Tuesday, December 28, 2010

Messing with the template. Again.

Longtime readers (yes, there are a few) will notice something different today; I've changed the blog template again. This is an old tradition here at Robot Guy, going back to the very first days of this blog; the stable template of the last few years has been the anomaly.

The change was forced on me by Blogrolling.com; while I had displayed rather a lot of blogrolls, many relied upon blogrolling.com to supply their innards. Alas, blogrolling.com has performed rather flakily for a while, and I just can't use it anymore. This is unfortunate, as the Space Blogroll was ensconced in blogrolling.com, and hundred of blogs out there were showing up as a big blank.

Well, Google Reader to the rescue. I had already entered as many blogs for which I could find an RSS feed as I possibly could over the course of the last several years. Now it turns out that Google Reader can produce a blogroll from the spaceblogs feed; the new Space Blogroll now resides on the left sidebar. I will probably be adding quite a few feeds to the blogroll over the next few months. I have also added the latest posts from space blogs and space news sites over on the right sidebar, also courtesy Google Reader.

Wednesday, December 22, 2010

An Inexpensive Space Infrastructure Improvement

Recently Jon Goff, Trent Waddington, and others have proposed some innovative space missions, space business strategies, and vital space technology tests that could be undertaken within the next few years. While I agree with much of what they propose, each proposal has a significant disadvantage: they all cost some entity a lot of money, more than seven digits after the dollar sign, up front.

One of the big expenses associated with any space mission is communication with the spacecraft. I think I have an idea that will bring down part of the cost of operating spacecraft communications with the ground, one that won't break the bank in up-front costs, and which will provide a boon to radio astronomy.

Currently NASA uses the Tracking and Data Relay Satellite System (TDRSS) for communications with manned missions and the Deep Space Network for communications with missions outside cislunar space. The TDRSS consists of several satellites in geosynchronous orbit and several ground stations each consisting of several large satellite dishes, and the DSN is three satellite dish complexes in California, Spain, and Australia.

It costs a lot of money to operate these big facilities. Anyone who wants to communicate with and control their satellite either has to build their own communications systems or rent the use of NASA facilities (or their Russian equivalents). As the number of launch providers and satellites launched increases, the scheduling on the TDRSS and DSN facilities will become tighter and tighter, the cost will rise, and eventually more and bigger dishes will need to be built.

However, there is another approach that takes advantage of an Army of Davids.

Currently there are about two million ham radio operators worldwide, and millions more electronics enthusiasts. There are many successful volunteer distributed processing efforts and open source collaborative software projects. And, take a bunch of widely-separated radio telescopes and combine their data using interferometry, and you end up with the equivalent of a telescope with an effective diameter equal to the largest distance separating the individual telescopes.

What if electronics/HAM/space enthusiasts worldwide were to be involved in a collaborative effort to improve communications with spacecraft? Each person involved would need a satellite dish, a way of pointing it controlled by software, and a connection to the internet. As a satellite orbits overhead and the earth turns below, multiple small radio dishes would work in concert to act as a much bigger virtual ground station. Each individual station would track a given satellite for a short time before moving to track the next one to come along, handing off communications to other stations down the line. Encrypted uplink and downlink data would travel over the internet (peer to peer?) to wherever the control center for a given satellite happened to be - a satellite could be controlled from a home office.

Satellite operators could then simply use an online satellite control and communications service. So who would be these satellite operators? Clearly the people operating billion dollar machines are going to continue to use the TDRSS and DSN for a while yet.

If we think of big expensive satellites like Hubble and Galileo as the equivalent to mainframe computers, then the "laptop computer" version of a satellite is the nanosatellite. For such small satellites, it simply doesn't make sense to have the same operational costs for communications as for the big satellites, nor does it make sense to use the big NASA facilities for such low-value targets.

However, a network of amateur ground stations running on open-source software over the internet would provide nearly worldwide coverage for these small satellites. Over time, as the network grows it would eventually become more robust (more fault-tolerant, better coverage, higher bandwidth) than the handful of big ground stations maintained by NASA and the Russians.

A second application of this satellite dish network exists. If a ground station isn't busy communicating with a satellite, it still has to point somewhere. If two such idle ground stations are separated by a thousand kilometers and pointing in the same direction, they can act as a single radio telescope a thousand kilometers across. The more widely-separated telescopes pointing in the same direction, the better the resolution. The network of small satellite dishes thus acts both as a highly-responsive very large array radio telescope and a robust satellite communication system.

Now, how will all of this be paid for? The individual dish operators would essentially be "Outernet Service Providers" (I wish I could claim credit for the clever name Outernet, but at least one company already had that idea; feel free to suggest another name). The software and electronics and so forth might be open source, but the service provider is still investing their bandwidth and electricity on a continuing basis, and they should be paid by the megabyte for their efforts. I'm thinking that the best way to do this is to set up an organization (non-profit? a business? more than one such entity?) that charges customers (astronomers or satellite operators) for data on a per-gigabyte basis and then reimburses the dish operators on a per-megabyte basis.

Such a network of ground stations would provide greater flexibility to researchers, and lower costs and improve service for satellite operators and their customers. The technology is fairly mature, the up-front cost for an individual ground station could be very low indeed (even zero), and there could be thousands of operational stations in as little as a year from now.

update December 23: Apparently the idea has some merit, because I'm not the only one to come up with it. Anonymous posted a comment over at HobbySpace about GENSO, a "software standard which allows each ground station on the network to communicate with non-local spacecraft and share data with the spacecraft controllers via the internet". This is an ESA-led project. And Trent Waddington points out the UNIFORM project being led by Japan.

Both of those projects share the idea of a distributed network of small ground stations linked by the internet for satellite data relay. I especially like the GENSO standard - at least someone has come up with the standard already, and they have 8 ground stations already operational (6 universities and two amateurs).

What I am proposing differs in scale - instead of a handful of ground stations, I'm proposing hundreds in the short term and eventually tens of thousands of ground stations. With that kind of redundancy and global coverage, one could conceivably make an entire hemisphere into a vast virtual deep space antenna or radio telescope. The availability of such a large network would lead to a dramatic drop in communications costs, extra redundancy for the Deep Space Network, and much greater flexibility in scheduling time for the big radio telescopes.

And while the GENSO and UNIFORM efforts are ESA-led and Japan-led, I'm calling for the open-source community to step up. I'm thinking that this is a job for the crowd at Hack A Day.

Thursday, December 09, 2010

Falcon 9 test launch #2

What follows is the launch of the Falcon 9 carrying the Dragon capsule, followed by the post-flight press conference.