This is the script of a radio show broadcasting soon on KFJC in Los Altos, California. It is an introduction to the constraints InterPlanetary Networking seeks to address…
Konstantin: Welcome to this, the first in an ongoing series of short Science Spots here on KFJC. We’re going to be bringing you short little nuggets of science news, with an emphasis on space and astronomy. I’m your host for Science Spots, Konstantin Kalaitzidis.
This week, we our topic is critical to space exploration, but may not be very familiar to many of our listeners: the development of an inter-planetary network for space communications. Our guest this week is Mike Snell from the San Francisco Bay Area Chapter of the Internet Society. Welcome to the show, Mike.
Mike: Thanks Konstantin. I’m very happy to be here.
Konstantin: So tell me, Mike, why is the Internet Society interested in space communications?
Mike: The Internet Society is chartered with advancing the internet. Most people think that’s primarily in the public policy arena where they’ve been very active in fighting efforts to censor the internet, but they’re also very active in furthering the technology that powers the internet, as witnessed by their key role in driving the adoption of the next generation of Internet Protocol addressing, IPv6. InterPlanetary Networking is simply a way of making data communication more reliable and secure in the constrained environment of outer space. New sets of protocols have been developed over the last fifteen years or so that can make networking in space a lot more effective. In fact, Disruption & Delay Tolerant Networking protocols are currently deployed on the Internatoinal Space Station and handle all earth to ISS communications for not only NASA, but also the Japanese and German Space agencies deployed there.
Konstantin: Wait a minute—why can’t we just use the internet as it is in outer space?
Mike: The internet we use has been wildly successful. Most of us rely upon it daily to work, to study, to entertain ourselves. But it’s easy to forget that the internet was based upon some assumptions, and it only works well when those assumptions are valid.
Konstantin: And those assumptions would be?
Mike: The core sets of rules that make the internet work: the Internet Protocol (or IP), which lays down the directions for how packets of data are sent from location to location on the network, and, more importantly—the Transport Control Protocol (or TCP), which insures that your data actually gets to its intended destination, and provides ways for errors to be dectected and packets to be resent, if necessary—assumes that a pretty benign environment exists in order for everything to work properly. Because these are the most important protocols deployed on the internet, you often hear it referred to as a TCP/IP network. The most important assumptions are:
1) That end-to-end connectivity exists between devices attempting to communicate with each other. We’ve all experienced what can happen when that assumption gets violated. Ever experienced a dropped cell phone call? Something very similar happens in a TCP/IP network when that end-to-end connectivity gets broken. Just like a dropped cell phone call, you have to establish the entire communication session again. Redial the number, and try to figure out where you left off in the conversation.
2) TCP/IP also assumes relatively fast round-trip times (RTT) between the sending and receiving hosts on the network. This is also called round-trip-delay. The term “host” here can be almost anything: your desktop Mac/PC; Google’s website, a tablet, a smartphone—or even a temperature sensor in a modern building. Because the speed of light is so pickin’ fast, this is rarely a problem for most internet applications we use here on earth—unless—we are trying to use applications that are sensitive to delay, like voice, or video.
Most of us have experienced occasionally frustrating or even humorous voice or video problems when trying to use services like Skype or Google Hangouts to communicate. Because the data must be transmitted live, you suddenly become very aware of signal delay causing chopped up video or garbled audio.
Streaming services like Netflix try to overcome this problem by storing a reserve of data locally on your device before playing the video. Network congestion can cause delay—so you may have noticed instances when even this reserve of video data is not enough to deal with signal delay. Your Netflix video pauses while your buffer of data gets rebuilt enough to continue streaming video to you.
Think about outer space and the distances between the planets. At these distances the speed of light becomes downright pokey. Round-Trip-Times between planets can become so long that TCP/IP literally breaks. The RTT for network traffic between earth and Mars, for instance, varies between 7 minutes, when they are closest together as they orbit the Sun—to as much as 46 minutes when they are furthest apart.
There are other, more arcane issues like bit error rates and symmetry of outgoing and inbound bandwidth—but just looking at these two primary issues of continuous end-to-end connectivity and relatively low signal delay—can demonstrate how significant the challenges are that need to be overcome in order to make interplanetary networking successful.
Konstantin: Thanks a lot, Mike for that introduction to InterPlanetary Networking. Next show, we’ll be looking at the basic principles behind InterPlanetary Networking, and how the Mars rover program solved a really serious problem by employing those basic principles. In the meantime, you can find out more about InterPlanetary Networking on their website: www.ipnsig.org.