Image Processing Reference
In-Depth Information
Rates vs. time for 25-35 requirement
Rates vs. time for 25-35 requirement
B
B
Rates vs. time for 15-25 Hz requirement
Rates vs. time for 15-25 Hz requirement
A
A
60
60
70
70
60
60
50
50
50
50
40
40
rate 0.26
rate 0.27
rate 0.26
rate 0.27
veto
40
40
30
30
rate
veto
rate
veto
30
30
20
20
10
10
0
0
time (s)
time (s)
0
0
103
206
309
412
515
618
721
824
927
0
96
192
288
384
480
576
672
768
864
960
C
C
Rates vs. time for 35-45 Hz requirement
Rates vs. time for 35-45 Hz requirement
Contribution of the DCT sub-triggers
to the final trigger for the Occ = 7
Contribution of the DCT sub-triggers
to the final trigger for the Occ = 7
70
70
12
12
60
60
PMT1 : 15 25
PMT2 : 15 25
PMT3 : 15 25
PMT1 : 25 35
PMT2 : 25 35
PMT3 : 25 35
PMT1 : 15 25
PMT2 : 15 25
PMT3 : 15 25
PMT1 : 25 35
PMT2 : 25 35
PMT3 : 25 35
PMT1 : 35 45
PMT2 : 35 45
PMT3 : 35 45
10
10
50
50
rate 0.28
rate 0.27
rate 0.26
t
rate 0.28
rate 0.27
rate 0.26
rate 0.25
veto
8
8
40
40
6
6
30
30
4
4
20
2
10
0
0
time (s)
23456789 0 1 2 3 4
0
0
96
192 288 384 480 576 672 768 864 960
Fig. 15. Rates obtained in the test detector during an auto-calibration process (graphs A, B
and C). A comparison of a contribution of "fired" DCT coefficients generating the
sub-triggers for all three PMT channels and for various trigger rates requirements
(right-down). There are no significant differences in a contribution of a fixed coefficients for
different PMT and various configuration of trigger rate requirements
high and the attenuation factor has to be increased (the acceptance lane is narrowed down).
The graph B shows the process, when the initial parameters are optimal and the acceptance
lane is not modified (only the veto threshold is tuned). The graph C shows the process when
the initial parameters give too low trigger rate and the acceptance lane is changed three times.
10. Conclusion
The pattern recognition technique implemented parallel with the standard threshold detection
may improve an efficency of a registration of rare events, especially for a single PMT in the
surface detector, when the coincidence technique cannot be longer used.
The optimized algorithm of the spectral trigger based on the Discrete Cosine Transform with
veto and auto-calibration procedure has been successfully implemented into the FPGA and
showed the perfect stability in the real detector. Measurements in the test detector confirmed
assumption for a selection of limited amount of DCT coefficients and a stability of algorithm
for arbitrarily selected acceptance lane of the spectral trigger rate.
Although 6 surface detectors from the Pierre Auger Observatory have been used for the tests,
the spectral trigger is being developed more generally for future ground EAS arrays using -
other than the present Pierre Auger Observatory - only one PMT per station.
11. Acknowledgement
The author would like to thank the Pierre Auger Collaboration for being allowed to use a
PAO infrastructure and a test-hexagon and for getting the data made available. The successful
installation and preliminary tests of the new Front End Boards with the DCT trigger would
Search WWH ::




Custom Search