Jeff has been looking at splitter beam problems we’ve been experiencing as of late. The splitter is not able to process all beams because one of our two data streams is having a problem where it will miss buffers, thowing throws the blanking signals out of sync. Maybe he could figure out how many buffers were missed and then correct for it later.

Also, NTPKCr is running really slow right now.

Dave had no news for this week.

Matt and Bob have been dealing with a few database woes, specifically with our servers crashing. They are attempting to run diagnostics on our server, so if there is some down time in the next couple days Eric, Bob and Matt will just be working on getting everything back up. Two of our machines have been crashing once a week. It seems like it is most likely a hardware problem. Maybe we should think of buying hardware, or just get creative with what we have.

Tomorrow morning Matt will start updating the operating system. We’ll have a temporary webpage up while we’re down so that people will still be able to connect to SETI@home. SO no worries there!

Andrew recently got a permanent domain name for the SERENDIP server and he’s working on populating a database with candidates. In the process’s he’s had a couple of ATA drives fail. It is odd that Eric is not experiencing the same drive failure problems. In light of this, Should we consider building up the gutted machine that is laying right behind us? With a little memory and some TLC, it could be of use to us.

If you are interested in helping rebuild this machine, check out the Help Wanted Forum in the next couple of days.

Josh now has the figures of merit for his algorithms. There are five algorithms that he currently uses on the database to mark RFI. For example, one of them checks for pulses that are seen twice in the same polarization, and therefore is unlikely to be RFI. The figure of merit he defined is the percent of real pulses that make it through the algorithm over the total pulses that make it through the system. If that number is greater that one, then we can tell it is a pretty good algorithm. Most of the algorithms he has checked are between 1 and 2, but we don’t yet know the combined figure of merit for all the algorithms.

Eric’s proposal was sent in, so he is back to work as usual. He is running the current Zone RFI algorithm, which is “good but not perfect.” Right now he is only working on Gaussians (There aren’t too many Gaussians to work with so the code is running pretty fast!). Once he gets this done, he’ll put the spike and Gaussian zones in the database so Jeff can start testing. Then on to pulses, triplets, and periods, oh my!

Bob summed up his weeks in two words: “database problems.”

Last week Eric, Andrew and Dan went to hear retired Berkeley Electrical Engineering professor Dave Messerschmidtt talk about his new idea of what ET’s signal processing might look like. He suggested that perhaps we should look more closely for spread spectrum signals. He argued spread spread spectrum is very efficient, and as such it might be the method that ET would use. Maybe computation is a non-issue for extremely advanced civilization, and they would be most interested in getting the information transferred with the max information per Hz. It is much harder to look for spread spectrum computationally. Prof. Messerschimtt seems to think that ET would transfer a transcendental number, like pi. It would be fun to let him try out his algorithms and see what he comes up with. Maybe we could give him some Arecibo data that he could play around with, and maybe he’ll come up with a searchable space?

There’s a few other events that will be happening this week
There will be a talk on RFI mitigation from a visiting Australian postdoc who will explain his method for suppressing RFI. This Wednesday, at 2 pm.

The Space Sciences Lab Colloquium will be on Friday, and there will be a talk about transient experiments (which includes ET). This Friday, at 3pm.

And finally, we’re all going to go and watch the disaster-apocalypse “2012” later this week. Expect lots of cool special effects, but no sensible portrayals of science. It should be fun!

That’s all this week fro SSL! Thanks for reading!