Guest EditorialCopyright © 1998 by Noel Cedric WelsteadThere has been much recent discussion on the SETI email lists about a new class of microwave receiver, which resides on a plug-in card and disappears into your computer. Such receivers (of which the WinRadio manufactured in Australia is a prime example) are nearly ideal for a host of applications. Unfortunately, SETI is not one of them. The problem has to do with the high levels of interference generated internal to modern, high-speed computers. Although shielding on computer-card receivers may be adequate to reject most such interference, SETI is about detecting impossibly weak signals -- the sorts that computers generate in abundance.
SETI League Eastern Australia Regional Coordinator
email nwelstead @ seti.org.au
As an example of the sorts of spurious signals which plague SETI researchers (amateur and professional alike), see the images on the SETI League "What We've Heard So Far" page. Three of these displays show a very exiting "hit" which turned out to be some spurious signals from a computer I had on the work bench. In particular, the signals were coming from a SCSI hard disk controller card that was exposed from the case shielding. My main reason in expounding all the above, is to indicate that SETI is based on the detection of very weak signals that have been integrated out of the noise. Using FFT methods we can pull these almost "non existent" signals out from the background noise.
I believe that a receiver like the Win Radio systems would be a great device for the Amateur radio / computer enthusiast who wants to "scan" around a bit, but forget it for SETI use. The fact that the radio is surrounded by the computer (a really RFI noisy piece of equipment) means that it will get all sorts of low level signals from the computer hardware. These signal even though at a very low level will accumulate into "hits" if you integrate the received signal enough times. In other words, a low noise environment is very important. When you look at how much gain we are using at the front end looking for those elusive signals, you can pick up all sorts of strange apparitions from all sorts of places. Including your own equipment.
The traces on my FFTDSP came from a very weak signal emanating from an oscillator on the SCSI controller. It was a harmonic from the low frequency crystal Oscillator that clocked the data registers from the hard disk drive, very weak but there. I have seen these "wigglers" before. Shown to me by Jill Tarter from the SETI Institute. They were probably the same sort of signal that has plagued reseachers with super sensitive receiving equipment..
The scientists at the Parkes Radio Observatory constantly complain about the RFI they get from their on-site computers. If they get it so can we.
My advice to the ARGUS station builders is use equipment that is well shielded from the outside world. Use the minimum number of system components as possible. As other SETI League members have discovered, it's better to down convert using one stage of conversion, not two.
Use lots of shielding to reduce those false hits, they only cause stress when they occur. I should know , we have had a few signals that have almost defied our efforts to find out where they came from. Only by carefully searching the local environment did we eventually track them down. None of them ETI's (unfortunately). But you never know!
Editor's Note: Several of the companies which produce receivers-on-a-card also manufacture well shielded external receivers, which merely plug in to the host computer via a serial, parallel, or SCSI interface. Rosetta Labs, for example, now offers its top-of-the-line WinRadio under two model numbers: WR-3100i (for Internal) and WR-3100e (for external). Such external receivers should be less prone to the type of computer RFI which Mr. Welstead describes than their internal brethren.
 email the Webmaster | entire website copyright © The SETI League, Inc. this page last updated 4 January 2003 |
 Top of Page |