Go to the U of M home page
School of Physics & Astronomy
Probe Mission Study Wiki

User Tools

Trace:
calibration_simulations

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
calibration_simulations [2018/10/10 04:41] – mauriziotomasicalibration_simulations [2019/02/20 04:56] (current) – Put an "outdated" warning message at the top of the page. mauriziotomasi
Line 1: Line 1:
 +**This page contains outdated information**. Refer to the report «Simulating dipole calibration for PICO», by Maurizio Tomasi, for the most up-to-date information {{::tomasi-calibration-note.pdf|}}.
 +
 ====== Calibration simulations ====== ====== Calibration simulations ======
  
Line 6: Line 8:
 ===== Hypothesis ===== ===== Hypothesis =====
  
-We assume to calibrate the instrument using the solar dipole of the +I assume to calibrate the instrument using the solar dipole of the 
-CMB, whose peak-to-peak amplitude is ~7 mK. We use+CMB, whose peak-to-peak amplitude is ~7 mK. I use
 [[https://github.com/hpc4cmb/toast|TOAST]] to produce the timelines, [[https://github.com/hpc4cmb/toast|TOAST]] to produce the timelines,
 and [[https://github.com/ziotom78/dacapo_calibration|DaCapo]] to and [[https://github.com/ziotom78/dacapo_calibration|DaCapo]] to
Line 80: Line 82:
 ===== Running DaCapo ===== ===== Running DaCapo =====
  
-We assume that the calibration code to be used in a real mission would+I assume that the calibration code to be used in a real mission would
 include a component-separation step able to reduce the bias due to the include a component-separation step able to reduce the bias due to the
 contamination of the dipole signal by Galactic large-scale contamination of the dipole signal by Galactic large-scale
-structures. Therefore, in our simulations we assume that the only+structures. Therefore, in our simulations I assume that the only
 signal in the sky apart from the dipole is the CMB itself. From the signal in the sky apart from the dipole is the CMB itself. From the
 point of view of calibration, CMB acts as a noise signal which is point of view of calibration, CMB acts as a noise signal which is
Line 96: Line 98:
 algorithm works as follows: algorithm works as follows:
  
-- Apply an extended version of the destriping equation, which +  - Apply an extended version of the destriping equation, which reconstructs the baselines of 1/f noise as well as the gain factors. This uses the following assumptions: 
-  reconstructs the baselines of 1/f noise as well as the gain +    * 1/f baselines have zero mean 
-  factors. This uses the following assumptions: +    * Gains are constrained by a model of the dipole, which must be provided as input 
-  * 1/f baselines have zero mean +  - The equation is solved iteratively using the Conjugate Gradient algorithm. At the end, the following estimates are available: 
-  * Gains are constrained by a model of the dipole, which must be +    * 1/f baselines 
-    provided as input +    * Timeline of gain factors 
-- The equation is solved iteratively using the Conjugate Gradient +    * Sky map (temperature only) without the dipole 
-  algorithm. At the end, the following estimates are available: +  - Iterate the algorithm again, subtracting the sky map obtained in the previous step from the timelines, in order to remove any systematic due to spurious signals (CMB, foregrounds). 
-  * 1/f baselines +  - When the 1/f baselines, the gain factors, and the sky map no longer change, save all the results and quit. 
-  * Timeline of gain factors +  - DaCapo must be run once per each detector, so that the sky map produced as output is typically noisier than the maps expected from a typical configuration of detectors.
-  * Sky map (temperature only) without the dipole +
-- Iterate the algorithm again, subtracting the sky map obtained in the +
-  previous step from the timelines, in order to remove any systematic +
-  due to spurious signals (CMB, foregrounds). +
-- When the 1/f baselines, the gain factors, and the sky map no longer change, +
-  save all the results and quit. +
-- DaCapo must be run once per each detector, so that the sky map +
-  produced as output is typically noisier than the maps expected from +
-  a typical configuration of detectors.+
  
-We employed the DaCapo algorithm in the following pipeline:+I employed the DaCapo algorithm in the following pipeline:
  
-- Use TOAST to produce timelines using two detectors (0A and 0B) along +  - Use TOAST to produce timelines using two detectors (0A and 0B) along the boresight direction, the nominal scanning strategy and realistic noise, and save them to FITS files. The gain used in this step is 1.0 (constant) for both detectors. TOAST produces a sky map (including the dipole) using the Madam map-maker (a destriper) and data from both detectors 0A and 0B. 
-  the boresight direction, the nominal scanning strategy and realistic +  - Run DaCapo on the FITS files produced during the previous stage. As the algorithm works on single detectors, the code must be run twice (once for detector 0A and once for 0B). 
-  noise, and save them to FITS files. The gain used in this step is +  - If DaCapo converges, re-run TOAST as in the previous step, but this time use the gains calculated by DaCapo instead of assuming a constant (1.0) gain. 
-  1.0 (constant) for both detectors. TOAST produces a sky map +  - At the end of the simulation, take the difference between the maps produced by TOAST in steps 1. and 3. and compute the power spectrum. These spectra represent the error caused by an imperfect estimate of gain drifts during the mission.
-  (including the dipole) using the Madam map-maker (a destriper) and +
-  data from both detectors 0A and 0B. +
-- Run DaCapo on the FITS files produced during the previous stage. As +
-  the algorithm works on single detectors, the code must be run twice +
-  (once for detector 0A and once for 0B). +
-- If DaCapo converges, re-run TOAST as in the previous step, but this +
-  time use the gains calculated by DaCapo instead of assuming a +
-  constant (1.0) gain. +
-- At the end of the simulation, take the difference between the maps +
-  produced by TOAST in steps 1. and 3. and compute the power +
-  spectrum. These spectra represent the error caused by an imperfect +
-  estimate of gain drifts during the mission.+
  
 Our simulations assume that each detector has a calibration factor Our simulations assume that each detector has a calibration factor
Line 193: Line 174:
 ===== Output maps ===== ===== Output maps =====
  
-As sketched in the previous section, we produce full-sky maps with+As sketched in the previous section, I produced full-sky maps with
 TOAST, calibrate them with DaCapo and re-run TOAST with the gains TOAST, calibrate them with DaCapo and re-run TOAST with the gains
 estimated by DaCapo to get a new full-sky map which contains the estimated by DaCapo to get a new full-sky map which contains the
Line 201: Line 182:
 I have run two sets of simulations, using the following assumptions: I have run two sets of simulations, using the following assumptions:
  
-- Two noiseless receivers (0A and 0B); +  - Two noiseless receivers (0A and 0B); 
-- Two W-band receivers with nominal white noise and 1/f noise with +  - Two W-band receivers with nominal white noise and 1/f noise with knee frequency 10 mHz.
-  knee frequency 10 mHz.+
  
 The differenced maps produced in the two cases are quite different: The differenced maps produced in the two cases are quite different:
  
-<IMAGE HERE>+{{ :systematicswg:cal_error_maps.png?600 |}}
  
 Of course, the scale of the effect considers the presence of two Of course, the scale of the effect considers the presence of two
Line 213: Line 193:
  
 Computing the BB spectrum of the map in the two cases and confronting Computing the BB spectrum of the map in the two cases and confronting
-it with the input map shows the scale of the effect:+it with the input map shows the scale of the effect (as well as two 
 +power spectra assuming only tensor modes, provided as a reference):
  
-<IMAGE HERE>+{{ :systematicswg:calibration_spectrum_bb.png?600 |}}
  
-In order to produce this plot, we rescaled the spectrum by 2/N, where +In order to produce this plot, I rescaled the spectrum by 2/N, where 
-N is the number of W-band detectors in the focal plane of PICO: this+N is the number of W-band bolometers in the focal plane of PICO, and 
 +then I performed a NET-weighted average over all the cosmology bands 
 +from 60 GHz (band 7) to 220 GHz (band 22): this
 assumes that the gain fluctuations are negligibly correlated among assumes that the gain fluctuations are negligibly correlated among
 detectors. This is the case for the case with realistic noise, but not detectors. This is the case for the case with realistic noise, but not
Line 225: Line 208:
 and 0B: and 0B:
  
-<IMAGE HERE>+{{ :systematicswg:gain_timeline.png?600 |}}
  
calibration_simulations.1539164507.txt.gz · Last modified: 2018/10/10 04:41 by mauriziotomasi

Page Tools