Copy
View this email in your browser
June 19, 2017
Dear friends,

I am happy to report that all of the students in the Spring 2017 SETI course did very well.  They presented the result of their work to the class on June 9 with 16 students giving 14 presentations on 12 planetary systems.  Among the most interesting candidate signals was a short pulse that we initially thought came from a single direction on the sky.  After further inspection, the same signal was identified in data recorded when the telescope was pointed in another direction, indicating that a local emitter was most likely responsible.  There were several other interesting candidate signals and students did a great job at recognizing some of the sources of interference.  Additional work is required to identify the sources of the remaining candidate signals.  To keep the data volume manageable, we analyzed only about 25% of each scan in class, which amounts to tens of terabytes of data.  We will examine the other 75% this summer to confirm the most interesting candidate signals and perhaps detect additional candidates. 
Yolanda discussed candidate signals in the Kepler-283 system during her presentation on June 9.  Here, she described the flags that we use to identify radio-frequency interference.
Micah showed time-frequency diagrams of candidate signals recorded while observing the Kepler-22 system.
This year, we used the data-processing pipeline written during the Spring 2016 SETI course, with some improvements.  Each student contributed a new piece of software that helps with the analysis of candidate signals.  One student wrote a program to extract signal properties from the MySQL database where candidate detections are stored.  Some students computed statistics or generated histograms of the properties of the 11,387 candidate signals that we identified.  Most students wrote code to quantify the properties of candidate signals, e.g., measuring the signal width in the time and frequency domains, or measuring the evolution of the signal power as a function of time.  Some students investigated the motion of GPS satellites because GPS signals appear in our data.  Because we stored all of the software on the Git collaborative platform, several students downloaded the code written by their classmates and used this software to analyze their data set.  Next year, I hope to refine and build on these analysis tools.
The Spring 2017 UCLA SETI class with donor Michael Thacher.  This year, most of the students wrote software to improve the quantitative analysis of candidate signals.
During final presentations and one-on-one debriefings, several students noted that they struggled to keep up with the range of new concepts that we cover during the short ten-week quarter.  The criticism is entirely fair.  Some students started the course with no programming experience at all.  Others started with some programming experience, but in a language other than Python.  Several students had never used the Jupyter interface, a Unix-like operating system, or the Git/GitHub environment.  Some students had to learn all of these skills in addition to concepts in radio astronomy, signal processing, and statistics.  Structured this way, the course is rather demanding.  Nevertheless, the students reported enjoying the course and feeling motivated to learn new skills.  Some of the students indicated that they plan to take additional computer science courses to perfect their programming skills.

Many students supported the suggestion of having an optional discussion section in addition to the two weekly sessions in the computer lab.  The discussions could be used to offer additional training in Unix/Linux, Python/Jupyter, Git/GitHub, etc.  It is a great idea and I will think about ways to implement it.  However, staffing the discussion section is almost certainly going to be a challenge.  We were all quite fortunate in Spring 2016 and 2017 that Adam Greenberg joined us as an unofficial teaching assistant (TA) without any additional compensation.  But Adam will be graduating soon, so we may not have a TA at all next year.  I suspect that negotiating a TA position for the SETI course will be difficult.  TA positions are scarce and they are typically reserved for the large general education courses or lower division courses.  I will continue to try to secure some sort of graduate student funding for SETI work in the hope that another graduate student could play a role similar to the role Adam played.

Some students suggested that the course ought to be taught every quarter, or that the course ought to be offered as a two-quarter sequence.  Admittedly, the current course format does not allow us to cover many interesting aspects of SETI, such as the history of the field, sociological aspects of the search, or other search techniques.  It is possible that I will be able to inject more of these topics in the coming years as other parts of the course become more familiar and as I gain a better appreciation for the time required for each module.  I am grateful for the good ideas and suggestions on how to improve the course and I will strive to implement as many of them as possible.

Warm regards,

Jean-Luc Margot

 
Copyright © 2017 UCLA SETI Group. All rights reserved.


Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list

Email Marketing Powered by Mailchimp