Hacker News new | past | comments | ask | show | jobs | submit login

I'm amazed that they're able to disable a voltage regulator remotely. Did they just build in relays for every single component in case they wanted to rewire everything on the fly or was this something they specifically knew they might want to do 45 years ago?



Also I assume there isn't a web service running there with a "enable/disable" endpoint to control stuff. Wonder what the control stack looks like.

Turned up this from https://voyager.gsfc.nasa.gov/Library/DeepCommo_Chapter3--14...:

> The signal consists of 16-bps, Manchesterencoded commands, biphase modulated onto a squarewave subcarrier frequency of 512 hertz (Hz).

There are references to fuller descriptions of the command set but I haven't found them yet.


I can't help but wonder how possible it would be (or would have been at some point in the past) for a person with evil intent to send unauthorized commands to these probes. How are signals authenticated? At this point, is it just that there are only a handful of parties capable of sending a strong enough radio signal to a precise-enough location in space, and who know the position of these probes to the degree of accuracy required?


Well, I think the first obstacle would be acquiring a sufficiently large antenna. Given that the large dishes used in the DSN are 70m (~230ft) across, I suspect somebody would notice…


They'll notice! Good thing you're doing it in North Korea where they can't do anything about it.


I buy that. But what about shortly after launch?


Your guess is as good as mine I suppose. Not many people have access to a huge antenna with huge transmitting power.

As for authentication: with 16 buts per second and 44 hours round trip, they're not doing some sort of handshake to agree an encryption key. But given these were launched during the Cold War, maybe some basic encryption perhaps, with key of course hardcoded before launch?


I highly doubt it's even authenticated. They're probably just relying on anyone with enough skill and money having easier/better things to attack.

The death of Voyager would not hurt any of America's strategic objectives. It would just make some nerds really bummed, approximately as many in your own country as in America, so what have you gained?


Just getting a piece of infrastructure that can lock onto the probe's telemetry signals is incredibly difficult.


You'll need to hack NASA's deep space network first I think. On your own and with a normal mortal's budget you're not going to be able to do much of anything.


Nowadays sure. What about shortly after launch?

Say you’re the USSR and want to quietly piss in NASA’s cornflakes. Or hell, say you’re Putin—today—and want to do the same thing. Could you?


Shortly after launch it was almost certainly not known how communication worked. And if you had a spy that could get you that info, you could just get them to press the wrong button(s) instead.


Putin has limited resources. He'd rather spend them on his stupid war, and also building other weapons and warships, rather than a giant radio antenna and a team of engineers trying to hack a 45-year-old spacecraft's communications protocol, just to screw with NASA.


I’m asking about the technical measures used to authenticate commands to these probes. The bit about Putin was just about framing the question.


Some random hobbyists (or even small nation-states) aren't going to be able to build a radio antenna large enough to communicate with deep-space probes like that. That's all the "technical measures" needed.


Got a Time Machine?


Yeah, I'd say it's actually pretty common to make spacecraft avionics as remotely switchable as possible. For one thing, there's likely to be a considerable amount of redundancy so that a single component failure doesn't ruin your whole mission. It's good practice to have a way to isolate any component that is anticipated to fail. FDIR is a known acronym: fault detection, isolation and recovery.


This could be as simple as disconnecting a zener diode from ground, but yeah, I'd be a lot interested in knowing more technical details.


They probably implemented fail-safe relays with circuitry to disengage automatically under hazardous conditions but also controlled remotely.


Exactly, and to have the foresight so long ago to architect it this way, brilliant.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: