I knew Vint Cerf was working on bringing the Internet to space. Universe Today, Google and NASA are Working on an Interplanetary Internet, has some info about it, though I wouldn't call it details.
"The Interplanetary Internet project is primarily about developing a set of communication standards and technical specifications to support rich networking in space environments,' Cerf added. This all sounds very interesting, but the challenges with building such a system require some novel techniques. How do you deal with the limitation of the speed of light? After all, it can take light 40 minutes to travel to-and-from Mars, and up to 12 hours to Pluto and back. How do you cater for planetary rotation? The transmitters/receivers won't always be on the correct side of the planet. What happens if a satellite signal is blocked by a planet, the Sun or a moon? Vint Cerf says the disruption of data transmission has to be confronted with a delay- and disruption-tolerant networking system, otherwise known as DTN."
In a semi-related story, apparently we're getting really good at satellite communications. Physorg reports Satellites approach the Shannon limit. "Satellites are achieving unparalleled efficiency with a new protocol, DVB-S2. The performance of DVB-S2 satellite systems is very close to the theoretical maximum, defined by the Shannon Limit. That efficiency could be pushed even further by network optimisation tools and equipment recently developed by European researchers."
1 comment:
We've been "very close" to Shannon for a while now - arguably since Turbo codes become practical at least a decade ago. In fact it can be easily exceeded these days by using time-space coding techniques (e.g. MIMO). The largest constraint on what we can do is the limits of processing power. Of course I could be biased in that view.
Back when I was working (indirectly) for NASA/JPL there was development on a protocol named "SCPS" which was intended to help mitigate some of the problems traditional internet protocols have with space based links. This was aimed at communication via orbital satellite networks. Of course, as you move further out, the problems get worse. In addition to latency increasing, bandwidth goes down (more noise and less power). Some current research on quantum coupling may provide an alternative that gets around the whole speed of light problem.
Post a Comment