Quick Links
Calendar
| Sun | Mon | Tue | Wed | Thu | Fri | Sat |
Categories
Archives
XML/RSS Feed
Statistics
Total entries in this blog:
Total entries in this category:
Published On: Aug 05, 2007 08:38 AM
Total entries in this category:
Published On: Aug 05, 2007 08:38 AM
What I've been doing
Data analysis, more single-father mode
living, and a cold.
I've simply been too busy with work and too tired
at home to do any blogging in the past week, so rather than make individual
entries and backdate them, I'll just post an entirely new entry summarizing what
went on this past week.
I work on the Low Energy Telescope (LET) on the STEREO spacecraft . The LET Instrument is a solid state detector telescope designed to measure elemental abundances in solar energetic particles at low energies (~1-70 MeV/nuc). The instrument contains in its onboard memory three response matrices. Each matrix corresponds to one of three particle penetration ranges in the instrument, and each matrix maps the response of two detector layers to the passage of a charged particle through the detectors as a delta-E vs. E' response. From the location of a particle's signals on a matrix, the instrument can identify the particle's charge and assume a mass. Energy is then calculated from the sum of energies deposited in the instrument. In principle, a particle's charge can be determined directly from the detector signals (assuming they're properly calibrated) and a given range-energy relationship, but doing so would take a lot of CPU cycles by the LET onboard processors, which can't even do floating point calculations natively anyway. (Calculations onboard are done in log-space anyway, using integer-rounded log tables.) The matrices take up a lot of memory, but using them takes up very little CPU.
I generated the response matrices used by the LET instruments. The matrices currently onboard the spacecraft were generated prior to launch using range-energy calculations and extensive test data. However, now that the STEREO spacecraft are in space, the environments that the instruments are in are significantly different than prior to launch, necessitating fine-tuning and recalibration. By early last week, Rick L., one of my Caltech colleagues, had finished calculating new gains and offsets for the solid state detectors using in-flight ADC STIM data (calibration pulses), and with these new gains and offsets, I could see how the matrices might need to be adjusted.
The following screenshot shows the kind of thing I've been working on since last week:
The window at left shows a graphical representation of one of the LET matrices (the colored bands), overlaid with particle data (the black dots) from a solar energetic particle event from December 2006. The matrices themselves are 128x400 arrays of integers, but they're usefully represented by the graphical figures, which I distribute to my colleagues for review. The y-axis is the delta-E measurement for a given range (energy deposit in the "first" detector), and the x-axis is the E' measurement for the range (residual, or remaining, energy deposited in the "second" detector). Energy deposits are governed (or approximated) by the Bethe-Bloch equation, which itself can be simplified to range-energy relationships in silicon. The result is that particles of various energies, charges, and masses lie on bands or tracks on the delta-E vs. E' plot, ranging in this case from protons at lower left (yellow band) to iron (green band at top) and nickel (yellow band above the iron band). The LET instrument identifies protons, He-3, He-4, C, N, O, Ne, Na, Mg, Al, Si, S, Ar, Ca, Fe, and Ni, shown as the alternating yellow, red, and green bands. The various shades of blue are background bands, and the uncolored (white) region at lower right shows "backward-going" particles or other noise. The orange rectangles are "livetime STIM" boxes for calibration purposes.
As colorful as the figure is, however, it's merely a graphical representation of a large matrix that itself is generated by range-energy curves (physics), tweaks to those curves to match actual data, and a lot of IDL code that I wrote to calculate those curves and to "paint" the matrices according to those curves. A tiny snippet of that code is in the window at right.
That said, all of my work at the office for the past couple of weeks has been to take Rick's most recent gains and offsets and particle data from some largish events that STEREO has detected recently, and to tweak the matrices further. Tweaks usually involve changing numbers on the multitude of tables I have which define the curves on the matrices. I've gone through dozens of private revisions alone in my office, and last week and this week, I released plots to my colleagues showing the new matrices and how the data lie atop the matrices. Those plots would result in additional requests for matrix modifications, followed by further tweaking alone in my office, and some of the requested modifications have resulted in actual changes to my code, rather than changes to tables. I also had to tweak my code -- both the matrix generation code as well as the data analysis code -- so that the turnaround time between tweaking the matrices, viewing them with data, and re-tweaking them would be shorter.
I released a third public revision yesterday (Friday), just before leaving work for the long weekend. Hopefully, this will be the last revision. If so, I'll convert the matrix file to a base64 representation of the file and submit it to my colleagues Andrew and Rick C., who will be able to include it with the next major firmware upload to the spacecraft. Once that's done, we may soon declare the LET instruments "commissioned" for the flight.
-----
The STEREO work has taken up most of my time since the last entry in this photoblog. Generally, since William was born, I don't like to bring work home, so I can keep my home life strictly devoted to family. There have been occasions when I've had to work at home, however, and this matrix work did occupy a few evenings.
I'm usually stricter about no-office-work on weekends than I am on weeknights. Last Saturday, while I was with James at My Gym, Hsuan and William went shopping at Target for some Valentine's day items for William to bring to daycare on Valentine's Day. While there, Hsuan found out that Target would be opening up a shipment of Nintendo Wii's the following Sunday morning, and they would start handing out numbers to anyone in line at 7:00 AM that Sunday morning to get one of the 22 (or was it 32?) Wii's in the shipment. Hsuan knew I've been casually interested in the new generation of gaming consoles. (I haven't played console games since the early 1980's, though I've played computer games often since my undergraduate days.)
Hsuan planned to buy me a Wii in secret, as a Valentine's Day gift for me, but William blabbed the secret to me on the way home. Hsuan and I talked about it. I wasn't sure how much interest there remained in the Wii and whether people would be camped out overnight in front of Target. If so, I didn't think it was worth her effort, but if she got there at 5:30 AM without a big line, I thought it might be worthwhile. When she got there at 5:30 AM or so, she was number 2 in line, and by the time the store opened, I think there were still 7 units unclaimed. So, there's strong interest still, but it's not impossible to get one any more.
After 9:30 AM Mass on Sunday morning, I started setting up the Wii, with William hovering around being impatiently patient. I wanted to route the audio and video cables from the Wii through my Harman Kardon home theater amplifier to the plasma TV, but after some struggling, I concluded that none of the RCA video inputs on the amplifier work any longer. I was disappointed; I will eventually have to upgrade the amplifier to a newer model, though I'll likely wait until we upgrade our cable and DVD player to high definition versions.
Meanwhile, I ended up hooking the Wii directly to the plasma TV via the TV's second A/V inputs, and from then on, everything worked splendidly. We -- mostly William and I -- spent most of Sunday playing the included Wii sports games -- boxing, tennis, baseball, bowling, and golf. I worry that the Wii's infrared sensor, which I taped to the top of the TV, may not detect me very well, given that the television is somewhat recessed in the home theater cabinetry, but it seems to detect William better, probably because he's shorter and more thoroughly in view of the sensor.
One result of our first day of play is that I spent the first half of this week with a sore elbow and back. I play tennis, baseball, and boxing rather vigorously.
------
Last weekend was cold and wet, so I didn't go out for my weekend 2-mile walk with Bessie. In fact, I haven't exercised at all since the week before.
------
Oh, and for Valentine's Day, I ended up buying Hsuan a dozen long-stemmed roses. These were the higher-quality, thicker-stemmed, very expensive kind of roses that the flower shop doesn't sell at "special" lower rates, but they should last a lot longer than the cheaper kinds. I gave William and James each a rose to give to Mommy.
However, we gave her the roses on the 13th, because Hsuan had to leave on Wednesday for another business trip that lasted through Thursday afternoon. That left me back in single father mode Wednesday night and Thursday morning. I brought William to his My Gym class on Wednesday evening, and afterward, I passed by McDonalds again for the boys' dinner. After dinner, bathing the boys, and putting them to bed, I did a little house work and then went to bed myself, bringing the boys to bed with me to save me the trouble of having to get up in the night if they woke up. Of course, I spent the night trying not to be kicked in the face by either boy on either side, as well as trying not to roll over them...
The photo is of me with James on the family room couch, before I went to bed, and after he woke up during one of his usual night-time awakenings. I brought him out of the boys' bedroom so he wouldn't wake up William.
Thursday morning, I had my hands full with the boys, including my daily chauffeur job of bringing them to daycare, that I had to skip my usual Thursday morning STEREO conference call. At this point, though, I had already submitted another matrix revision via e-mail, so I had made my contribution for the morning.
-----
This Thursday evening, I had my monthly Board meeting for the boys' daycare center. Since we were already past the halfway point in our fiscal year, I decided I'd do a half-year budget and finance assessment, to see how well we were meeting our budget goals. I found we're behind expectations for our income, but we more than make up for that shortfall by keeping our payroll and payroll-related expenses in line. I wrote up a report and presented it to the rest of the board. I expect to begin outlining a budget for next year before April.
-----
To top it all off, I've had a head cold for the past few days, with a lot of sinus congestion and headaches. I'm popping vitamin C like it's candy.
-----
Not specific to this week in particular, but Hsuan's snoring has gotten worse in recent months (at least when she doesn't use an anti-snoring throat spray), to the point that I've gradually given up trying to get James to sleep through the night by weaning him off of having her in the boys' bedroom. She gets that duty almost every night after I go to sleep because it's dangerous for me to drive the boys through rush hour traffic to daycare if I'm falling asleep at the wheel. On the other hand, she usually goes to sleep before I do, and given that I don't go to sleep before 2 AM on most nights, I take care of James' wake-ups while she sleeps and while I'm still awake.
Some nights, after I finally go to bed, I just pray for James to wake up crying so I can get some sleep alone in our bed...
-----
Oh, finally, we decided not to apply to the big Pasadena parochial school. We figure the odds are too long, and we've already got the regular private school applications done.
And, William has his yellow belt test next weekend. We think he's ready.
I work on the Low Energy Telescope (LET) on the STEREO spacecraft . The LET Instrument is a solid state detector telescope designed to measure elemental abundances in solar energetic particles at low energies (~1-70 MeV/nuc). The instrument contains in its onboard memory three response matrices. Each matrix corresponds to one of three particle penetration ranges in the instrument, and each matrix maps the response of two detector layers to the passage of a charged particle through the detectors as a delta-E vs. E' response. From the location of a particle's signals on a matrix, the instrument can identify the particle's charge and assume a mass. Energy is then calculated from the sum of energies deposited in the instrument. In principle, a particle's charge can be determined directly from the detector signals (assuming they're properly calibrated) and a given range-energy relationship, but doing so would take a lot of CPU cycles by the LET onboard processors, which can't even do floating point calculations natively anyway. (Calculations onboard are done in log-space anyway, using integer-rounded log tables.) The matrices take up a lot of memory, but using them takes up very little CPU.
I generated the response matrices used by the LET instruments. The matrices currently onboard the spacecraft were generated prior to launch using range-energy calculations and extensive test data. However, now that the STEREO spacecraft are in space, the environments that the instruments are in are significantly different than prior to launch, necessitating fine-tuning and recalibration. By early last week, Rick L., one of my Caltech colleagues, had finished calculating new gains and offsets for the solid state detectors using in-flight ADC STIM data (calibration pulses), and with these new gains and offsets, I could see how the matrices might need to be adjusted.
The following screenshot shows the kind of thing I've been working on since last week:
The window at left shows a graphical representation of one of the LET matrices (the colored bands), overlaid with particle data (the black dots) from a solar energetic particle event from December 2006. The matrices themselves are 128x400 arrays of integers, but they're usefully represented by the graphical figures, which I distribute to my colleagues for review. The y-axis is the delta-E measurement for a given range (energy deposit in the "first" detector), and the x-axis is the E' measurement for the range (residual, or remaining, energy deposited in the "second" detector). Energy deposits are governed (or approximated) by the Bethe-Bloch equation, which itself can be simplified to range-energy relationships in silicon. The result is that particles of various energies, charges, and masses lie on bands or tracks on the delta-E vs. E' plot, ranging in this case from protons at lower left (yellow band) to iron (green band at top) and nickel (yellow band above the iron band). The LET instrument identifies protons, He-3, He-4, C, N, O, Ne, Na, Mg, Al, Si, S, Ar, Ca, Fe, and Ni, shown as the alternating yellow, red, and green bands. The various shades of blue are background bands, and the uncolored (white) region at lower right shows "backward-going" particles or other noise. The orange rectangles are "livetime STIM" boxes for calibration purposes.
As colorful as the figure is, however, it's merely a graphical representation of a large matrix that itself is generated by range-energy curves (physics), tweaks to those curves to match actual data, and a lot of IDL code that I wrote to calculate those curves and to "paint" the matrices according to those curves. A tiny snippet of that code is in the window at right.
That said, all of my work at the office for the past couple of weeks has been to take Rick's most recent gains and offsets and particle data from some largish events that STEREO has detected recently, and to tweak the matrices further. Tweaks usually involve changing numbers on the multitude of tables I have which define the curves on the matrices. I've gone through dozens of private revisions alone in my office, and last week and this week, I released plots to my colleagues showing the new matrices and how the data lie atop the matrices. Those plots would result in additional requests for matrix modifications, followed by further tweaking alone in my office, and some of the requested modifications have resulted in actual changes to my code, rather than changes to tables. I also had to tweak my code -- both the matrix generation code as well as the data analysis code -- so that the turnaround time between tweaking the matrices, viewing them with data, and re-tweaking them would be shorter.
I released a third public revision yesterday (Friday), just before leaving work for the long weekend. Hopefully, this will be the last revision. If so, I'll convert the matrix file to a base64 representation of the file and submit it to my colleagues Andrew and Rick C., who will be able to include it with the next major firmware upload to the spacecraft. Once that's done, we may soon declare the LET instruments "commissioned" for the flight.
-----
The STEREO work has taken up most of my time since the last entry in this photoblog. Generally, since William was born, I don't like to bring work home, so I can keep my home life strictly devoted to family. There have been occasions when I've had to work at home, however, and this matrix work did occupy a few evenings.
I'm usually stricter about no-office-work on weekends than I am on weeknights. Last Saturday, while I was with James at My Gym, Hsuan and William went shopping at Target for some Valentine's day items for William to bring to daycare on Valentine's Day. While there, Hsuan found out that Target would be opening up a shipment of Nintendo Wii's the following Sunday morning, and they would start handing out numbers to anyone in line at 7:00 AM that Sunday morning to get one of the 22 (or was it 32?) Wii's in the shipment. Hsuan knew I've been casually interested in the new generation of gaming consoles. (I haven't played console games since the early 1980's, though I've played computer games often since my undergraduate days.)
Hsuan planned to buy me a Wii in secret, as a Valentine's Day gift for me, but William blabbed the secret to me on the way home. Hsuan and I talked about it. I wasn't sure how much interest there remained in the Wii and whether people would be camped out overnight in front of Target. If so, I didn't think it was worth her effort, but if she got there at 5:30 AM without a big line, I thought it might be worthwhile. When she got there at 5:30 AM or so, she was number 2 in line, and by the time the store opened, I think there were still 7 units unclaimed. So, there's strong interest still, but it's not impossible to get one any more.
After 9:30 AM Mass on Sunday morning, I started setting up the Wii, with William hovering around being impatiently patient. I wanted to route the audio and video cables from the Wii through my Harman Kardon home theater amplifier to the plasma TV, but after some struggling, I concluded that none of the RCA video inputs on the amplifier work any longer. I was disappointed; I will eventually have to upgrade the amplifier to a newer model, though I'll likely wait until we upgrade our cable and DVD player to high definition versions.
Meanwhile, I ended up hooking the Wii directly to the plasma TV via the TV's second A/V inputs, and from then on, everything worked splendidly. We -- mostly William and I -- spent most of Sunday playing the included Wii sports games -- boxing, tennis, baseball, bowling, and golf. I worry that the Wii's infrared sensor, which I taped to the top of the TV, may not detect me very well, given that the television is somewhat recessed in the home theater cabinetry, but it seems to detect William better, probably because he's shorter and more thoroughly in view of the sensor.
One result of our first day of play is that I spent the first half of this week with a sore elbow and back. I play tennis, baseball, and boxing rather vigorously.
------
Last weekend was cold and wet, so I didn't go out for my weekend 2-mile walk with Bessie. In fact, I haven't exercised at all since the week before.
------
Oh, and for Valentine's Day, I ended up buying Hsuan a dozen long-stemmed roses. These were the higher-quality, thicker-stemmed, very expensive kind of roses that the flower shop doesn't sell at "special" lower rates, but they should last a lot longer than the cheaper kinds. I gave William and James each a rose to give to Mommy.
However, we gave her the roses on the 13th, because Hsuan had to leave on Wednesday for another business trip that lasted through Thursday afternoon. That left me back in single father mode Wednesday night and Thursday morning. I brought William to his My Gym class on Wednesday evening, and afterward, I passed by McDonalds again for the boys' dinner. After dinner, bathing the boys, and putting them to bed, I did a little house work and then went to bed myself, bringing the boys to bed with me to save me the trouble of having to get up in the night if they woke up. Of course, I spent the night trying not to be kicked in the face by either boy on either side, as well as trying not to roll over them...
The photo is of me with James on the family room couch, before I went to bed, and after he woke up during one of his usual night-time awakenings. I brought him out of the boys' bedroom so he wouldn't wake up William.
Thursday morning, I had my hands full with the boys, including my daily chauffeur job of bringing them to daycare, that I had to skip my usual Thursday morning STEREO conference call. At this point, though, I had already submitted another matrix revision via e-mail, so I had made my contribution for the morning.
-----
This Thursday evening, I had my monthly Board meeting for the boys' daycare center. Since we were already past the halfway point in our fiscal year, I decided I'd do a half-year budget and finance assessment, to see how well we were meeting our budget goals. I found we're behind expectations for our income, but we more than make up for that shortfall by keeping our payroll and payroll-related expenses in line. I wrote up a report and presented it to the rest of the board. I expect to begin outlining a budget for next year before April.
-----
To top it all off, I've had a head cold for the past few days, with a lot of sinus congestion and headaches. I'm popping vitamin C like it's candy.
-----
Not specific to this week in particular, but Hsuan's snoring has gotten worse in recent months (at least when she doesn't use an anti-snoring throat spray), to the point that I've gradually given up trying to get James to sleep through the night by weaning him off of having her in the boys' bedroom. She gets that duty almost every night after I go to sleep because it's dangerous for me to drive the boys through rush hour traffic to daycare if I'm falling asleep at the wheel. On the other hand, she usually goes to sleep before I do, and given that I don't go to sleep before 2 AM on most nights, I take care of James' wake-ups while she sleeps and while I'm still awake.
Some nights, after I finally go to bed, I just pray for James to wake up crying so I can get some sleep alone in our bed...
-----
Oh, finally, we decided not to apply to the big Pasadena parochial school. We figure the odds are too long, and we've already got the regular private school applications done.
And, William has his yellow belt test next weekend. We think he's ready.
Posted: Sat - February 17, 2007 at 12:35 AM