Weekly Meeting Updates 5/3/10

Jeff is working on data quality; we may need to get new data for alpha's pointing model, but Dan isn't too worried about the inevitable small corrections. Jeff thinks the biggest data quality issue at present is Matt's radar blanking project.

Matt's been hard at work trying to figure out if there's an imperfection in our software blanking code; results from running data through the software blanker have not been matching those of hardware-blanked data. While Dan suspects that maybe the software blanker just works better, Matt and Jeff are still suspicious that the software blanker may have a bug.

Matt has the latest developments: Waterfall plots indicated that the tests he had been running were not optimal for comparing the hardware-blanker to the software-blanker, but this had the unintentional side effect of proving that the software blanker was functioning properly. Matt discovered, through a conversation with Eric, that his problems detecting artificially injected data were the result of a minor error. He is optimistic for the next week of tests.

Dan and Eric discussed adding an autocorrelation test function to SETI@home. Quoth
Dan, "Jerry Harp believes that ET wants to broadcast a message that has a signpost (meaning it's easy to see) but also message content. These will both be in the same signal. Traditionally, you send a narrowband signal, and it's surrounded by a signal at lower power, wider band. People would see the narrowband and then they’d dig deeper and find the wideband. The problem is you have to decide how much power to put in the carrier and how much to put into the message. Jerry has another way: send a message, and a delayed copy of the message. That degrades the signal because you’re broadcasting two things at once. The way you search for it is in delay space, when you shift them you find out that they perfectly align. The idea is there’s something easy to find and then also something hard to find with content. It’s an easy thing to add and we could put it in there. It might add 5% more computation."

That's all for this week. Keep your eyes on the stars.