Underground fiber optics spy on humans moving above | Hacker News Reader

Underground fiber optics spy on humans moving above

wired.com

209 pointsmardiyah68 comments

That has been used 8 years ago at least to get info on Hamas tunnels across the Gaza-Israel border: https://www.timesofisrael.com/how-the-tunnels-in-gaza-are-du...

That was an interesting read, thanks for sharing.

I couldn't find any links, but I remember reading about a system (or proposed system) working on this principle from the 1980s. It was meant to be deployed in the Fulda Gap region to precisely locate moving vehicles.

Agreed, old news. Also think about apartment buildings with coax https://ieeexplore.ieee.org/document/6083650

"Human intruder detection with Leaky Coaxial cables for wide area surveillance system" 2011

https://en.wikipedia.org/wiki/Time-domain_reflectometer

Paranoid memes below

https://www.youtube.com/watch?v=Kr7AONv3FSg

https://www.youtube.com/watch?v=aPMULk7AhWw

> Because the speed of light is a known quantity, the Penn State researchers could shine a laser through a single fiber optic strand and measure vibrations at different lengths of the cable by calculating the time it took the scattered light to travel. The technique is known in geoscience as distributed acoustic sensing, or DAS.

It turns out Marauder's Map was enabled by an underground array of optical cables.

In the electrical world, this is called time domain reflectometry.

In the optical world we call it OTDR (optical time domain reflectometry). What an imaginative bunch we are /s.

You can get a tiny box to do it, to find fiber breaks and bad splices.

There's also OFDR now, frequency based. My understanding of it is pretty hand-wavey, but it's a lot more sensitive.

> pretty hand-wavey

What do you mean?

"hand-wavey" refers to (in this context) an understanding that is missing a lot of detail. The term refers to the notion that if someone asks you to explain a specific piece of the theory, you'd have to respond with "enhhhhh...." and a wave of your hand

Yep, pretty much exactly this. I understand it enough to use it as a tool and do some rudimentary troubleshooting, but to explain how it works would require some waving of the hands to fill in my knowledge gaps.

In the mid-90's I briefly worked for a company who built and installed security systems using this principle, and it looks like they are still going: https://www.remsdaq.com/solutions/integrated-security-system...

Besides acoustic waves, fiber measures strain, pressure and temperature. One proposed application is to line a water ar oil pipeline to quickly locate a leak. Currently pressure sensors every few miles. The you have to visually inspect the segment for the leak. Harder to do for buried pipelines.

Just look for all of the dead grasses where the spill is occurring for oil pipelines or patches of lush grasses compared to the other areas for water pipes. Or is that too simplistic?

That indeed works, but is a time consuming manual process, pipes arent always leaking clearly (visually), and that only works when a failure has occured. I built a few fibre based instruments three or four years ago that could detect and gauge changes in the pipe caused by wall thinning (before leaks have occured), small leaks (too small to notice soil changes) and also fouling. Longevity and robustness of the sensors was always an issue though

It's not always that easy to see. I'm on the board for my city's utilities and we recently had a leak in our 48" water main that was 380,000 gallons/day. We found out about it because of all the water coming up in someone's front yard.

After repairing it and doing pressure tests, we now know we have another leak in another part of the main. However, there's all sorts of roads and other things build along the way so it's hard to know where it is, especially if it's not as big of a leak.

This could be a lower-cost way to do the border wall with Mexico in isolated areas. Use this to detect activity of interest, then dispatch a fast drone.

> then dispatch a fast drone.

To do what, exactly?

That's a policy decision.

Spoken like a true defense contractor!

Clearly you havent watched many Hollywood movies.

To zoom and enhance

To feed of course.

Buried seismic sensors have been used by CBP for years now, there’s an entire program dedicated to using them specifically for detecting tunnels as well.

For detecting persons, it’s just an evolution of the existing process with the massive camera setups.

Sensed/taped footfalls -> closest CBP station -> agent response

Why bother when, aside from the Posse Comitatus Act[0], there are already extremely advanced optical and synthetic aperture satellites that can discern a license plate on the Earth's surface flying around the globe?

Or solar powered, regular-assed Ring doorbells (once they get that heat problem fixed). As a side effect, then CBP will be subsequently able to find their lost car keys out there.

[0] - https://en.wikipedia.org/wiki/Posse_Comitatus_Act

I imagine this system wouldn't be defeated by waiting until a cloudy day?

I believe that the synthetic aperture satellites can pierce cloudcover, probably with reduced minimum features. I'm not privy to any special information about this though, so don't really know wtf I'm on about.

But also consider its the desert, you know?

Those deserts still have a lot of clouds: https://www.esa.int/Applications/Observing_the_Earth/Space_f...

They just don't come together enough for much rain.

I did not know these ranges. Thank you, that is very interesting.

It looks like some do and some don't, proving my assumptions rather uninformed. The southern US border area looks to range from about 60-25%, distributed more towards the high end of the range.

I'm aware of coastal deserts like the Atacama which have exceptionally low precipitation but lots of foggy conditions, but I don't know of any like that in the US areas described.

Regardless, I still imagine that satellite, drone, and observation posts are the smart way to create the absurd Iron Curtain down there. Not that we should...

I worked on a project where a fence based fiber optic intrusion detection system was interfaced with a drone-in-a-box type solution. I think Airobotics had a few of these kinds of deployments in Australia.

It's still so cool to know that in high security buildings it's common to run fiber optic comms cable inside of pipes carrying natural gas. Seems like that would largely take care of the potential for seismic eavesdropping or wiretapping haha.

> the technology can’t exactly identify a car or person. “You can say if it's a car, or if it's a truck, or it's a bike. But you cannot say, ‘Oh, this is a Nissan Sentra, 2019,’” says Stanford University geophysicist Ariel Lellouch, who uses DAS but wasn’t involved in this study but did peer-review it. “Anonymity of DAS is one of the biggest benefits, actually.”

Good.

I'm sure, like all people have different gait, that an acoustic profile can be matched against.

With sufficient data, acoustic patterns could be learnt to identify the type of vehicle down to its model. You might be able tell it is a Sentra but not _whose_ Sentra.

True, and with even more sufficient data, such as context and other acoustic profiling like those from the tires and one very well might be able to detect a specific Sentra.

> acoustic patterns could be learnt

Maybe, but this is assuming a lot about the fundamental SNR of the measurement system, etc. It might go the other way, e.g. you can't even reliably tell it was the same vehicle twice, just a roughly similar one.

> But you cannot attribute it to a specific person

Until someone figures out how to do gate analysis from the data :).

This is a very cool seismography technique though. Could be using fiber while it’s “dark” to perform a data gathering use for other sciences. Like a trans oceanic cable to get a new picture into our oceans with multiple data points.

There was a paper in science [1] this year looking at seismic events over a submarine link. The fibre also doesn't need to be dark in fact.

[1] https://science.sciencemag.org/content/371/6532/931

Stanford geophysics was the first university to run a optical fiber for studying ambient sound. Now there are dozens of these for research.

> By shining a laser through the fiber optics, the scientists could detect vibrations from above ground thanks to the way the cable ever so slightly deformed.

> Civil engineers are already using DAS to study soil deformation, and biologists are even using offshore fiber optic cables to listen in on whales. (Sound propagates as a vibration, after all.) “It's just really exploding in terms of the applications,”

I wonder if ISPs are using residential fiber as an optical microphone [1] array to listen to their customers? If they are not, then I could see this as a future threat. I wonder how one might inexpensively insulate a fiber-optics cable to prevent unwanted eavesdropping.

[1] https://www.sciencedirect.com/science/article/abs/pii/092442...

Photonics engineer here: DAS isn’t sensitive or spatially accurate enough to eavesdrop like that. Also, fibre is very rarely to the home. Normally it’s to a connection box just outside the property and then co-ax to the modem. Lastly, fibre telecoms uses lots of repeaters and switches that DAS can’t see past (needs optical line of sight).

Thank you for your insight, those are all great points. The fiber to my home physically runs through my home to a box that is located inside of my home. I didn't realize that most people have the box located outside of their home.

I've had fiber into an apartment and into my current home. I don't think it is that unusual.

At least in some European countries fiber goes to a GPON terminal in the apartment and then proceeds via Ethernet cable to the home router.

I'd guess the co-ax cable is probably a US-only practice.

Virgin Media (UK) use coax. There are some FTTP installs (with coax to the router), but most, as far as I'm aware, use coax for the last leg to the house and from the wall to the router.

I've just had BT FTTP installed, and have a fibre coming into the building. They use GPON. It's wired into a modem ("Optical Network Terminal") that Openreach provide, which then presents the connection via cat5. You then plug your router in, which then uses PPPoE to establish the connection.

I was speaking from a UK perspective which, as another commenter says, is mostly co-ax from a point on the pavement. 'Actual' FTTP is coming in now, like HyperOptic, starting with apartment blocks where they can get lots of customers per one new civil engineering job. Most consumer telecoms fibre networks in the UK were laid in the 90s and don't go right to the home.

If you're in the lucky situation of having true FTTP, then it's still very unlikely such fibre could be used as a 'microphone' because of the sensitivity and spatial resolution issues. It would also be pretty obvious it was being used because it would probably interfere unless they were very careful with bands. So don't worry, DAS isn't going to be used to listen to your voice. Much easier to do that with your phone!

By the same principle it should be possible to listen to reflections of the signal propagating in the Ethernet cable that external vibrations introduce. But I guess the noise in the cable is much greater and the frequency of the signal is like 6 order of magnitude smaller making this mostly theoretical.

In the US it’s not called fiber if there’s coaxial in the line. If that were true everybody with a cable modem would essentially have fiber.

> In the US it’s not called fiber if there’s coaxial in the line

My FIOS ONT has a coax out, so that's probably not entirely true.

Yeh that seems to basically be what we call fiber in the UK!

When I lived in Chattanooga back in 2011, the Electric Power Board offered fiber to the home with an offer of a Gigabyte up and down---hence the original use of Gig city for Chattanooga. I think this was even before Google showcased fiber to the home (was it in Oklahoma City at first? I don't recall)

We had to have a router with a fiber port to connect, then Ethernet via cat 5 or WiFi.

Any idea if the fiber being used for DAS is simultaneously capable of carrying data, or do they pick a strand that is unused?

There are a few different flavours of physics you can use for DAS, and they 'use up' different areas of the spectrum. As long as what you want to do for telecoms is using completely separate wavelengths (channels or bands) then yes you can do both. Most DAS providers and users prefer to just use un-used/dark fibres though as it avoids any headaches.

I have fiber to my home. It runs from the side of my house back to a box at the edge of the park down the street. Inside that box its combined with the feeds to all of my neighbors into a smaller number of runs back to the CO. On the side of my house is a steel utility box that contains an ONT which converts the signal into ethernet. The ethernet cable runs to the other end of my house where I have a router.

I guess theoretically it could sense when I operate the garage door and drive in and out, maybe if I take out the trash. But the fiber runs through conduit. And given they way the fibers are multiplexed down the street in practice its hard to infer much from that fiber.

Same here, the fiber gets to my home.

At least for me, my fiber install is fully inside my house.

My ONT (buried inside a BGW-320 that I can't bypass -_-) is fully inside as well. I think a lot of AT&T's setup is going this route as well from what I've read.

I'm a layman network tech with no college degree and no grasp of physics beyond very-high level high school science, but I can almost guarantee getting voice audio from a fiber drop on a GPON network where the NID is outside or in the basement, and using the setup in the paper, is currently impossible. The paper you linked seems to require purpose-built hardware on either end of the fiber, and for the transmit strand to reflect into the receive strand, which is not at all how GPON works.

The paper I linked was also from March 1991. I imagine that the technology has changed since then. Here is another article on fiber optic microphones designed for operation in harsh environments [1]. It is from 1999.

[1] https://ui.adsabs.harvard.edu/abs/1999SPIE.3852..106K/abstra...

The abstract on that paper makes it seem like the study pertains to medical equipment, not telecommunications.

ISPs can’t get residential fiber to work in many cases, using it as a microphone is beyond mass capabilities.

Not to say that if the alphabet boys are interested in you they might do something similar (but in that cases there are likely cheaper methods).

The real world answer is 99% of consumers carry a battery powered microphone and GPS beacon with remote telemetry, so there's no need.

Some also pay money to put internet connected microphones in their house.

Energy companies like these for collecting data in harsh environments (e.g. hot geothermal wells) which would degrade conventional electronics quickly.

Also with magnetic influences, like thermo sensors in large transformators, breaks or motors.

Aggregating and processing enough fiber data might be able to work towards identifying mass, contact patch size and shape, position, and velocity.

Looking to learn

Interesting new use for existing tech, somewhat clickbait headline. Also, the term is "Jerry-rigged". [Edit: I stand corrected.]

Both are actually fine. https://www.dictionary.com/e/jury-rigged-vs-jerry-rigged/

Reply on news.ycombinator.com