Links:
RHESSI Simulation Control Parameters
Part II, Thermal-Nonthermal Simulation Example
Sample Simulated Spectrum Test
For a sample time, we'll use the flare of 4-jul-2002 06:20 UT
hsi_switch, /sim ;sim eventlist - true aspect
o = hsi_eventlist(sim_ut_ref='4-jul-2002 06:22:36')
model = {gaussian_source_str} ;Input to hsi_modulate_point_source
model.amplitude = 1.0
model.xysigma = [0.0001, 0.0001]
model.xypos = [0.0, 0.0]
spp = [1.35, 2.68, 13.72, 4.21] ;Nonthermal part of ospex fit
o -> set, sim_spec_pars = spp ;spectral parameters
o -> set, sim_spec_model = 'f_bpow' ;broken power law (the default)
o -> set, /sim_use_spectrum ;Use the spectral pars input
o -> set, sim_background = 0 ;No background (the default)
o -> set, sim_xyoffset = [825.0, -325.0] ;position of 25-50 keV source
o -> set, sim_model = model
o -> set, sim_energy_band = [20.0, 25.0]
o -> set, /sim_diagonal_srm ;Use diagonal energy response
Get the data, and write it out
ev = o -> getdata(sim_time_range = [0.0, 24.0]) o -> Write, 'hsi_point_source_0.fits', time_range=[0, 0], energy_band=[0, 0]The [0, 0]'s insure that all of the simulated photons end up in the file
Do the image:
hsi_switch, /flight ;
img_obj = hsi_image(ev_filename = 'hsi_point_source_0.fits')
image = img_obj -> getdata(xyoffset = [825.0, -325.0], $
energy_band = [20.0, 25.0], $
det_index_mask = [1, 0, 1, 1, 1, 1, 0, 0, 0], $
time_range = [0.0, 24.0], $
pixel_size = [0.25, 0.25])
img_obj -> set, image_alg = 'clean'
img_obj -> set_no_screen_output
image1 = img_obj -> getdata()
Stats for 50% contour ( See the contour )
Box Flux Area Centroid (X,Y) Peak (X,Y) St Dev (X,Y) Peak 0 17.827 9.75 824.88 -325.12 825.03 -324.98 0.79 0.81 2.7758
FWHM about 3 arcsec.
Ok, add some background..
o -> set, sim_background = 10 ;background = 10*HESSI_BACKGROUNDThis is a relatively high background, but S/N is approximately 35 to 1. Results are about the same.
o = hsi_eventlist(sim_ut_ref='4-jul-2002 06:22:36')
model = {gaussian_source_str}
model.amplitude = 1.0
model.xysigma = [0.0001, 0.0001]
model.xypos = [0.0, 0.0]
spp = [10.0, 3.0, 400.0, 2.0]
o -> set, sim_spec_pars = spp
o -> set, sim_spec_model = 'f_bpow'
o -> set, /sim_use_spectrum
o -> set, sim_background = 0
o -> set, sim_xyoffset = [825.0, -325.0]
o -> set, sim_model = model
o -> set, sim_energy_band = [100.0, 300.0]
o -> set, /sim_diagonal_srm
o -> set, sim_a2d_index_mask = [bytarr(9)+1]
ev = o -> getdata(sim_time_range = [0.0, 24.0])
o -> Write, 'hsi_point_source_1.fits', time_range=[0, 0], energy_band=[0, 0]
Do the image
hsi_switch, /flight
img_obj = hsi_image(ev_filename = 'hsi_point_source_1.fits')
img_obj -> set, image_alg = 'clean'
img_obj -> set_no_screen_output
image = img_obj -> getdata(xyoffset = [825.0, -325.0], $
energy_band = [100.0, 300.0], $
det_index_mask = [0, 1, 1, 1, 1, 1, 0, 0, 0], $
time_range = [0.0, 24.0], $
pixel_size = [1.0, 1.0])
Stats for 50% contour ( See the contour )
Box Flux Area Centroid (X,Y) Peak (X,Y) St Dev (X,Y) Peak 0 35.657 40.00 824.52 -325.28 825.04 -324.93 1.52 1.71 1.5016
FWHM about 6 arcsec.
image2 = img_obj -> getdata(det_index_mask = [1, 1, 1, 1, 1, 1, 1, 1, 1])
Gives noise...
Add background, here S/N is approximately 10 to 1.
o -> set, sim_background = 10
Not much changes
hsi_switch, /sim
o = hsi_eventlist(sim_ut_ref='4-jul-2002 06:22:36')
model = {gaussian_source_str}
model.amplitude = 1.0
model.xysigma = [0.0001, 0.0001]
model.xypos = [1.0, -2.0]
model = [model, model]
model[1].xypos = [-1.0, 2.0]
model[1].amplitude = 0.1
spp = [1.35, 2.68, 13.72, 4.21]
o -> set, sim_spec_pars = spp
o -> set, sim_spec_model = 'f_bpow'
o -> set, /sim_use_spectrum
o -> set, sim_background = 0
o -> set, sim_xyoffset = [825.0, -325.0]
o -> set, sim_model = model
o -> set, sim_energy_band = [20.0, 25.0]
o -> set, /sim_diagonal_srm
ev = o -> getdata(sim_time_range = [0.0, 24.0])
o -> Write, 'hsi_point_source_3.fits', time_range=[0, 0], energy_band=[0, 0]
hsi_switch, /flight
img_obj = hsi_image(ev_filename = 'hsi_point_source_3.fits')
img_obj -> set, image_alg = 'clean'
img_obj -> set_no_screen_output
image = img_obj -> getdata(xyoffset = [825.0, -325.0], $
energy_band = [20.0, 25.0], $
det_index_mask = [1, 0, 1, 1, 1, 1, 0, 0, 0], $
time_range = [0.0, 24.0], $
pixel_size = [0.25, 0.25])
2 arcsec boxes ( See the boxes )
Box Flux Area Centroid (X,Y) Peak (X,Y) St Dev (X,Y) Peak 0 19.695 14.25 825.86 -327.09 825.97 -326.97 1.03 0.98 2.4874 1 2.4763 14.25 824.00 -323.74 826.00 -324.17 1.03 0.94 0.46693
Add noise, as before (S/N is 35 to 1)
2 arcsec boxes ( See the boxes ) Box Flux Area Centroid (X,Y) Peak (X,Y) St Dev (X,Y) Peak 0 19.319 14.25 825.85 -327.13 825.94 -327.05 0.97 0.96 2.4413 1 2.0226 14.25 824.56 -323.47 826.18 -324.28 0.61 1.09 0.44604 2 1.0682 14.25 829.03 -323.25 829.88 -323.11 0.00 0.52 0.24671 3 0.72948 14.25 822.07 -326.67 821.30 -326.04 0.00 1.14 0.18022
hsi_switch, /sim
o = hsi_eventlist(sim_ut_ref='4-jul-2002 06:22:36')
model = {gaussian_source_str}
model.amplitude = 1.0
model.xysigma = [0.0001, 10.0]
model.xypos = [0.0, 0.0]
model.tilt_angle_deg = 30.0
spp = [1.35, 2.68, 13.72, 4.21]
o -> set, sim_spec_pars = spp
o -> set, sim_spec_model = 'f_bpow'
o -> set, /sim_use_spectrum
o -> set, sim_background = 0
o -> set, sim_xyoffset = [825.0, -325.0]
o -> set, sim_model = model
o -> set, sim_energy_band = [20.0, 25.0]
o -> set, /sim_diagonal_srm
ev = o -> getdata(sim_time_range = [0.0, 24.0])
o -> Write, 'hsi_long_source_1.fits', time_range=[0, 0], energy_band=[0, 0]
Here is what the source should look like:
hsi_switch, /flight
img_obj = hsi_image(ev_filename = 'hsi_long_source_1.fits')
img_obj -> set, image_alg = 'clean'
img_obj -> set_no_screen_output
image = img_obj -> getdata(xyoffset = [825.0, -325.0], $
energy_band = [20.0, 25.0], $
det_index_mask = [1, 0, 1, 1, 1, 1, 0, 0, 0], $
time_range = [0.0, 24.0], $
pixel_size = [1.0, 1.0])
Not so great, increase photon flux by 10.
What if there are 2 long sources?
model.amplitude = 1.0 model.xysigma = [0.0001, 10.0] model.xypos = [0.0, 15.0] model.tilt_angle_deg = 30.0 model = [model, model] model[1].xypos = [-15.0, 0.0]
Not so good, maybe Pixons will help
Comments to: jimm@ssl.berkeley.edu
21-Oct-2004, jmm