Friday, August 31, 2012

Some Info On The Photos Curiosity Takes, Stores & Transmits

In a world where we can throw gigabytes of information around in seconds, we get a bit blasé about the technical achievement it takes to get that done. Case in point: The Mars Curiosity Rover fields 17, count em, 17 cameras. For the most part these cameras store their data in "raw" or unprocessed format (I am not going to get into raw and jpeg and what is best, let's just take it on faith for the moment that raw is big, jpeg....not so much)what NASA/JPL has to deal with however is the size of Curiosity's data stream which is huge by comparison is an amazingly small 30meg per Sol. No, you are reading that right. Curiosity can stream 30 megabytes of data a DAY. (Hell, one frame off my ancient dslr in raw is bigger than that!) So you can imagine that the rover's controllers are not snapping away like a bunch of Japanese tourists on the GW bridge. Nor are they downloading everything that the rover takes either. Why? Because that 30meg a Martian day also has to carry all the other scientific data that Curiosity generates as well.

Well you must be saying, how do they know what to download or upload....depending on you point of reference (confused yet? lol) Well Curiosity does the same trick your camera does. When you look at the pictures in your camera, you are actually looking at the "thumbnail" (a low resolution representation)of each photo. The rover just sends thumbnails from which decisions are made as to what gets sent from the rover. Another trick they use is one I have harped on for years (people that have taken my photography courses have heard this one before) Megapixels are a myth, you don't need them. Yes, I have heard the arguments, more megapixels clearer better pictures. True in part, but unless you are printing wall sized pictures, you many never see the difference between 6 and 16 megs. Most people never print a picture so anything over 1 megapixel is a waste and to prove that point Curiosity's cameras for the most part are on 2 megapixels! Oh I know...sensor size and all that, but still it is vastly smaller picture which takes up little space on the rover's memory cards. But most importantly is the kind of hoops NASA/JPL controllers have to jump through each day just to get data back and forth!

You really should read this article here about the cameras and what they do. It is fascinating material. Go here for more on the Economist


kallamis said...

An old sci-fi concept comes to mind here. Not sure of all the ramifications and tech deals, but I'm sure someone out here will know.
What about a sat relay system. Numerous sats set at different points for relaying info. We should really do this anyway throughout the system.
Not sure if that would allow for more info at a time, but it's an idea.

Beam Me Up said...

Really excellent question Kall. In truth that is already what is taking place. Curiosity is no where near strong enough to push data straight to Earth (though I am hearing hints from the amateur community that they can hear the carrier at times) so there is a complicated dance taking place with the orbital probes already flying. Adding to this network would help some, but there is a point at which no matter how many craft were utilized, there would be little if any improvement. It really is a factor of distance for the most part. Already the dishes used to capture data are huge. I would almost hazard a guess that we would have to go into high Earth orbit or one of the Lagrangian points and build rectifying systems several kilometers on a side for any practical range and data improvements. Realistically they would have to be built on this side of the data stream. The logistics alone for getting the needed material to the L points are staggering, I can not even begin to visualize the infrastructure, alone, that it would take to relocate this effort to Mars.

And I find myself slowly sliding off the point, so I will leave it there.

kallamis said...

Hmm interesting. And I sure wouldn't want the job of aligning all that to work myself.
I may be ahead of the curve here, but have we advanced at all on laser sending?
That was kind of where I was headed anyway, just kind of avoided it for lack of latest info on the subject. I know that we can use lasers for sending information over short distances, but have we ever experimented even to anyone's knowledge with using it in space.
Light is the fastest way I can think of to transmit information. But getting and keeping the stream tight enough I believe could present a problem, not to mention everything that can get in the way.
I know we can bounce one off the moon plates, but what I am talking about is a lot different, but there would be no initial atmosphere to start the beams dispersion.
I wonder if it would be possible to somehow send a message laser within a magnetic field that would disperse before the laser, and could be reconfigured at each relay before it was sent forward again.
Or maybe I'm nuts.

Beam Me Up said...

Oh my yes, the advancements in laser data have been staggering, most importantly in the density of data transferred. However one of laser's weak points is that of alignment. The higher the density the finer alignment that is required. If you add millions of miles of latency compounded with up to 20 minutes of delay and you begin to see the mountain that has to be climbed. Plus laser does not handle attenuation all that well. Plus remember the medium itself is the data carrier. With radio, you still have near light speed. Radio can take vast amounts of attenuation and still provide a usable data channel because the medium does NOT carry the data but simply provides a "carrier" for the data channels. There will be times when laser will be mandatory. One example will be if they ever get a space elevator running. In this scenario lasers are a must. Why you ask. Well let me put it to you this way. What is this called. You have a long wire that you induce a voltage in. Then you spin it in a magnetic field. What happens and when..Accually I cheated a bit, you really don't have to induce a voltage in the long wire because one will be induced anyway, but if you do it first, something different will happen.

This is why we will need lasers, but there is the reason that radio waves are still popular with NASA