Is there intelligent life somewhere within our universe? How would we find out? We could go looking, if we had spacecraft capable of at least faster than light speeds. Or we could wait for them to come and visit us, if they have such craft. But in spite of all the media hype about Roswell, N.M., area 51, or little green men in flying saucers, ET has not come calling.

We could look for evidence, which we are now doing using the ground observatories as well as the Hubbell Space Telescope. To date, we have discovered evidence of giant gas planets the size of Jupiter orbiting distant stars in our galaxy. But the possibility that we could see a planet capable of sustaining life - as we know it - is extremely remote. The distances are too vast, and our telescopes, even the Hubble, are just not good enough.

Our best possibility is to listen. We are constantly sending signals into outer space. Some are deliberate, pulsed signals designed to tell a distant civilization that there is something here. The rest is leakage - radio, television, radar and microwave. These signals leave earth in all directions, steadily growing weaker but, in theory at least, capable of infinite travel. The most powerful signals, radar and microwave, have only been traveling for some 50 years, so these can only be some 50 light-years out. The assumption is that if there is another intelligent civilization somewhere in the universe, they will also be sending signals, either accidentally or for a purpose, and maybe we can detect these signals. Even if such a civilization is very far away, if it is also much older than we are, then signals could be reaching us now. So we listen, not knowing what kind of signals to expect or where they will come from.

If you point a radio telescope at the sky, you can hear all kinds of signals. Some come from our galaxy. Others come from the atmosphere. What we are hearing is static, very noisy at low frequencies because of the galaxy and also at higher frequencies due to atmospheric noise. Between the two noisy regions is a relatively quiet area, from about 1 GHz (Gigahertz, or billion vibrations per second) to 10 GHz. This part of the radio spectrum is just above the frequencies used for pagers and many wireless phones. But a 10 GHZ span is a very wide frequency range to explore.

Nature has given us a hint about a smaller frequency range we should monitor. The simplest stuff of the universe, neutral hydrogen gas, emits radio signals at 1.42 GHz. Another molecule in interstellar space, the hydroxyl, or OH, emits at about 1.64 GHz. These two combined make up the compound of water (HOH, or more commonly H2O). The frequency range between these two emissions, from 1.42 to 1.64 GHz, is a quiet region of the spectrum called the “water hole.” Life as we know it requires water to exist. Where would we expect water-based intelligent civilizations to meet? Around the water hole, of course. Out of the millions and millions of frequencies possible, this provides a nicely limited range of frequencies to start the search.

The leading search for intelligence has been the University of California at Berkeley program named SERENDIP (Search for Extraterrestrial Radio Emissions from Nearby Developed Intelligent Populations). SERENDIP has been in operation for 19 years, beginning with SERENDIP I in 1979. That initial program used a 100-channel spectrum analyzer at the UC Berkeley Hat Creek Observatory. Since that time, the program has undergone a series of sequential improvements. In 1992 it was moved to the 1,000-foot dish at Arecibo, Puerto Rico, the largest radio telescope in the world. Currently, SERENDIP IV examines 168 million channels every 1.7 seconds in a 100 megahertz band centered at 1.42 GHz.

Collecting this much data poses a horrendous analysis problem. There are billions and billions of data points requiring complicated calculations to extract a possible significant signal out of the cloud of background noise. What was needed was the use of the best supercomputer to analyze the data. This was not available, nor could the SERENDIP project afford the tremendous cost. In 1996, two scientists with the project conceived the idea of using the Internet as a supercomputer to process the data. Thousands of small computers, each processing one small piece of the data, could achieve the same results as a very large supercomputer. Where would they find thousands of small computers? Easy. At my house, or yours or thousands of ordinary people who have home computers. Much of the time, these computers have flying toasters or some other screen saver program running while they are not processing some specific task. With careful planning, these systems could be enlisted into the world’s largest distributed network.

Thus was born the SETI@ Home project. The Berkeley labs created the software, which is free to anyone who wants to participate in the project. The developers hoped to enlist 100,000 home users in the project. At latest count, there are 2,766,375 from all over the world. The numbers of users by country ranges from 1,246,342 in the U.S. to 8 in Liberia.

I am currently running SETI@Home. Becoming involved is simple. Log on to www.setiathome.ssl.berkeley.edu and click on “download software.” You will be asked what platform you are using (Windows 95/98, UNIX, Linux, etc.) and the programs will be downloaded. It takes a maximum of 5 minutes. Remember to write down the name of the download file and where it is located. The download process will explain this. Double-click on the file to start the extraction and setup. When you are ready to register, you will be asked for a nickname or “handle” to identify your work, your email address, and whether this computer is at home, work or school. When registration is complete, the first “work unit” will be downloaded. This is a 350K file which takes 1-2 minutes to download. Everything is automatic from this point on. The program runs in place of your usual screen saver. The graphics are very nice, and, if you like, you can follow the progress of the analysis.

Each work unit is a 107.4-second block of data. Your computer performs mathematical computations called fast fourier analyzes the data, looking for particular types of strong signals at various combinations of frequency and bandwidth. To process this work unit will require between 2.4 and 3.8 trillion calculations. How long this takes depends on the computer system being used. Some of the really hot systems at the major computer labs will do this in as little as 14 minutes. The average for all systems on the network is about 21 hours. However long it takes, it does not interfere at all with my normal computer use since it runs only when I am not doing anything else. When a work unit is completed, the program will notify you. At your command, it will make the connection to Berkeley, upload the results, and download a new work unit.

What if the work unit I am processing contains signals from another civilization? Unless I happen to be looking at the screen while a significant signal was being processed, I would not know it. But when the results are uploaded to Berkeley, that work unit would immediately be sent to another computer for verification. Only after two or more analysis report the same results would that block of data be a candidate for even further more detailed analyses. Should it finally be determined that it was a valid signal, I would be credited as a “co-discoverer.” But since there have been 82,180,408 work units processed to date without any exciting results, I’m not going to hold my breath.

The world’s largest supercomputer, the IBM ASCII White, is rated at 12 terraflops (trillion floating point operations per second). SETI@ Home, with all 2,766,375 units in operation, is rated at 15 terraflops. So I am proud to say that I am a very tiny part of the world’s largest and fastest supercomputer network. And who knows, ET might really be calling us.

(Manning is a retired instructor in microcomputer systems technology at Haywood Community College. He can be reached at amanning@asap-com.com)

 

Back to Top
The Smoky Mountain News