want to keep everybody Oh, yeah. So, um, the ones I've got, I think one of them was ARM. Um, I haven't I I had a number of ones that were TPLink. I tried to focus all on the TPLink ones. Um, I've also got the the Mercury one, which as you saw is probably the same. Um, and then I bought some random um, there's one called Pixlink. Um, because everyone knows Pixlink Network here. Um, so, uh, I don't know what that one runs because I can find like so little on it. Yeah. Um, but obviously, yeah, uh, for the testing, I've obviously spent a lot of

time for in terms of the presentation on figuring out the limitations of the tech. Um, the next step is really digging into the firmware side and setting up that side of things. But yeah, you can chat to me afterwards. Cool. Coffee [Applause] time. Across the attack surface, scattered products and siloed views create blind spots that feel unstoppable. The deadliest risks are in these gaps where attackers move in. It's time to unify fragmented snapshots into one allseeing view of risk and unleash a platform born with one intention. Isolate and eradicate your priority exposures from IT infrastructure to cloud environments to critical infrastructure and everywhere in between. This is Tenable. Your exposure ends here.