Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/sas/8.0.0/doc/epevents/node16.html
Дата изменения: Wed Jul 2 05:21:44 2008
Дата индексирования: Fri Sep 5 20:47:55 2008
Кодировка:
Поисковые слова: lmc

XMM-Newton Science Analysis System


epevents (epevents-6.43) [xmmsas_20080701_1801-8.0.0]

Meta Index / Home Page / Output Files

Details about the pattern recognition

epevents performs a pattern analysis to identify and characterize split events: the charge distribution in split events is analysed and the detected patterns are described by four parameters:

1. PAT_TYP: PATTERN TYPE, the absolute value identifies the pattern, i.e. the number of pixels which belong to the pattern: single (1), double (2), triple (3), quadruple (4), ... negative if amplitude of the corresponding pixel should be ignored (e.g. if not main pixel or or also main pixel if it is at an edge). NULL value is set for trailing events which are excluded from the pattern analysis and MIPs.
2. PAT_IND: PATTERN INDEX, running number of pattern within a frame, starting with 1. All pixels within a pattern share the same (absolute value of) PAT_IND so this parameter is needed to identify split partners. For the pixel of a pattern with the highest PHA this value is positive; for the others it is negative. Unresolved patterns (e.g. those with PAT_TYP $>4$) contain only negative values. Singles are not regarded as patterns and thus have PAT_IND=0.
3. PAT_ORI: PATTERN ORIENTATION, zero for singles and unresolved patterns; otherwise for valid patterns a three-digit code, with the last digit specifying the orientation of the pattern and the first two the position of the pixel in the pattern; the pattern is inserted into a $3\times3$ pixel array with the main pixel (having the highest PHA value) at the center (2,2):
   • first digit: x position of the pixel on the pattern
   • second digit: y position of the pixel on the pattern
   • third digit: orientation of the pattern:
     • for doubles: neighbouring pixel (i.e. pixel with smaller PHA than the main pixel) to the right (1), bottom (2), left (3), top (4) with respect to the local maximum in the pattern
     • for triples: L-shaped distribution with the line of symmetry towards: lower right (1), lower left (2), upper left (3), upper right (4)
     • for quadruples: square-shaped distribution, offset towards: lower right (1), lower left (2), upper left (3), upper right (4)
   All coordinates refer to a chip-oriented coordinate system (ix,iy) with the CAMEX at bottom, seen from the X-ray source, i.e. the x coordinate is running towards the right and the y coordinate upwards, with the CAMEX at y = 0.
4. PAT_SEQ: PATTERN SEQUENCE, distance (in pixels) to neighbouring event in the same frame and column (along the readout direction) towards the CAMEX. PAT_SEQ = 0 indicates the first pixel read in a column. This information is used to determine the CTI as a function of the frame history - the charge shift during the readout of earlier events effects the CTI of the actual event.