Suppression of the low energy noise an improved approach
Download
1 / 119

Suppression of the low-energy noise: an improved approach - PowerPoint PPT Presentation


  • 101 Views
  • Uploaded on

Suppression of the low-energy noise: an improved approach. EPIC-pn detector noise. 50.7 ks closed FF (rev 059). Quadrant 0. adu. Suppression of the EPIC-pn detector noise. Original method (“method 1”), implemented in the SAS task epreject :

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about ' Suppression of the low-energy noise: an improved approach' - cicero


An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
Suppression of the low energy noise an improved approach
Suppression of the low-energy noise: an improved approach

Suppression of the low-energy noise


EPIC-pn detector noise

50.7 ks closed FF (rev 059)

Quadrant 0

adu

Suppression of the low-energy noise


Suppression of the EPIC-pn detector noise

Original method (“method 1”), implemented in the SAS task epreject:

Use closed exposures to determine the noise properties (spatially and spectrally) and use them as input for the Monte-Carlo filtering

Suppression of the low-energy noise


closed FF rev 462 23.2 ks

3

0

50

0

application of the original method (implemented in the SAS task epreject)

Suppression of the low-energy noise


Vela snr ff rev 534 38 8 ks
Vela SNR FF rev 534 38.8 ks

withoutepreject

Suppression of the low-energy noise


Vela snr ff rev 534 38 8 ks1
Vela SNR FF rev 534 38.8 ks

with epreject

application of the original method (implemented in the SAS task epreject)

Suppression of the low-energy noise


EPIC-pn detector noise in LW mode

closed, 61 ks, rev 790

Suppression of the low-energy noise


20

50 adu

EPIC-pn detector noise (LW)

Original method (“method 1”):

Suppression of events below 50 adu according to a reference noise distribution deduced from long calclosed exposures

Shown in blue is the result of this method applied to a (long) calclosed exposure, to illustrate the effect on a very faint X-ray source

100% suppression

Suppression of the low-energy noise


20

50 adu

EPIC-pn detector noise (LW)

Original method (“method 1”):

Suppression of events below 50 adu according to a reference noise distribution deduced from long calclosed exposures

Shown in blue is the result of this method applied to a (long) calclosed exposure, to illustrate the effect on a very faint X-ray source

99% suppression

Suppression of the low-energy noise


20

50 adu

EPIC-pn detector noise (LW)

Original method (“method 1”):

Suppression of events below 50 adu according to a reference noise distribution deduced from long calclosed exposures

Shown in blue is the result of this method applied to a (long) calclosed exposure, to illustrate the effect on a very faint X-ray source

90% suppression

Suppression of the low-energy noise


20

50 adu

EPIC-pn detector noise (LW)

Original method (“method 1”):

Suppression of events below 50 adu according to a reference noise distribution deduced from long calclosed exposures

Shown in blue is the result of this method applied to a (long) calclosed exposure, to illustrate the effect on a very faint X-ray source

70% suppression

Suppression of the low-energy noise


20

50 adu

EPIC-pn detector noise (LW)

Original method (“method 1”):

Suppression of events below 50 adu according to a reference noise distribution deduced from long calclosed exposures

Shown in blue is the result of this method applied to a (long) calclosed exposure, to illustrate the effect on a very faint X-ray source

50% suppression

Suppression of the low-energy noise


20

50 adu

EPIC-pn detector noise (LW)

  • This method works well for bright sources, but may cause an apparent deficit of photons between ~150 and 250 eV for faint sources

  • Other problem: inhomogeneous residual noise at ~20 adu

Suppression of the low-energy noise


Vela snr ff rev 740 29 4 ks

20 adu images after epreject

Vela SNR FF rev 740 29.4 ks

default noiseparameters adjusted

Suppression of the low-energy noise


Long closed LW exposures

Suppression of the low-energy noise


Suppression of the EPIC-pn detector noise

Alternative method (“method 2”):

Try to determine the noise properties (from closed exposures) at higher energies and extrapolate them down to 20 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790 (closed), all events

range used for fit: 40-100 adu

range shown: 20-400 adu

x log y log

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790 (closed), all events

range used for fit: 40-100 adu

range shown: 20-400 adu

x log y log

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790 (closed), all events

range used for fit: 40-100 adu

range shown: 20-400 adu

x log y log

for comparison: method 1

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

method 2, noise properties derived from rev 790 (closed) applied to rev 790

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

method 2, noise properties derived from rev 790 (closed) applied to rev 546

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

method 2, noise properties derived from rev 790 (closed) applied to rev 553

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

method 2, noise properties derived from rev 790 (closed) applied to rev 730

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

method 2, noise properties derived from rev 790 (closed) applied to rev 974

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

  • Method 2 would work well if there were no temporal changes of the low-energy noise

    Whenever the noise properties of the sample and template are (slightly) different, the subtraction leads to residuals in the spectra and the images

    Other problem: assumption of constant noise properties along CCD rows (required in order to get enough statistics) not valid for raw events:

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

20 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

21 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

22 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

23 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

24 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

25 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

26 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

27 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

28 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

29 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

30 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

31-35 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

36-40 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

41-45 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

46-50 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

51-55 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

56-60 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

61-70 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

71-80 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

81-90 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

91-100 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

101-120 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

121-140 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

141-180 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

after offset corrections

all events

181-250 adu

Suppression of the low-energy noise


Suppression of the EPIC-pn detector noise

New approach (“method 3”):

Part A: Remove noisy frames, identified by the number of low (20-30) adu events per CCD

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks

singles, 20 – 30 adu, frames with 1 event below 31 adu per frame and CCD

singles, 20 – 30 adu, frames with 2 events below 31 adu per frame and CCD

singles, 20 – 30 adu, frames with at least 3 events below 31 adu per frame and CCD

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks

(reconstructed) number of frames with no event below 31 adu per CCD

number of frames with 1 event below 31 adu per CCD

number of frames with 2 events below 31 adu per CCD

total number of frames

number of frames with more than 9 events below 31 adu per CCD

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks

=

+

singles, 20 – 30 adu, frames with 1 or 2 events below 31 adu per frame and CCD

singles, 20 – 30 adu, frames with more than 2 events below 31 adu per frame and CCD

singles, 20 – 30 adu, all frames

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks

=

+

singles, 20 – 30 adu, frames with 1 event below 31 adu per frame and CCD

singles, 20 – 30 adu, frames with more than 1 event below 31 adu per frame and CCD

singles, 20 – 30 adu, all frames

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks

removing frames with more than 1 event below 31 adu per CCD and frame:

number of frames removed

original number of frames

percentage of frames removed

Suppression of the low-energy noise


Suppression of the EPIC-pn detector noise

New approach (“method 3”):

Part B: Remove noise by rejecting invalid patterns and selecting only singles

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790, 61 ks closed, no offset corrections, same intensity scale

30-50 adu

raw data: all events

processed data: all “entries”

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

no offset corrections

non-singles

processed data

30-50 adu

Suppression of the low-energy noise


EPIC-pn detector noise (LW)

rev 790

61 ks closed

no offset corrections

singles

processed data

30-50 adu

Suppression of the low-energy noise


EPIC-pn detector noise (FF)

rev 462

23 ks closed

no offset corrections

all events

raw data

30-50 adu

Suppression of the low-energy noise


EPIC-pn detector noise (FF)

rev 462

23 ks closed

no offset corrections

all “entries”

processed data

30-50 adu

Suppression of the low-energy noise


EPIC-pn detector noise (FF)

rev 462

23 ks closed

no offset corrections

non-singles

processed data

30-50 adu

Suppression of the low-energy noise


EPIC-pn detector noise (FF)

rev 462

23 ks closed

no offset corrections

singles

processed data

30-50 adu

Suppression of the low-energy noise


Suppression of the EPIC-pn detector noise

New approach (“method 3”):

Part C: Try to determine the noise properties from the same exposure at higher energies and extrapolate them down to 20 adu

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks

In order to determine the noise properties, some areas have to be ignored:

mask for source regions to be avoided for noise determination

singles, 0.3-3.0 keV

source region mask

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks

In order to determine the noise properties, some areas have to be ignored:

noisy pixel mask

final mask, excluding source regions and noisy pixels

Suppression of the low-energy noise


Deriving the general properties of the detector noise from closed exposures

closed exposures 790 & 974, 73.6 ks

Suppression of the low-energy noise


Deriving the general properties of the detector noise from closed exposures

closed exposures 790 & 974, after removal of frames with more than 2 events/frame/ccd < 31 adu

Suppression of the low-energy noise


Deriving the general properties of the detector noise from closed exposures

common fit (for all CCDs) between 40 and 100 adu

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions included

red: data set to be corrected, blue: closed exposure

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions excluded

red: data set to be corrected, blue: closed exposure

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions excluded

average 45-55 adu flux

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions excluded

extrapolated flux at 20 adu

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions excluded

extrapolated flux at 20 adu averaged between all CCDs

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions excluded

spectrum after noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions excluded

spectrum after noise suppression, source regions excluded

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions excluded

spectrum before/after noise suppression, source regions excluded

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions included

original spectrum, source regions included

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions included

original spectrum, source regions included, after noisy frame removal

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions included

original spectrum, source regions included, after noisy frame removal and (spatially homogeneous) noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions included

original spectrum, including source regions, before/after noisy frame removal and (spatially homogeneous) noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

unfiltered data singles, 25 adu after noisy frame removal

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

every second single in CCD4 kept singles, 25 adu every 10th single in CCD4 kept

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

after noisy frame removal singles, 25 adu after spatially uniform noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

after noisy frame removal singles, 25 adu after spatially variable noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

unfiltered data singles, 20 adu after noisy frame removal

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

after spatially uniform noise suppression singles, 20 adu after spatially variable noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

unfiltered data singles, 21 adu after noisy frame removal

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

after spatially uniform noise suppression singles, 21 adu after spatially variable noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

unfiltered data singles, 22 adu after noisy frame removal

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

after spatially uniform noise suppression singles, 22 adu after spatially variable noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

unfiltered data singles, 23 adu after noisy frame removal

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

after spatially uniform noise suppression singles, 23 adu after spatially variable noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

unfiltered data singles, 24 adu after noisy frame removal

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

after spatially uniform noise suppression singles, 24 adu after spatially variable noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

unfiltered data singles, 25 adu after noisy frame removal

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

after spatially uniform noise suppression singles, 25 adu after spatially variable noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

unfiltered data singles, 26 adu after noisy frame removal

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

after spatially uniform noise suppression singles, 26 adu after spatially variable noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

unfiltered data singles, 27 adu after noisy frame removal

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

after spatially uniform noise suppression singles, 27 adu after spatially variable noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

unfiltered data singles, 28 adu after noisy frame removal

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

after spatially uniform noise suppression singles, 28 adu after spatially variable noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

unfiltered data singles, 29 adu after noisy frame removal

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

after spatially uniform noise suppression singles, 29 adu after spatially variable noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

unfiltered data singles, 30 adu after noisy frame removal

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

after spatially uniform noise suppression singles, 30 adu after spatially variable noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

singles, 100-130 eV

after noisy frame removal and spatially uniform noise suppression

after noisy frame removal and spatially variable noise suppression

original data

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

singles, 130-140 eV

after noisy frame removal and spatially uniform noise suppression

after noisy frame removal and spatially variable noise suppression

original data

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

singles, 140-150 eV

after noisy frame removal and spatially uniform noise suppression

after noisy frame removal and spatially variable noise suppression

original data

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

singles, 150-200 eV

after noisy frame removal and spatially uniform noise suppression

after noisy frame removal and spatially variable noise suppression

original data

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks included

singles, 200-300 eV

after noisy frame removal and spatially uniform noise suppression

after noisy frame removal and spatially variable noise suppression

original data

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions included

original spectrum, including source regions, after noisy frame removal and (spatially variable) noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions included

original spectrum, including source regions, after noisy frame removal and (below 28 adu: spatially variable) noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions excluded

original spectrum, excluding source regions, after noisy frame removal and (below 28 adu: spatially variable) noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks, source regions excluded

original spectrum, excluding source regions, after noisy frame removal and (spatially variable) noise suppression

Suppression of the low-energy noise


SNR 1E 0102, rev 803, LW, thick, 30 ks excluded

singles, 120-140 eV

unfiltered data

after noisy frame removal

after (spatially variable) noise suppression

Suppression of the low-energy noise



ad