Joined: 13 Feb 99
Since I last wrote, progress has been slow. This is partly because I took a few days off (hiking, of course) but also because working on birdie-related stuff is inherently slow since birdies are added at the beginning of the pipeline, so it takes a day or two to see the results of any change.
There were some bugs that caused too few signals to get generated for birdies; most had none at all. I fixed these and also increased the average power of birdies a bit, to get more signals.
The results are here: https://setiathome.berkeley.edu/nebula/birdies.php
I'd say they're encouraging. About half the top 100 multiplets are birdies, and lots of the top 1000. So if there's an ET signal like the birdies, we'd be able to find it with a reasonable amount of manually sifting through RFI. (Caveat: in these results we scored 100K pixels; when we score all 16M the birdies will be sparser).
This is good; it provides some objective confirmation that our RFI and scoring algorithms are mostly working. Much remains to be done, especially in scoring. Score is log(probability), and there shouldn't be scores of -100K. I may add birdie Gaussian signals.
I also add an important refinement to the scoring algorithm; I'll describe that in another post.
©2018 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.