NAME

SYNOPSIS

DESCRIPTION

Three distinct detection algorithms are available for this task.

The first detection method uses a simple algorithm, based on the trigger method on the GINGA satelite. This is specified by setting det_alg=GINGA, which is the default. The algorithm is as follows:

1. Integrate background rate for bgd_integ_time seconds, if no previous burst is overlapping.
2. Integrate light curve after background interval for delta_t seconds.
3. Determine if light curve is sigma above background.

The second algorithm is based on the HETE-2 burst trigger algorithm (2003AIPC..662...97T). The definition of "trigger" that is used below is a background interval followed by a foreground interval, followed by another background interval. The basic algorithm is as follows:

1. Iterate over light curve bins
2. Calculate the best trigger using current light curve time bin as the start of the foreground interval. Best trigger is one with highest sigma.
3. If sigma of trigger > theshold, calculate burst duration, sigma, total counts etc. Report as a burst in output file.

The individual trigger/sigma calculations in step 2 above are made as follows:

1. Fit a 1st order polynomial (a line) to the two trigger background intervals.
2. Calculate the expected background counts in the foreground interval by evaluating the fitted polynomial at each foreground time bin, and converting to estimated counts.
3. Calculate the sigma of the foreground given the estimated background counts.
4. If the calculated sigma > sigma, save trigger as a good trigger.

The third detection method is an occultation step detection. It is based loosely on the method described in: 2002ApJS..138..149H .

The output file is either a pure text file with several columns, and a comment header (comments are indicated as lines starting with '#'), or a FITS table with several columns describing each detected "event". The type of output file can be specified with the outtype parameter.

The columns in the output text file are as follows:

Burst Start Date (UTC)

YYYY-MM-DD

Burst Start Time (UTC)

HH:MM:SS

Burst Start Time (MET s)

float

Burst Stop Time (MET s)

float

Burst Duration (s)

Time that light curve spent <sigma> above background

Total Burst Counts

Sum of all counts betwee Burst Start Time and Burst Stop Time

Total sigma of burst

Sigma above background for entire burst

The columns in the output FITS file are as follows:

TIMESTART

[s] Event start time in MET

TIMESTOP

[s] Event stop time in MET

DATESTART

[UTC] Event start date

DATESTOP

[UTC] Event stop date

METHOD

Method used to detect events

RATE

[c/s] Event average rate, bkg subtracted

ERROR

[c/s] Event average rate error

RATESNR

Signal to noise ratio of the event

FORERATE

[c/s] Event average rate not bkg subtracted

FOREERROR

[c/s] Event average rate error not bkg subtracted

BKGRATE

[c/s] Background average rate

BKGERROR

[c/s] Background average rate error

BKGSTART

[s] Background start time in MET

BKGSTOP

[s] Background stop time in MET

BKGCOEFFS

Background rate fit coefficients

CHANBIN

WAM channel binning type (TH or PH)

CHANMIN

Lower channel number

CHANMAX

Upper channel number

T50

[s] T50 of the event

T50ERR

[s] T50 error estimate

T50START

[s] T50 start time in MET

T50STOP

[s] T50 stop time in MET

T90

[s] T90 of the event

T90ERR

[s] T90 error estimate

T90START

[s] T90 start time in MET

T90STOP

[s] T90 stop time in MET

DATAMODE

WAM datamode

DETNAM

WAM detector name

TLM_FILE

Telemetry file name

PARAMETERS

input_name [filename]

Input HXD WAM light curve (see also fhelp hxdmkwamlc)

outfile [string]

Output file name.

bgd_integ_time = 8 [int]

Background light curve integration time (in seconds).

delta_t = 1.0 [float]

Source integration time.

sigma = 5.5 [float]

Detection threshold in sigma.

(bgd_early_gap = 0) [int]

Gap (seconds) between 'early' background interval and foreground interval. det_alg=HETE2 only.

(bgd_late_gap = 0) [int]

Gap (seconds) between foreground interval and 'late' background interval. det_alg=HETE2 only.

(det_alg = HETE2) [0|GINGA|1|HETE2|2|STEP]

Burst detection algorithm

(use_trigger_set = no) [bool]

Use HETE-2 trigger sets instead of input trigger specification? det_alg=HETE2 only.

(trigger_set = MEDIUM) [0|SHORT|1|MEDIUM|2|LONG]

HETE-2 trigger set to use. These are based on the short, long and medium trigger sets from the paper:

Tavenner, T, et al - 2003AIPC..662...97T

det_alg=HETE2 only.

(step_window = 240) [int]

Window size in seconds for step fitting. det_alg=STEP only.

(step_delchi = 2.71) [real]

Change in delta chi-squared to be considered a '1-sigma' change in fit parameter error estimates. det_alg=STEP only.

(outtype = 1) [int]

Output file type (0=FITS table, 1=Text table)

(gaptol = 2.0) [real]

Gap tolerance (s).

(overlaptol = 10.0) [real]

Burst detection overlap tolerance (s). Bursts detected that are within overlaptol of one another will be merged.

(maxdur = 200.0) [real]

Maximum allowed burst duration.

(durest = yes) [bool]

Try to refine burst duration estimate (T50/T90 are always calculated)?

(leapfile = CALDB;$ENV{LHEA_DATA}/leapsec.fits) [file]

Leap seconds FITS table filename?

(num_event = -1) [int]

Number of event (-1=all,0=exit)

(event_freq = 10000) [int]

Event number printout frequency

(anl_verbose = -1) [int]

ANL verbose level (-1:full, 0:minimum)

(anl_profile = yes) [bool]

Enable ANL module profiling

(chatter = 2) [int]

Terminal chatter level (0:min, 2:norm, 5:max)

(clobber = no) [bool]

Clobber existing output file?

NOTES

LAST MODIFIED

March 2011