Most SETI programs scan the sky looking for strong signals. Any signals that are deemed interesting are put on a list for follow-up observations weeks, months--even years later. Long delays in verification of potential ET signals sometimes generate tantalizing, but ultimately frustrating, stories. The most famous is the "Wow Signal" detected at the Ohio State University Observatory in 1977. The signal was strong and had many of the characteristics that would be expected from a real ET signal. Over the quarter century since the detection, multiple attempts with greater sensitivity have found nothing. Yet for some people, the mystery remains.

The SETI program developed as NASAs Targeted Search, and carried on by the SETI Institute as Project Phoenix, is designed to eliminate such mysterious signals. This program observes Sun-like stars for relatively long periods of time at a wide range of frequencies and for a variety of signal types. The result is the most sensitive and comprehensive search ever conducted. Engineers and scientists developed a system of computers that collect and analyze data, and then immediately test any potential ET signals. This entire process is highly automated and choreographed by sophisticated control software. Human experience and judgment have been encoded in software to make rapid decisions about the signals detected, because signals are detected.

Since the technological, radio-communicating civilizations we seek could be many light years away, it is a safe bet that we will detect our own, terrestrial, radio signals. The challenge is to quickly determine whether a signal is coming from the direction of the star were observing or from somewhere near the Earth.

To determine whether a signal is coming from a particular star, we exploit a rhythm of our solar system, the daily rotation of the Earth. Our primary telescope, Arecibo Observatory, is at latitude 18 degrees and so moves at a speed of about 987 miles per hour (1579 kph). This motion causes a changing but very predictable Doppler shift in the frequency of a signal that would be coming from a star. Another observatory would have a different Doppler shift relative to the star. In our project, we use the Jodrell Bank Observatory in England. At latitude 53 degrees, JBOs telescope moves at about 624 miles per hours (999 kph). The relative velocities between the telescopes and the star change, and the rates of change are different at the two telescopes. This is a very predictable result of the Earths daily dance, and it causes a signature frequency waltz in an ET signal.

So far, the signals weve detected are of the "two left feet" varietynone has yet displayed the cosmic Doppler dance of a genuine ET signal. As we complete Project Phoenix by mid 2004 and then move to the Allen Telescope Array, our searches will continue to process data in real time. The beat goes on.