Lockheed is now tracking phones and walkie-talkies from space | Hacker News Reader

Lockheed is now tracking phones and walkie-talkies from space

jackpoulson.substack.com

344 pointsed-209196 comments

> including satellite phones, walkie-talkies, cellular towers, and GPS.

How can they track gps devices? I thought gps devices are passive in the sense that they don’t emit any signals but only read signatures from satellites.

Or am I reading the article wrong?

Every radio has a local oscillator that can be detected at some distance - from space however? I am highly skeptical.

I think the gist of the article is that they are detecting active transmitters, not the IF of receivers.

SMH. The majority of oscillators are shielded to prevent interference and to pass regulatory approvals. Civilian GNSS receivers are passive devices.

There are some dedicated GPS devices that phone home for things like tracking trucks.

Ah, but do they use cellular networks, which I presume is the standard, or some other technology?

Probably application-specific. Satellite for ships, cellular for commercial trucks, radio for remote mining trucks.

There are also panic buttons for remote backpackers that collect gps coordinates and send them to a satellite. All things I wouldn't call satellite phones, even if they're basically just more rudimentary versions. Looks like the quote was referring to something else, though.

They use cell service or sat phone or some kind of sat service like GlobalStar. And they're suggestively called trackers, so there is no expectation of privacy.

GNSS receivers are passive devices just like regular radio receivers, but in addition to regular radios they use math to process the deltas from GNSS satellite clock signals to pinpoint their position.

This isn't a function of GNSS modules. It's a combination of a cell phone module, a GNSS module (or a combination) with an embedded controller comprising an element of a field force/fleet management system.

GNSS jamming devices.

> (U//PROPIN) Additionally, HawkEye 360 satellites can detect RF signals in the GPS bands, as well as other GNSS systems, e.g. GLONASS. If detected by all three spacecraft, the RF energy from a GNSS interference can be geolocated to its point of origin using an adaptation of the company’s existing geolocation techniques. Due to the global coverage of the constellation, geolocation of GNSS jamming or spoofing can be conducted over denied areas without exposing airborne or terrestrial sensors or personnel to hostile conditions.

So you can target these jamming devices with force projection and disable them to improve location services for local assets (force projection can typically switch to ring laser INS or terrain guidance when GNSS is degraded or unavailable during terminal phase).

Just to be clear "force projection" is a euphemism for bombing the shit out of something?

Parking an aircraft carrier 2km from your shores counts I guess?

Yes. The employment of which is also euphemistically known as a "kinetic military action".

The article claims:

> HawkEye 360 satellites can detect RF signals in the GPS bands, as well as other GNSS systems, e.g. GLONASS

Seems like a misunderstanding. They claim detection in that frequency, not GPS devices that listen to that frequency. So jammers, probably.

Without knowing anything resembling specifics: (most) antennas are senseable -- they absorb a given frequency by design and ergo appear as 'black' when imaging a given spectrum.

Exceptions to (most) include plasma antennas, which can disappear on demand.

Most antennas actually have a large radar cross section (RCS), with simple dipoles being some of the worst. An infinitely thin dipole (a current filament) has a non zero RCS even though it has no physical cross section.

In theory, a reflector antenna can have a zero RCS, so putting a reflector behind a dipole will reduce the RCS. It’s totally non-intuitive.

"HawkEye’s current constellation of 21 satellites is trained to locate the sources of electromagnetic emissions with wavelengths ranging from roughly 2 meters down to 2 centimeters, with “Signals of Interest” including satellite phones, walkie-talkies, cellular towers, and GPS."

Despite the headline, the system does NOT track individual cellphones or GPS receivers. Cellular signal from a personal cellphone isn't strong enough to register at satellite altitude. The same goes for non-milspec walkie-talkies.

Depends on the satellite and the altitude. SpaceX/Starlink have claimed their new satellites (in LEO) will be able to provide limited service to existing and unmodified mobile phones with a view of the sky.

Amateur radio operators with just a moderate bit of gain (small uda-yagi antenna or the like) uplink to LEO satellites with ~5 watts all the time. At peak power a cell phone can be about 3 watts... but their antennas are not directional. So that loss of ~8dB of antenna gain has to be made up by increasing the sensitivity or gain of the satellites in LEO by increasing antena aperture or better front end low noise amplifiers. It's all very feasible given the clear line of sight.

What's less feasible is having enough of these at good positions overhead of a single cell phone to multi-laterate it's position using shared timestamps on the downlinked spectrum from multiple satellites.

I think it's entirely feasible with the budget of a medium space services corporation that can launch 21 satellites.

Civilian GPS receivers are passive reception devices.

> Cellular signal from a personal cellphone isn't strong enough to register at satellite altitude.

This isn't entirely true. A primary limiting factor is the frequency-specific gain of satellites' antenna arrays. With a big enough antenna and fast and sensitive enough ADCs, much more is possible. Sifting through TD-SCDMA is the fun part.

I can easily communicate with LEO satellites with a normal HT and a bit of gain on the antenna. It stands to reason that with better sensitivity receivers you would be able to pick up cell phones pretty easily, though identifying specific units is likely not yet possible.

"That is impossible" is literally the bread and butter of signals intelligence.

Measuring light bouncing off a window to hear conversations, using millimeter waves to see through walls, taking photos from space, and planes that could fly themselves were all the realm of science fiction at one point - while in the hands of intelligence agencies.

Still, the whole "this is what we have imagine what the gov has" is usually wayyy overrated of a take and IRL it's usually way more boring and incremental, closely tracking commercial industry.

In most cases it's their ability to spend money to build large teams to employ tech, the complete scale of operations (XKeyscore+TEMPEST comes to mind), and the care they take is where things like the NSA dominate vs the average ability of the best hacker. Their unique advantages are only rarely in the individual technological leaps, which either a) industry has no match of or b) well informed technical experts are unaware of, like the things you reference.

FWIW you almost certainly can at minimum generalize the location of a phone if you know roughly where to look via beamforming [0]

Massive antenna arrays and SIGINT satellites have massive funding and decades of history, and they've already shown interest in tracking literally everything else [1]

This isn't some star trek type of thing, this is a "would the USG dump enough money into SIGINT to implement it?"

I'm very confident the technology exists to roughly track specific locations of interest for specific devices if not more.

I understand people may disagree but it would be foolish to assume it's beyond the realm of possiblity.

[0] https://space.stackexchange.com/questions/55143/is-it-possib...

[1] https://www.aerosociety.com/news/eavesdropping-from-space/

> Measuring light bouncing off a window to hear conversations

Laser microphones are old too. Clear and Present Danger features the use of a laser mic and that was written in 1989.

>Cellular signal from a personal cellphone isn't strong enough to register at satellite altitude.

Sure it is. The iPhone can do that right now with emergency SOS.

Starlink does not communicate with iPhones.

Yea it does: https://direct.starlink.com/

Both of you guys are half right, Starlink does not currently talk to iPhones, but it will be able to as Starlink Direct is intended to appear as a regular mobile tower. The emergency satellite SOS feature on iPhones is with another constellation.

There are LTE-over-satellite systems, cellular signals absolutely can register at that altitude.

I just learned about the SBIRS mission that Lockheed built, which is a constellation of far earth orbit satellites that are constantly scanning every inch of the entire globe with IR cameras with missile/plane/boat/person(?) level resolution. And has been for at least nine years now.

Incredible.

Here is a video that explains it. https://www.youtube.com/watch?v=mDTnl4E9FiY

There are hundreds if not thousands of those up in the space. You can even rent one yourself as a civilian (order one to point the camera/SAR at a specific location).

Why was it so difficult to determine the MH370 flight path then?

Because who is looking in the middle of the Indian Ocean at commercial airliners. There’s far more important targets and areas for these systems to focus on

Yes indeed. You can see for example here what kind of lead times are we talking about: https://sar.iceye.com/5.0/productguide/ordering/

Any serious military is tracking all known imaging/SAR satellites and uses their fly-by times (which are entirely predictable) for moving stuff covertly. I.e. you want to be out in the open in Area 51 testing a new RQ-180 derivative only when no one can be looking at you right now.

Also, there are global ADS-B satellite relays, and frankly I don't know why this data wasn't logged for MH370 - it must be explained somewhere. There's a guy on Twitter who captures those on a custom antenna + BladeRF setups (this data is plugged into adsbexchange for oceanic routes coverage). AIS relies even more on satellite relays, as the curvature of the planet limits signal propagation much more than for airplanes. What you see on any AIS website is mostly from a satellite.

Someone disabled the transponder before they turned off into the ocean is why.

Apparently the ADS-B transponder on MH370 was received but stopped suddenly.

Interestingly there are 7 hours of logged Inmarsat Classic Aero mobile handshake data from (IIRC) the "smart engine" telemetry.

The logged transmission round trip GroundStation -> Sat -> Aircraft -> Sat -> GroundStation provided an non unique "arc of travel" across the globe that was used by Australia to narrow the search area to a truckload of ocean floor that was substantially smaller than the entire planet.

The archived versions of the logs saved by the Malaysian gov are:

https://web.archive.org/web/20160304040818/http://www.dca.go...

Alternating sat routing would have given a more definitive fix and removed the ambiguity.

Many have lost trust in the WSPR data and Inmarsat pings. (Example: https://twitter.com/JustXAshton/status/1699535126663688544)

Your source conveniently omits weird data in the logs that start at 18:39:55UTC (a bunch of zeros, like something errored out).

Satellite pings switch to IOR at 15:59UTC and never change back to POR despite the plane flying east towards it. Going east until 17:21 UTC the plane should have pinged POR but doesn't

That's the archive.org mirror of the openly declared edited from raw logs version released by Malaysia .. so not "my source" which, back in the day, were the raw logs from ground stations.

The specific issue I have with hyper focused attention on these specific logs and hoorah about "weird data" is that I haven't seen much in the way of properly comparative analysis of a super set of broad fleet Inmarsat logs ... these in general are the trailing edge of quality data, riddled with nuls, semi handshakes, line noise, etc.

Transponders are for positioning, black boxes for logging important flight data, diagnostic interfaces for mechanics in hangers ... and component level Inmarsat .. pretty much not more than a curiousity of "because we can" that's occasionally useful when a properly formed messages truly indicates an inflight issue.

Ping routing and ping times were never meant to be consistent nor used for gross "half-LORAN" positioning .. this is all happenstance after event signal engineering hack work.

No satellite was pointed at the path when MH370 was flying.

So it isn't actually true that it is "constantly scanning every inch of the entire globe with IR cameras with missile/plane/boat/person(?) level resolution" then, or it is true and they know exactly where the plane went down they just don't tell anyone about it for some reason. Its either or.

Constantly scanning != Constantly recording and storing in perpetuity.

I think "repeatedly scanning" would be a better phrase.

Every inch of the globe is imaged every N hours, by satellites making continuous observations. Most of the earth is not under sensor coverage at any particular instant. Both are true.

It’s possible that NRO has a later observation of MH370 than any that have been publicly disclosed, but it would be sheer luck to have, like, video of the disappearance.

Some say otherwise. I don't know what to make of it, but here you go: https://www.youtube.com/watch?v=AwaC4AXFqRI

The satellites were all too busy monitoring HN users.

Judging from this thread, yes.

I have direct experience being tracked by this technology. With photo evidence of a utility he360 truck in the field after several loop backs on the road. The technology can be used to swarm a target of interest. It is my opinion from experience all skilled ethical hackers and non ethical hackers are tracked. I say this knowing my surveillance is about to be doubled down now. Yay.

Edit: to their defense… my laptop was compromised at the time due to a Realtek chip supply chain hack that was persistent. The exploit was associated with “redgobot”.

Frequencies are combined to create unique footprints of any given target… ie car satellite or mobile connections, radio stations, devices, chips in head… you know the usual.

Get used to it, my friend. There’s a reason the TI phenomenon is growing, and it isn’t just baseless paranoia.

I do not identify as a TI btw. I identify as the average Joe American citizen that may or may not be aware of the undercurrent that defines their existence.

Maybe there was much less of this in 2014?

Assuming it was monitored by a classified system, would they release the information? Recall rumored IUSS role during the Titan implosion situation and consider Malaysia's position, geographically and geopolitically, between the US and China in the context of sharing monitoring capabilities. If those capabilities even exist.

bingo. it either landed somewhere secretly or crashed. In both situation there was nothing aboard, secret cargo or important politicians, that would justify unveiling of their capabilities to track everything everywhere at all times.

> Incredible.

Incredibly dystopian. Since at least nine years ago, there hasn't a single place on the surface of the planet where you had any total and absolute privacy.

Cameras are so cheap now, and prices will keep falling, that I think in the future pretty much everything will be filmed.

I'm not saying it's good or bad. But it is inevitable, and we need to adjust to that reality.

They are already flying over some big cities and constantly capturing everything in car-level resolution, then when a "crime" happens they can roll the tape back and see the where the car came from.

https://radiolab.org/podcast/eye-sky-2306

"Persistent Surveillance System" google blackholed of course

This is why everyone should vinyl wrap their cars. Then if anything happens, just peel it off.

Also why you should wear a ski mask in every bank, in case “something happens”

Perhaps the fact that it all of a sudden became acceptable (to put it mildly) to cover the lower half of one's face has some benefits after all.

After all, who could object to a safety measure?

A big point from the podcast isn't that the goal is to identify you or your car later as you might think. It's that they can "rewind" the footage and trace where the offender came from and show up at their doorsteps within minutes/hours. It's as much about traceability as identification. A vinyl wrap won't help that.

Yep, a video about it capturing a murder (blurry video from 7 years ago I'm sure it's even better today):

https://www.youtube.com/watch?v=mJLr0KMsRAA

David Brin has idea after thinking about future making surveillanace unavoidable. The solution is that everybody can spy on everybody else. The problem is that the government and corporations can spy on us, but are immune. I'm not sure if he means it as blueprint or inevitability with the price dropping.

I think he is wrong that everybody spying will help, when everybody discussing online doesn't. But it is useful to think about what world will be like when everybody can surveil.

Cameras are already cheap enough that even fairly poor people can afford 24/7 cameras in and outside their home.

The linked article shows they’re not wholly immune. Some publicly available technologies can be quite powerful as Bellingcat demonstrates.

It’s certainly an asymmetrical field but, especially moving forward, I doubt it will be an entirely one-way street.

The #MeToo movement made me realise that we will probably choose constant surveillance for our own protection, rather than have it imposed on us by some Orwellian government.

Can you elaborate on what this has to do with the #MeToo movement?

The fact that verbal allegations of events decades ago were enough to sink multiple people’s lives.

Hence, people will choose to record every minute of their lives out of fear of something similar happening to them.

Its been that way since probably the 1960s and has yet to materially impact your life I imagine.

Bold claims require bold backing.

It's also in the country's favor to overclaim their capabilities.

It may have materially impacted many people’s lives by allowing western governments the required intel to sabotage WMD programs of governments who believe in first strike attacks.

I am generally pro-situational-awareness of things in public spaces.

Fair point, its hard to quantify the number of events that didn't happen because of intervention.

"who believe in first strike attacks"

Which don't exist.

In the 1960s, high resolution images were definitely possible (and taken daily from Russian territories), but only when the plane was overhead.

This is benign compared to ubiquitous camera phones with unbelievable image resolution and location data, being uploaded for "free" to services with cutting-edge facial recognition capabilities, among others.

Only 9 years? Try 22 years. 9/11 changed the game. PATRIOT Act in the USA is still in effect. FISA courts rubber stamping any search order in the name of “national security”.

The idea of privacy disappeared a long time ago. FB/Cambridge Analytica/IG/TT/SC is just the surface.

There are plenty of places. Under a forest canopy. Underground, under a large overhang from a cliff. And those are just the types of privacy you don't create for yourself, or you don't naturally get through large structures.

They aren't necessarily convenient places, but it's not like you can't find any privacy just because there are cameras up high pointing down. It's actually probably far less intrusive and privacy destroying than the electronic device you likely carry everywhere with you that reports you location and could (if it doesn't already) keep track of when it can't report and relay it when it can again.

If you think the trade-offs of that make it worthwhile most the time, consider that there may be trade-offs here that are positive that you are discounting with what we're discussing here.

How about: walkie-talkie with decent crypto?

Suppose we design walkie talkies such that two or more units can be hooked up together in a daisy chain. While the devices are daisy chained, a button can be pressed on the head walkie talkie which will generate a random symmetric key, like 256 bit AES or whatever. This is sent over the daisy-chained bus to all the others. They flash an LED or something to indicate they accepted the key. After that they can be split up and will all use that key.

Everyone has a walkie-talkie with decent crypto in their pocket already. The problem is portable radios have limited transmission range, so you'd need some sort of a ... maybe network of towers that could pick up and re-broadcast the signal? Hmm.

Such systems exist with even more advanced key management schemes. Eg: you lose a handset, it gets removed, all other handsets have their traffic key rotated immediately over the air.

This satellite can find walkie-talkies

SBIRS is the upgrade, previously it was DSP, since the 70s.

https://missilethreat.csis.org/defsys/dsp/

What purpose does the highly elliptical orbits provide that a more circular orbit doesn't? Does it allow higher resolution when it swings in closer, and that the higher orbit is over parts of the globe of little interest?

I think it's usually the opposite. The satellite spends much more time observing interesting stuff in the higher part of the orbit and much less time zipping by uninteresting stuff in the lower part.

that does make sense. US targets of interests do tend to be northern hemisphere oriented.

Probably higher resolution over more desirable areas as well as more stable orbits.

Historically, it's for dwelling over the interesting long side and rushing past the uninteresting short side.

https://en.wikipedia.org/wiki/Molniya_orbit

Look at where the sats dwell, mostly over the North Pole. Now think about the great circle routes between the USA and the USSR.

Says a thing or two about our medias that it's never mentioned.

So every plane, train and automobile that went missing in the past 9 years did so without this system offering help?

When have trains gone missing?

It sounded nice

There is a movie called closely watched trains which is humorous but not relevant.

Well….

https://www.cbc.ca/player/play/2675415775

Perhaps they went missing when it was cloudy?

wouldn't some IR go through clouds? I thought that was the point of using IR.

That being said, the more important question is, how's the temporal resolution.

From the article: "trained to locate the sources of electromagnetic emissions with wavelengths ranging from roughly 2 meters down to 2 centimeters"

These types of signals wouldn't be obscured by clouds, presumably.

Yes. These systems are for control, not general purpose law enforcement.

Not yet

These systems are generally never used to solve low-level crimes. When they are, parallel-construction is used to avoid publicizing their existence. These systems are reserved for direct threats to national interests.

“…your scientists were so preoccupied with whether or not they could, they didn't stop to think if they should.”

I think they thought about it and decided they should.

Do you have a link to something discussion the resolutions you mentioned? The video didn't seem to mention them when I watch, unless I missed it?

At the end of this video: "We never forget who we're working for."

If you have to put this disclaimer into your work, we already have a problem.

This is why I've said from the beginning that we absolutely know who hit Nordstream. We know the boat, and we probably know the operators by correlating the boat movement with communications records and financial surveillance.

Whoever they were, they're enough on our side(i.e. the west) that we chose not to name and shame them.

My guess is a Ukraine NGO but it could have been anyone with a couple hundred thousand dollars given the equipment required to accurately plant the explosives is so common now.

Wait until someone realizes what Starlink with cameras could do.

I mean, Planet's been doing that since before Starlink had a constellation, and what it can do is "image every inch of the globe at least once each day".

https://www.planet.com/

> Incredible.

Incredibly worrying, as a member of a nation state with conflicting interests with the USA.

Pretty sophisticated surveillance technology - though it's worth remembering that you can't use it to look back in time and the footprint of each satellite is probably fairly small? I imagine that number is highly classified.

The info on the foreign buyers is pretty spicy, though it's generally known that UAE's high-level relationship with the USA is built upon recycling a good chunk of their oil money back into the military industrial complex (Wikileaks Cablegate said $19 billion/year as of 2010) - which is only one part of their overall Wall Street/London investment portfolio, but it does have a special significance.

> "HawkEye’s advisors have helped lead a large percentage of U.S. military and intelligence organizations — including the Central Intelligence Agency’s technical surveillance programs — and have included two former members of Congress who pivoted into lobbying, Norm Coleman and Lamar Smith. And so one can only conclude that HawkEye’s surveillance support for Gulf dictatorships is not an anomaly, but rather a corporate extension of official U.S. foreign policy."

UAE is also known for buying Israel's NSO Group (Pegasus etc.) surveillance tech:

https://gulfstateanalytics.com/pegasus-as-a-case-study-of-ev...

> Lockheed is now tracking phones

I get that the satellite can track the radio emissions from phones, but can they differentiate between different phones? - pick up the signal and track device by SIM IMEA/ICCID etc

It is probably because even if the signal is within a certain band the band still has a frequency legenth longer than a simple decimal point. Channels operate in a range of frequency for example. Even if a device 5g and it needs to connect to a tower it will do its best to replicate the upstream / downstream signal with little noise as possible. However the noise is the difference between the devices, they are distinct and this is how they are probably determining the device.

Also radio waves do not just go away. We are probably incredibly loud planet to any radio wave sensitive alien/creature.

Since 4g lots of devices talk on the same freq at the same time and they do math to decode separate logic streams of data - so freq in itself is not enough

I imagine it is distinct enough if you're coned into a specific place. They are probably not looking at the global as a whole but like a radio telescope pointed back at Earth.

I think you're right here. To give a hypothetical from current events, the US/NATO forces may not have direct access to information from Russian cellular networks, but a bunch of cell signals in an unusual place could indicate a troop movement or something.

Right, you could be able to see not only the number of cellphones but also the group.

I have a outdoor air quality sensor that spikes for five minutes each day at certain times. I know each day when my neighbour goes out for a cigarette. On a similar note Strava data once gave away locations of military bases.

If you aren't putting in any effort to obfuscate your data you are not doing enough. There is a reason why our VIPs and anti-censors both do it. Encryption and other privacy tools alone aren't enough. The IETF and others are putting great effort into encryption but not enough here. Recently Google announced MLS support, hopefully as that is adopted you'll see more effort with Pluggable transports and making it easy to hide data over more than TLS.

With certain kinds of antennas you can differentiate distinct emitters on the same frequency if there is enough spatial separation between them.

Just like with a single radio telescope you can simultaneously watch two closely located radio-galaxies emitting on the same frequency.

If this [1] works, why shouldn't locating mobile phones from space work?

[1] https://arstechnica.com/tech-policy/2023/03/fcc-aims-to-help...

How does this work? Don't regular Starlink dishes use like 100W or so? How does a phone have enough power to do that?

I can see text messages, maybe, like the iPhone... but real time voice too?

I think these guys have explainers on their site:

https://lynk.world/

Yes, it was not for high bandwidth applications. Nevertheless when you can receive something like SMS it seems possible to locate you phone with similar technology.

Real time voice and data downloads of about 14Mbps from LEO test satellite to unmodified phones by AST Spacemobile.

https://www.theverge.com/2023/9/19/23879527/att-cellular-sat...

Extraordinary claims require extraordinary evidence.

Can someone furnish evidence that ordinary cell phones are actually trackable... from space? When 5g is typically 200 ft, at best ~4km; LTE being ~100km at best... and the literal record for the lowest satellite orbit (Tsubame; which merely sustained it for 7d) being 2711.5km?

Yes, sure, maybe at the obscenely low signal received in space, one could pick up some kind of signature as opposed to it being even marginally usable for tx. But to infer that some kind of high-fidelity tracking could be done with that? Come on.

Even with phased arrays, you think it's not possible? I didn't check link budgets on and off target or anything, just curious.

Also, the ISS orbits at ~450km, so I'm not sure where you're getting that record lowest orbit from. Microsatellites will decay pretty quickly there, but these spy satellites are typically bigger with thrusters for orbit maintenance.

> Extraordinary claims require extraordinary evidence.

The fact that every government in the world is stumbling over themselves to become customers?

Suppose I wrote a blog post saying that the US government had invented a way to read your thoughts from a satellite — would you believe it merely because they would be a paying customer?

I can communicate via a satellite using my amateur radio walkie talkie using a few watts. Imagine what’s possible with nation state budgets.

Because that satellite is emitting a heck of a lot more electromagnetic radiation than your phone…

Ah yes, the fabled power of 250mw

https://www.amsat.org/two-way-satellites/so-50-satellite-inf...

No it is not.

Remember that 5G also uses the same frequency bands that 4G does, with a slight increase in range over 4G. 5G doesn't mean short range; 5G mmwave does.

Your altitudes for satellites are too high. Low earth orbit is defined as being below 2000km and very low orbit as being below 450 km. Starlink is at about 550km. Sources: https://en.wikipedia.org/wiki/Low_Earth_orbit https://ts2.space/en/what-is-the-altitude-at-which-starlink-...

>the literal record for the lowest satellite orbit (Tsubame; which merely sustained it for 7d) being 2711.5km

Source? Wikipedia says it was 167.4 km.[0]

Heck, even the ISS is well below 2700 km. It orbits at roughly 420 km altitude.[1]

[0] https://en.wikipedia.org/wiki/Super_Low_Altitude_Test_Satell...

[1] https://www.heavens-above.com/IssHeight.aspx

Tangential: I have argued that privacy is largely an artifact of the lack of technology. As technology increases, the total amount of information which can be kept private shrinks.

Also, the total number of criminal laws (including violations of treaties) is huge. One of the reasons we aren’t all prosecuted is because of the lack of prosecution resources and technology. As technology and automation increase, more average people will increasingly get prosecuted.

Not necessarily worried about the outcome of this Lockheed satellite system, but it is another tool which could contribute to this trend.

As long as we don't automate the court systems, that should rate limit prosecutions. On the other hand, we're already automating things like early release through machine learning scoring of likelihood of recidivism.

It's not like many US cities are really interested in prosecuting criminals in the first place.

This is a take so bad it’s borderline negligent.

If you are referring to the “progressive prosecutors”, I suggest you actually spend some time to hear long-form interviews with them. They have a genuine interest in lower crime and crime rates; they just don’t believe that the current pattern of prosecutions and warehousing criminals together in mass numbers achieves the goals.

It’s perfectly reasonable to criticize their solutions and outcomes, but to say they have no interest in prosecutions is flat wrong.

Chris Hayes interview with Chesa Boudin: https://www.msnbc.com/msnbc/amp/ncna1307198

California recently passed a law banning employees from confronting shoplifters, exempting trained security guards. So now businesses have to comply with this nonsense or be prosecuted for stopping crimes. I don't believe the state has any interest in prosecutimg crime.

https://sd15.senate.ca.gov/news/governor-signs-senator-corte...

Ordinary employees should never be the frontline defence against criminals.

That is what trained security guards are for.

That's a very blanket statement that I believe is quite wrong.

Hopefully we don't end up in a society like that.

Where minimum-wage cashiers are pitted up against violent criminals in order to protect the theft of a company's insured products.

I think the point should not be that we expect cashiers to stop criminals. Rather the issue is that the state forbids their intervention. Why should people be disallowed from protecting their private property?

They may have such an interest, but their misguided methods are absolutely destructive.

Courts don’t need to be automated. With enough laws and enough evidence, prosecutors will have sufficient leverage for plea bargains, which already make up the vast majority of criminal cases.

And the use of ML to assist bail / parole is aside from my point. Those are attempts to get a more accurate (potentially less biased) evaluation of already existing cases. I’m talking about the increase in cases.

The point being that there are a limited number of human beings, including prosecutors, that can make the system work, even plea deals are rate-limited.

Plea bargaining is a dangerous game IMO.. It incentivises a suspect to plead guilty because even if they're not they's a possibility to get a much higher sentence if the court thinks otherwise.

So people can be coerced into pleading guilty. I'm glad we don't have this system where I live.

It was considered unethical when the US was founded, but the role and ethics of public prosecutors have evolved in the past 200 years.

Also worth pointing out that local prosecutors are elected, so there are tons of perverted incentives in trying cases. When prosecutions get backlogged, voters get frustrated and will vote for change.

The 13th amendment does exclude punishment for a crime...

> I have argued that privacy is largely an artifact of the lack of technology

That's certainly true.

Think about trying to "disappear" now. 50 years ago, it wasn't so hard. Police data wasn't available online. People didn't carry devices that were constantly connected to the mobile network. CCTV cameras and facial recognition didn't exist.

Now? It would be far, far harder.

I understand every concern about privacy but 50 years ago crime rate was a lot higher in almost every country. We as humans were more violent with lack of tech or no tech. This recent 50-30? years is the most peaceful time in history by far.Tech improves our lives, and somehow makes us busy and makes us wayy less violent. Privacy is the cost here? does it worth it? I really don't know

Excessively strict, persistent and widespread monitoring technology also takes away the human "leeway" element to enforcing laws, controlling for infractions and pursuing for incrimination. I don't think most are conceived in a way that assumes absolute compliance, most people probably break multiple during their day without even realizing.

The text only mentions satellite phones, so the headline might be a bit click-baity.

I assume tracking mobile phones from space would be way harder and more expensive, although the progressive addition of satellite connectivity to recent iPhone models might help with that?

> The text only mentions satellite phones

iPhones 14+ are satellite phones to a limited degree. [1]

1. https://support.apple.com/en-us/HT213885

> "I assume tracking mobile phones from space would be way harder and more expensive"

I dunno which way I'd bet on that; on one hand any cellphone signal sent upwards is wasted energy, wasted battery - the cell towers are sideways. And cell towers are local so phones will try to use as little power as possible to get to the closest mast also to save battery, 5G picocells can be down to 100 meters. But you can't be sure to radiate sideways when phones are used at all sorts of angles, can you?

On the other hand, signal going upwards has clear line of sight and rapidly thinning atmosphere. A Google result tells me that cellphone base towers can "typically reach up to 25 miles and sometimes up to 45 miles", and this article[1] says cellular macrocells can be up to 100 miles (diameter?). Wikipedia[2] says "Mobile phones are limited to an effective isotropic radiated power (EIRP) output of 3 watts" and Reddit[3] says you can reach the International Space Station as an amateur with 5-10 Watts and a good aerial.

So ... maybe? a tuned sensitive receiver constantly listening for moments of phones doing a high power ping?

[1] https://www.emnify.com/blog/5g-small-cell

[2] https://en.wikipedia.org/wiki/Wireless_device_radiation_and_...

[3] https://old.reddit.com/r/HamRadio/comments/rvzj11/can_i_make...

Cell phones don't have directional antennas. It is impressive that they fit the anisotropic antennas. You can hold the phone in any orientation and move it around.

The directionality of phone network comes from the towers which have arrays of antennas pointing in different directions.

Also, the cell phone doesn't know where the towers are. Especially with all the smaller cells that may not know where they are.

Expensive, yes, but difficult, not as much as you think.

I'm assuming they are operating in low earth orbit. Which means that the satellite could be around 200-300km above the earth. Apart from the atmosphere its direct line of sight.

You have the advantage that you are not really trying to decode data, just figure out where something is, so you can use a large phased array antenna to electronically sweep an area. Because you're in orbit, you can localise a RF source as you fly over.

in terms of RF power, GPS is transmitted at ~40 watts from ~20200km I haven't done the maths, but as its inverse square, I'd punt that GPS signal is weaker than a phone signal at 200km

SAR is nice for satellites too.

https://www.mathworks.com/help/radar/ug/spaceborne-synthetic...

How much POWER do the GPS Satellites output on the 1575mhz L1 frequency?

In the frequency allocation filing the L1 C/A power is listed as 25.6 Watts. The Antenna gain is listed at 13 dBi. Thus, based on the frequency allocation filing, the power would be about 500 Watts (27 dBW).

Now, the free space path loss from 21000 km is about 182 dB. Take the 500 Watts (27 dBW) and subtract the free space path loss (27 - 182) and you get -155 dBW. The end of life spec is -160 dBW, which leaves a 5 dB margin.

And if you really get into it, you'll discover ALL of the following represent the same approximate signal strength for GPS on the face of the earth (m stands for milliwatts and m2 stands for meters squared):

-160 dBW, -130 dBm, -135 dBW/m2, -105 dBm/m2, -223 dBW/Hz, -163 dBW/MHz, -193 dBm/Hz, -198 dBW/m2/Hz, -138 dBW/m2/MHz

Once you figure out why they're all the same, you're well on your way to understanding power, power density, and power flux density as it relates to GPS. For those that wish to quibble, I am assuming an even distribution of power density over a 2 MHz C/A bandwidth.

http://gpsinformation.net/main/gpspower.htm

thanks for this, my maths is a bit weak and I last did this sort of calculations in 2008.

I got around -79dbm for a phone at 200km, now I think thats probably a bit high. Even so even -90dbm is certainly something thats workable with

This capability has existed at least 70 years and improved over time. See also: TRUMPET, Rhyolite, Magnum, Mercury, Orion.

A $10 MIO investment seems like nothing in that area.

Can anyone explain how this works? Are these satellites in a similar orbit to GPS satellites? Do signals from cell phones etc include timestamps? Or is there a high resolution way of detecting the direction of a signal?

Maybe just plain old triangulation? The constellation consists of 21 satellites. If multiple satellites receive the same signal, the difference in arrival can be used to precisely locate the source.

Iirc, SAR satellites in polar orbits measure Doppler effect of distress beacons to compute coarse latitude. It’s been working for decades, so I imagine recent advances in signal processing might be able to significantly improve precision.

modern DoA algs like https://en.m.wikipedia.org/wiki/MUSIC_(algorithm) can use a small phase coherent array of antennas to estimate DoA by, in grossly simplified terms, measuring minute phase differences across the array as a signal arrives: https://www.diva-portal.org/smash/get/diva2:724272/FULLTEXT0...

You'd need unique signature of cell phone signal to separate it from other cell signals

In terms of timestamp, when multiple satellites measure different times of receipt from same cell signal they could reverse compute the location

Could probably incorporate differences in received signal strength as well

At least thats my guess

I don't think you'd wanna use timestamp info from the sender, since you have no control over the accuracy of that, even if it was available

There's a new paper related to this RF signal identification technology by MIT [1]:

[1] Score-based Source Separation with Applications to Digital Communication Signals:

https://paperswithcode.com/paper/score-based-source-separati...

Hopefully aliens also have some powerful SIGINTs, so they could detect American military and exercise certain actions against the north American warmongers.

The name HawkEye reminds me somehow of the movie Eagle Eye from 2008 (which is also about tracking people).

I think the title is misleading and should be changed. Most phones don’t have satellite capabilities.

There are satellites now where 4G/5G phones can roam to them. This is just listening, not roaming. The title look correct.

This capability isn't new.

Detecting GPS spoofing would be useful in Ukraine?

It’s not possible to track ordinary cell phones this way. They’re tracking stuff that is much higher energy afaiu.

Yes, “cell phones” but like, satellite ones.

They claim to be able to track FRS ptt radios, which implies a sensitivity to .5w transmissions, and iirc a phone will put out max .2w (on 4g). Lower, yes, but not massively so, so if it really does work on FRS I wouldn't be shocked if it worked on a conventional smartphone. The phone operating in a higher frequency range than ~470mhz frs/gmrs is admittently an additional challenge, but since the thing runs up to 15ghz I wouldn't count on it being unable to track a phone

Depends on radio but FRS are now allowed up to 2W on lower/GMRS channels.

https://lynk.world/ And AST and Starlink are all working on it for consumers which is far more complex (actually pushing a 5G signal to the phone and getting a response back in orbit). Just identifying that there is a phone transmitting is far simpler.

This is just an incredibly confused comment. 5g simply does not allow for transmission distances of 200+km. Period.

They might make some kind of addition to the 5g standard or something. Because yes, indeed, there are such things as satellite phones and satellite internet.

Please consider being more technical here and instead of just believing a random website or parroting one of Elon's audacious claims (of which its hopefully become painfully clear that none can be taken as truth on face value).

I’m, no. They can get the signal. I’ve spoken with them and the issues they have is more about red shift type effects of the satellites not being geosynchronous, and not being able to alter the phones designs but meet what’s on the market.

Ok, so some shit maths:

assuming a phone transmits 0.5watts at 1m, inverse square law[1] says that it will be 0.0000000000125 watts at a distance of 200km.

Thats -79dbm, so kinda shitty wifi level of signal. GPS by contrast is -125 dbm. (logarithmic scale) Your phone can pick out a GPS signal reliably with a tiny chip antenna. Imagine having a 4 meter phased antenna array.

[1] https://www.omnicalculator.com/physics/inverse-square-law

Yes, sure; but you're leaving out issues of signal fidelity due to noise.

How much power in the carrier spectrum does the tree next to you emit? That just makes it a question of sampling length and statistics.

Reply on news.ycombinator.com