 |
||||||||||
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
|
 |
|
|
|
|
|
|
|
|
|
|
SETI
- Allen Telescope Array
- Optical SETI
- Interstellar Message Construction
- Project Phoenix 1995-2004
- Background Science
- History of SETI
- Center for SETI Research Team
- Other people doing SETI
- Hollywood's View of SETI
Technology and Time
by Dr. Peter Backus, Observing Programs Manager
We had a visitor the other night, a Silicon Valley engineer. We talked about the science and the technology behind SETI, and how things have changed over the years. The system we developed for the NASA Targeted Search and later, Project Phoenix, filled nine equipment racks. Our New Search System processes nearly three times as many channels in less than half the rack space.
To show our visitor how much things have changed, I retrieved a circuit board from the old system. This impressive looking artifact is the most complex board from that system. In its day, it performed Continuous Wave signal detection on 14 million channels in one polarization. At about 16 inches (40 cm) on a side, the board is unusually heavy, and filled with components on both sides. Several rows of custom built memory modules stick out from the surface and on one side of the board, a daughter board roughly half the size of its mother rides above connections at the top and bottom of the main board. Rows of memory modules fill this smaller board.
While we were admiring the complexity of the old board, Ben Sanchez brought in the equivalent component from the new system, a board that he helped design and build. What a contrast! The board for the new system fits in a standard PC, and has only a few components, dominated by a single Xilinx Field Programmable Gate Array, its workhorse. While it may not look nearly as impressive as the board which it replaces, appearances can be deceiving. Our new board can process eight million channels in two polarizations, and perform more signal processing on those 16 million channels than the old board could on 14!
Technology has certainly progressed as seen in the two generations of circuit boards. But that progress continues, and our next search system will use very few (if any) custom circuit boards. The next system, SETI on the Allen Telescope Array or SonATA, will do almost all processing in software. We can apply our engineering resources to expanding the search in bandwidth, signal types, or other areas. All of this is thanks to Moores Law of technology and time.