Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Mars to Earth at an average of 29 kbits/s

mars.jpl.nasa.gov

Post reply on HN
Print view View original

21–30 of 88 posts

Re: Mars to Earth at an average of 29 kbits/s

#21

Post by vbtemp »

iwwr wrote: post #18
vladd wrote: post #15

I'm wondering how do they prevent non-NASA entities from sending commands to the rover - are they using some sort of cryptographic signatures when sending the commands?

You need access to the big Deep-Space Network of radio dishes.

1. You can hack into the DSN, and try to do it that way.

2. Set up an 80-meter dish in your back yard, with a team of NASA RF experts.

With step (2), the authorities will zero in on you in a second, so the emphasis is definitely on the cyber security of the ground infrastructure and not on SSL-over-space-link-communication.

Re: Mars to Earth at an average of 29 kbits/s

#23

Post by vbtemp »

fsiefken wrote: post #12

Does anyone know more about the architecture of this space radio network? For example does it use tcp/ip on top over radio or is it packet switched as with amateur packet radio?

Generally they use the CCSDS specifications, CCSDS is an international body with the space agencies of each major country contributing. Google for CCSDS, they have a bunch of open standards. Space networks really aren't quite at the level of fully mature, layered, multi-hop packetized networks... It's a Hard Problem.

Basically, TCP/IP doesn't scale to meet the constraints of deep-space networks (high latency, asymmetric data rates, intermittent connectivity). It just breaks down. That's why there's a lot of active research in DTN (delay/disruption tolerant networking and ad-hoc networking protocols). ION (Interplanetary Overlay Network) is the JPL implementation of the DTN protocols, and has been flown in a demonstration capacity on some extended-mission/end-of-life Deep Space missions. Much of it has been open sourced at this point.

Re: Mars to Earth at an average of 29 kbits/s

#24

Post by ChuckMcM »

So one wonders at what point you uplink to a satellite with ejectable flash drives which, when full, detach, make a gravity sling shot pass to leave orbit and then burn for a rendezvous with earth 9 months later. If I did the math right a 1TB drive returned from Mars in 9 months would be an effective bandwidth of 470K bits per second (or 47K bytes per second)

Re: Mars to Earth at an average of 29 kbits/s

#25

Post by jk4930 »

rwhitman wrote: post #20

I read this the other day and found it interesting: http://en.wikipedia.org/wiki/Interplanetary_Internet Apparently NASA had plans for launching an orbiter specifically as an optical communications hub for Mars but scrapped it in 2005: "As of 2005, NASA has canceled plans to launch the Mars Telecommunications Orbiter in September 2009; it had the goal of supporting future missions to Mars and would have functioned as…

Now it's ESA's turn to try it: http://www.esa.int/esaMI/ESOC/SEMM5IHWP0H_0.html

(OT: Note the picture at the bottom, the astronaut is one of ours: Austrian Space Forum, a private space R&D org.)

Re: Mars to Earth at an average of 29 kbits/s

#26

Post by vladd »

vbtemp wrote: post #21
iwwr wrote: post #18

Earlier quoted context omitted.

You need access to the big Deep-Space Network of radio dishes.

1. You can hack into the DSN, and try to do it that way. 2. Set up an 80-meter dish in your back yard, with a team of NASA RF experts. With step (2), the authorities will zero in on you in a second, so the emphasis is definitely on the cyber security of the ground infrastructure and not on SSL-over-space-link-communication.

(2): there are plenty of third world countries or foreign government agencies where placing a 80-meter dish is not a problem.

Although the process seems a bit intimidating - http://deepspace.jpl.nasa.gov/dsn/features/dsnbuilt1.html

Re: Mars to Earth at an average of 29 kbits/s

#28

Post by pothibo »

vladd wrote: post #15

I'm wondering how do they prevent non-NASA entities from sending commands to the rover - are they using some sort of cryptographic signatures when sending the commands?

This kinda raises question. I for one would not use encryption just to maximize bandwith efficiency. Deep space program are no where close to being a tactical advantage over another country. I wouldn't see why someone would like to screw it up.

On the other side, there's always a moron out there to crash the party. So even if I wouldn't use encryption, I wouldn't be surprised if they use it.

Re: Mars to Earth at an average of 29 kbits/s

#29

Post by vbtemp »

vladd wrote: post #26
vbtemp wrote: post #21

Earlier quoted context omitted.

1. You can hack into the DSN, and try to do it that way. 2. Set up an 80-meter dish in your back yard, with a team of NASA RF experts. With step (2), the authorities will zero in on you in a second, so the emphasis is definitely on the cyber security of the ground infrastructure and not on SSL-over-space-link-communication.

(2): there are plenty of third world countries or foreign government agencies where placing a 80-meter dish is not a problem. Although the process seems a bit intimidating - http://deepspace.jpl.nasa.gov/dsn/features/dsnbuilt1.html

Not just assembling the dish - you also have to know the ephemeris schedule, bit coding scheme (it changes based on circumstances), packet formats, software infrastructure... In the great scheme of things, my friend, Cuba or Zimbabwe secretly erecting an 80 meter dish just to f--k with a rover on mars isn't really a concern. Even though ESA or China or Japan might conceivably have the theoretical capability to do this... there are other things to worry about.

Re: Mars to Earth at an average of 29 kbits/s

#30

Post by zandor »

ChuckMcM wrote: post #24

So one wonders at what point you uplink to a satellite with ejectable flash drives which, when full, detach, make a gravity sling shot pass to leave orbit and then burn for a rendezvous with earth 9 months later. If I did the math right a 1TB drive returned from Mars in 9 months would be an effective bandwidth of 470K bits per second (or 47K bytes per second)

Sounds like a modern take on the Corona satellites.

http://en.wikipedia.org/wiki/Corona_(satellite)

Post reply on HN