The next big Wi-Fi standard is for sensing, not communication (2021)
staceyoniot.com
staceyoniot.com
> The link-rate per stream is 44 Gbit/s, with four streams this goes up to 176 Gbit/s.
Yes, it can't penetrate walls, but is huge for point to point communication. Think VR headset bandwidth, file transfers between devices, wireless backhaul between buildings. It's not just for spying on your cat.
I think it's important to remember that it's not magic, it's just radio waves. You can "sense" with a lot more resolution if you use a camera and light bulb. There are plenty of eyeballs, cameras, and lights all around you right now. Even worse, someone could be looking in your window with binoculars.
(Visible) Light has a frequency of around 400-790 THz, which explains why it has a inherently better resolution.
> Li-Fi connections are broadcast over the air through a light-emitting diodes (LED) broadcaster and support rates up to 100Gbit/s ... Li-Fi can also serve to identify an object’s indoor position more accurately than Wi-Fi or GPS used today (less than 2cm and less than 3 degrees of orientation while it is providing real-time localization (less than 34ms). This accuracy is vital in multiple applications such as navigation In-Door Robots and Drones, Virtual Reality (VR), Augmented Reality (AR), and Gaming, among others ... it cannot go through walls, and thus a private local area network (LAN) can be created by lighting up a closed room ... Any organization that needs to keep information within the four walls, such as military bases and banks, can use the technology to keep data restricted to a single room.
> They claim visible light travels faster than radio waves
I wasn't able to find it on the lifi website but seems like some poor copywriter got something wrong
> their thousand dollar router can only do 100Mb/s
Who's they? In any case, since lifi.co is advocating Li-Fi as a whole so it makes sense they talk about theoretical speeds, whereas a specific router from some manufacturer might not be able to reach those speeds, considering how new the technology is
In the end, I imagine Li-Fi doesn't have much of a future outside of niche markets because it has to compete with wifi which is everywhere (and it can't supplant wifi because e.g. it doesn't work in your pocket)
Another big advantage of LiFi is that the usage of light allows LiFi connections to occur almost instantaneously because light travels at extremely fast speeds. This results in faster transmission of data and faster internet connections – about 100 times faster than speeds achievable by WiFi.
https://lifi.co/lifi-products/lifimax/
$1098 Guaranteed fast data transfer with 100 Mbps download & 100 Mbps upload.
I hope then you can dissable the lower frequencys, because now you have 5 or more APs in your appartement.
Can we stop downplaying privacy issues please?
With this technology, it will be integrated into your router (and clients?), so turning those off will mean to turn off WiFi altogether. Added to that, I'll bet that most people wouldn't even begin to imagine that WiFi could be used for sensing people, and they'll continue using WiFi without being aware for the grave privacy implications.
It's rather unreasonable to think that you won't be able to turn off band on router. Especially given fact that 60Ghz is heavily shielded by _doors_, so any usable router would require fallback 2.4/5/6 Ghz connectivity.
Even if Neighbor 2 turns off their router, Neighbor 1 could passively use the radio waves from Neighbor 3's router to surveil reflections from Neighbor 2, including keyboard typing, heartrate/breathing, location and physical activity.
Previous discussion: https://news.ycombinator.com/item?id=27123493
It makes it suddenly ubiquitous when all new WiFi devices use the new standard.
Wrapping my house in tin foil is sounding a lot less conspiracy theorist now! Actually my house is going to be stuccoed - I wonder if they are using a metal mesh that I can ground. Hmm...
There needs to be a balance, location information will be very useful for a lot of things, but only if we have it fully implemented. I want things to sense when I fall and call for help for me. (If I was at a big risk I'd have a button, but for the average young person the risk is non-zero but very small). I want my home automation stuff to figure out where they are and configure themselves. I want my routers to suggest that I'd get better coverage by moving it.
Yes there is privacy concerns and they need to be addressed, but don't lose the good with it.
The Verge includes a whole section on stuff you agree to when purchasing a new laptop when reviewing them. Maybe it's a good idea for us to start insisting on some level of privacy regulation on new tech. This is purely regulatory right, nothing preventing things like VR--just preventing FB from knowing everything you're doing while using their VR headset, and selling that information to others. Maybe that creates an economic constraint, but this seems like a choice consumers can make. For example Amazon sells 2 versions of every Kindle: ad supported and not.
It can be asked once at the device level, but regulated information needs to be specified at the highest level (e.g. "location information about an individual"), and then the concrete technical implementation left unspecified.
"Does it, or does it not?" is more important than "How?" Because legislation specifying "How?" will always lag technology by decades.
(But then again, this probably should have been done in the mid-90s, once it was apparent which way cookies were going)
Who on earth is actually not an advocate for privacy?
> I want things to sense when I fall and call for help for me.
There's no need to sacrifice privacy to achieve this. [0]
> I want my routers to suggest that I'd get better coverage by moving it.
This is really more an issue of proprietary software than a privacy concern, no?
> Yes there is privacy concerns and they need to be addressed, but don't lose the good with it.
Communications are by default expected to be private. Highly public and non-private communications are advertised as such since the use case is so specific and often times, nuanced. In almost all circumstances there is hardly any "good," at all without best-practices privacy protections included.
This standard makes my wifi router into an internet connected, closed source blob, light bulb and camera that is impossible to obscure (without killing my internet connection).
Sure, I can wire my whole house for ethernet. And, while I’m behind the drywall, I might as well rip off all the drywall put a layer of Al.
Here’s a business idea - drywall with aluminum fibers embedded in. For those of us who don't want our sexy times recorded by our router.
EDIT: made wifi’s ability to see through walls explicit per _jal’s comment.
QuietRock has drywall that combines soundproofing with RF shielding (steel plate), but it's expensive and targeted at military use cases, https://www.quietrock.com/sites/default/files/QuietRock_530R.... Aluminum radiant barrier can be layered with standard drywall.
There is conductive carbon paint for RF shielding, e.g. http://www.emfrf.com/shielding-of-electric-fields-in-a-resid... & https://www.zokazola.com/emf_reduction.html
Other materials: https://news.ycombinator.com/item?id=29902926
Your creepy neighbor in the apartment next door is getting a new toy.
500 terahertz = 600 nm (visible light)
60 gigahertz = 5 mm
6 gigahertz = 50 mm
5 gigahertz = 60 mm
2.4 gigahertz = 125 mm
- Unless 2.4 and 5 GHz are taken off the standard, my router will see me through the walls.
- At 60 GHz, if I have wifi, my router can see me
- To make our concerns clear to you, even at 125 mm resolution, you can tell if I like my wife to drill my bottom with a strap on. Or you can tell if I piously say my prayers at night (something that is increasingly political). Or if the immigrant next door spanks his kids. Or if his 15 year old daughter is “exploring” her body.
And for what benefit exactly?
So an ad agency can track me and know Im struggling with my diet, hovering over the fridge door?
Plenty of non-immigrants (won't even dignify that with a label like Americans, because immigrants can become Americans too) spank their kids or hit their wives etc....
I think your fears are real, but I also think it's important to be careful with your examples :)
The community would then share privacy-enhancing I/O profiles for every kind of device. If years later e.g. my adversarial lightbulb pivots to brokering kompromat SIGINT, I want to filter that out, and I want not to be the first one to write a filter like that..
Does it change the internal state of the lightbulb’s logic? No.. but hopefully it would even be able to simulate the state loop of the lightbulb well enough to guess what to filter out.
Of course there may also come a time when somebody starts to sell $20, 60 ghz-spectrum-only-visible “human activity fakers” to disrupt the collection of such data. Maybe with a “Honey, I’m home!” package being the most popular, lol
The ancients had a solution for this (brick walls).
Frequently we create the tool in order to kill with it.
After that we also use to tool to create.
We're equally capable of good and evil and yes, we are frequently good, too.
Over these years I also realized how little you can do to actually stop research, no matter what your opinions of it are. Even if something is officially forbidden, you can be sure someone else works on it if it's interesting enough, so in the end you're at a disadvantage. There are many examples of it nowadays.
If you look at my comment, you can construe it as even being more negative and cynical, since I'm saying that instead of creating a tool and finding ways to kill with it, we're frequently building tools just to kill with them, and only later we find ways to use them for something non-lethal.
Edit: Turns out there are other comments below mentioning this.
What do you mean? Most consumers have a wireless router, and that works for their needs.
Even if that is a concern, I don't see sensing as being a benefit.
If you don't have the expertise to run ethernet cable yourself, just find local companies that install alarms - they have lots of ways to run and hide cables; it's their job after all!
It's much worse: neither crack nor connection is required. The technology is entirely passive, it only uses reflections from Wi-Fi radios, same as radar. No connection to the router is required, hence the acronym DFWS (Device-Free Wireless Sensing).
To be clear, this can be done today with $20 ESP32 WiFi devices + custom firmware, i.e. any motivated attacker can already see through the walls of homes and businesses. The Wi-Fi 7 Sensing draft standard is proposing to make this available to everyone.
Perhaps we need a celebrity to help demo SENS transparency of their home walls, to help consumers and regulators understand the implications. That could motivate research investment in privacy controls.
More details: https://news.ycombinator.com/item?id=29901979
As your sentence before this says, it already is available to everyone. The only thing wifi 7 opens up is someone with less technical know-how being able to flash openwrt and install some program that extracts and/or visualizes the sensing data, but $20 ESP32's isn't much of a cost to someone looking to use radio wave sensing maliciously right now.
It's like you could make a magic flashlight that sees though walls, vs. flipping a switch that turns every wall to glass. It's not a perfect metaphor, sure, but the point is that it's kind of a big step to flip that switch, throw open those floodgates, and to do it with so little fanfare.
Exactly. After 10 years of research and hundreds of published papers, there is a huge gap to be bridged with consumers, regulators and lawyers, before dramatically changing the definition of "fitness for purpose" in wireless networking, https://dhalperi.github.io/linux-80211n-csitool/#external
If a maliciously controlled router wants to track you it already can because your phone, smart watch, and laptop already broadcast themselves.
This technology may not seem useful to the purpose of WiFi on it's face but that couldn't be further from the truth. Knowing the population of a room in terms of devices but also people is useful for WiFi deployment planning and power level optimization. 2000 people with 2 devices is different from 4000 individual people.
People already have WiFi enabled lights and other sensors so providing a standard for object & people detection will make those use cases even better.
E - Expectations
L - Lifetime
M - Motivation
E - Environment
R - Reward
:-)
Yet another reason I prefer Zwave for my home automation communications. Sure you could probably use it in the same way, but it won't be handed to you on a silver platter like the wifi stuff discussed here!
Other than that, I do have a lot of ESP32s at my place sensing for activity via IR as well as by creating an FFT-"audio" log. No sound just FFT aggregated over one second, but all these devices store the data on my home server, not in some cloud.
I guess author wants to classify sounds in his environment too.
For example my bathroom has two IR sensors (https://www.amazon.de/dp/B08LBDYPYD), one inside the shower, one outside of it. This lets me log the time and duration of when I take a shower or how often I shave.
It also has an I2S MEMS microphone which collects the data for the FFT where the goal is to replace the IR in the shower with it since the water sound is enough information.
Also temperature and humidity, where the latter also gives a good indication of when I showered and when I forgot to close the window some time after showering.
My bedroom also has the same "setup", where the FFT shows me how big of a problem I have with snoring. Eventually I can add some logic on the server to slightly wake me up if I snore, but currently I'm just logging. The PIR serves as a measurement of how much I'm moving at night without requiring a camera for this. Though I have another ESP32 under my bed fixated at the slatted frame which contains an 9dof-IMU and logs any significant movement which may be related to quality of sleep. Also the magnetic field which is why I have a strong magnet attached to my bed tablet, so I know when it changed position. Even though the tablet knows that itself because it is logging its own accelerometer every minute.
I live alone, so all of this is no problem.
At this point I'm mostly only logging, and only using the shower and bed data to see when I used them, which I can then see in Grafana. I think the data will only gain value when watched over a long time frame.
You'd also be surprised how powerful smart cards are these days. the aforementioned NXP chip has 586 KB ROM, 144 KB EEPROM and 11 KB RAM, crypto coprocessors for RSA/ECC/DES/AES, and a 32-bit CPU.
60 Ghz is heavily blocked by walls, glass and doors, even more than 5Ghz, so that's very unlikely scenario.
Wi-Fi sensing works with 2.4 Ghz and higher.
As many APs today are already cloud dependent I can totally see where all the sensed data will be stored and out of the owners control.
“Your honor, wifi data clearly shows the defendant was…”
Through all these years, ethernet always works, wifi is more often than not the culprit of connectivity issues. Moving my chromecast on to ethernet (which was one of the closest wifi devices to the router) has lead to higher quality, less stalling.
i've had my 5ghz wifi disappear because of weather radar, can't say the same for my cabled connections.
> it has been shown that SENS-based classifiers can infer privacy-critical information such as keyboard typing, gesture recognition and activity tracking ... since Wi-Fi signals can penetrate hard objects and can be used without the presence of light, end-users may not even realize they are being tracked ... individuals should be provided the opportunity to opt out of SENS services – in other words, to avoid being monitored and tracked by the Wi-Fi devices around them. This would require the widespread introduction of reliable SENS algorithm for human or animal identification.
Would this require a worldwide database of biometric signatures for each human that opts out?
You will be able to take legal action against the café, but it won't make a difference.
SWAT teams, home burglars and the NSA will all be clients for the data.
Once the granularity gets fine enough, movements involved in prayers, sex and elimination of waste can all be monitored remotely by state actors.
Malware from around the world will now be able to transmit details of your body movements back to their controllers.
Imagine being able to blackmail you because malware detected you having sex when your partner was away or performing Muslim prayers in an islamophobic society.
It may also increase market demand for light-based wireless networking (Li-Fi) which has the advantage of being stopped by existing walls.
First Li-Fi tablet was announced at CES 2022, https://electronics360.globalspec.com/article/17597/first-li...
You mean like how we got ourprivacy back through innovation in web browsers?
Also we already have wifi-proof paint.
QuietRock 530RF drywall will block sound and RF, https://www.buildsite.com/pdf/pabcogypsum/QuietRock-530-Inst...
Or on a room level? How do you control for RF leaks? How do you handle windows and doorways?
I’m fascinated.
Windows = RF-blocking conductive mesh: copper (expensive) or nickel-copper-polyester.
[1]: https://www.linksys.com/us/support-article?articleNum=317804
[2]: https://www.businesswire.com/news/home/20171114005542/en/Nex...
I agree with other comments that question the ratio of benefit vs. risk. But I’m an old fogey who feels similarly about other “smart” home devices.
But there's a big difference in custom firmware used in a few products and IEEE standardization in every new WiFi access point.
We need more funding for countermeasures, https://ans.unibs.it/projects/csi-murder/
> Imagine that someone wants to illegally track the position of a person inside a laboratory, for instance to measure how much time is spent doing different activities at different desks, as depicted in the upper picture. How much effective can this attack be? ... With CSI-MURDER, the localization becomes impossible because results will seem random, thus preserving the person privacy without destroying Wi-Fi communications.
400+ (!) papers on wireless sensing, over the last decade https://dhalperi.github.io/linux-80211n-csitool/ & https://groups.csail.mit.edu/netmit/sFFT/soda_paper.pdf & https://dspace.mit.edu/handle/1721.1/103715
I think it is important to note that different people will have different inputs into benefits and risk. E.g. a 24 year old male living alone probably doesn't have a use case for elderly fall detection, etc. But, a 75 year old woman living alone probably does.
Also, I am of the opinion that is possible to do this properly (i.e. securely and privacy preserving). Case in point: people carry microphones (in phones) that can record any audio with them... but they do it because they trust that their phone is secure and private.
I cannot fathom any good use for this technology.
There isn't a use case for this that isn't abusive outside an industrial setting.
I think if ISPs push for these in routers we will see comeback of rewiring houses with ethernet points.
Back in the real world it would be frightingly effective.
How do you know if your sepf made faray cage is leaking?
Are you going to get an expensive spectrum analiser? Do you know how to use it?
I can go on the street and find a random person who understands networking withing 10 minutes. I have never met a RF engineer in my life.
Using ruler. If holes are smaller than 5mm, it will block 60Ghz.
[1] https://github.com/CTA-Ripple/Ripple-v.0
[2] https://9to5google.com/2022/01/06/google-soli-radar-ripple/
https://staceyoniot.com/two-companies-show-where-our-uwb-fut...
> Outside of fine-grained location tracking, UWB is also used for radar sensing. Google is using UWB for its Soli radar, which is embedded in the most recent Nest Hub display to track respiration during sleep ... Novelda, argues that UWB sensing is more effective than the traditional motion sensors that hotels have historically deployed because they can sense a person even when they are still by “seeing” micromovements made by the person’s breathing. Lenovo embedded the Novelda chip into its laptops as a way to keep the screen on only when a person is in front of it, which helps save on battery power.
http://law2.umkc.edu/faculty/projects/ftrials/conlaw/kyllo.h...
This seems primarily about providing tracking capabilities to operators whereas ToF provides tracking to the people who are actually trying to do positioning. The privacy implications are worrying.
https://www.cisco.com/c/en/us/td/docs/wireless/controller/te...
The Linksys docs even mention how you may have to tune the sensitivity to filter out motion from your neighbors...
Security
Environment Control
Wake on approach, walk away lock
Conference room occupancy
Gesture control
Elderly monitoring
Baby monitoring
Smart housekeeping
Advertisement signage efficiency
Motion detection
Proximity detection
People counting
Human activity recognition
Vital signs (e.g. heartrate) detectionHere we go
"Tactical suggests there are 14 heartbeats in the room"
"During the interview the candidates heart rate increased during questions on their reasons for leaving the last position"
ohhh. There is sooo much here :-)
That discussion is illegal under current laws. (you cannot use a lie detector in interviews) Well at least in the US, check your local laws.
Short term smart bots that can navigate and clean a cluttered household with little kids and toddlers knocking about.
Longer term, maybe nanobots.
It could be Rosie from the Jetsons but more likely it’s Amazon analyzing my poop density to sell me some goofball supplements.
Philip K. Dick would be proud.
When pressed to come up with a motivating vision, the big idea was lights turning on automatically when you enter a room.
I guess? Like a lot of R&D, it seemed a solution in search of a problem.
Was this really it? Or did they have some use-case not currently covered by motion sensors?
I can see some potential use for security (cheap REX sensors can be easily tricked, allowing unauthorized access into secure areas). The fix is obvious (stop cheaping out on your REX sensors) but I can envision someone trying to bill wifi sensing as the hip new solution for REX systems.
Or, you know, if a cop wants to stalk their ex-gf, or an isp tech support nerd wants to stalk a customer. Or any script kiddie or petty criminal can get hold of your wifi password or drop his own AP in your house.
Nothing to see here. I mean, what could possibly go wrong?
But yeah, I think we can all expect this will be used to monitor behaviour in stores, to spy on people in homes, to rob us and to throw more "relevant" advertising in our faces.
brings new meaning to beam forming! I like it.
I can think of few real world applications. Eg. I have autism and air filter. Air filter working on full power is very uncomfortable for me, but with position sensing, it can be configured to run more often when I'm not in room. Motion sensing isn't exactly a solution, because I can be very still.
Automated light management is sensible too. It can also be used to automatically turn off lights if I fall asleep.
> Great article. To get out ahead of this, I’ll go ahead and pre-write this for the IEEE. They’re going to need it in a few months.
> “The IEEE has heard the concerns of consumers and takes privacy very seriously. We recognize that the potential value of the proposed standard is enormous, but that it must be paired with technologies that prevent it’s misuse in an ever more dangerous cyber world. As such, we are delaying the implementation of this technology in the .bf standard, and we will re-address it once we have come to a consensus across all stakeholders including privacy and consumer advocates.”
(2012) military use case, https://www.popsci.com/technology/article/2012-07/seeing-thr...
(2017) industrial use case, https://www.digitaltrends.com/cool-tech/wi-fi-radiation-tran...
(2020) slide #20 on use cases for Wi-Fi 7 SENS, https://www.intel.com/content/dam/www/public/us/en/documents...
(2021) HN threads on "seeing through walls" radar for human monitoring: https://news.ycombinator.com/item?id=28479057 & https://news.ycombinator.com/item?id=27123493
(2021) IEEE working group, https://arxiv.org/pdf/2103.14918.pdf
> This paper reports ongoing efforts by the IEEE 802.11bf Task Group (TGbf), which is defining the appropriate modifications to existing Wi-Fi standards to enhance sensing capabilities through 802.11-compliant waveforms.
(2021) panel discussion on RF sensing, https://staceyoniot.com/what-you-missed-at-the-rf-sensing-ev...
> we could use terahertz spectrum to detect specific molecules and in turn use terahertz frequencies and radios as a way to track specific ingredients in food or pollutants in the air.
(2021) EU-funded CSI countermeasures research in privacy protection, https://ans.unibs.it/projects/csi-murder/
(2022) product launch ahead of the 2024 standard, https://www.theverge.com/2022/1/3/22864783/sengled-smart-hea...
> The Sengled Smart Health Monitoring Light is a dual Wi-Fi / Bluetooth bulb with built-in health monitoring using radar technology. It’s designed to track your sleep and certain biometric measurements, such as heart rate and body temperature, as well as other vital signs. The bulb can work alone, or multiple bulbs connected via a Bluetooth mesh network work together to create a virtual map that can help detect human behavior.
> In 2012, her work on Sparse Fourier Transforms was chosen as one of the top 10 breakthroughs of the year by Technology Review ... In 2014, on the celebration of Project Mac's 50th anniversary, her work on X-ray vision was chosen as one of the "50 ways that MIT has transformed computer science."
This is awesome!
I've been frustratedly trying to tell people this for a few years now. Glad it's finally "coming to light".
^^^ this pun has something to do with electromagnetic radiation
Co-founders are welcome to contact me.