Difference between revisions of "LE algo"
| Line 44: | Line 44: | ||
|width="420pt"| | |width="420pt"| | ||
[[Image:run_31660_track12_3_off_drift23.png|thumb|420px|Drift time ch23, without interpolation]] | [[Image:run_31660_track12_3_off_drift23.png|thumb|420px|Drift time ch23, without interpolation]] | ||
| + | |} | ||
| + | |||
| + | Below: As above, for the interpolated data, but only events where a track can be fitted are selected (Events with ADC data going above pedestal mean + 5 sigma are selected. Events with 5 or more hits are selected. 20 samples starting 10 samples before the threshold crossing are sent to the LE algo after subtracting mean pedestal. The threshold for the LE algo is 3 sigma.) | ||
| + | |||
| + | {| border="0" cellpadding="2" | ||
| + | |width="420pt"| | ||
| + | [[Image:run_31660_track12_3_tr_drift23.png|thumb|420px|Drift time ch23 (tracks), with interpolation]] | ||
| + | |width="420pt"| | ||
| + | [[Image:run_31660_track12_3_resids23.png|thumb|420px|Residuals ch23, with interpolation]] | ||
| + | |} | ||
| + | |||
| + | {| border="0" cellpadding="2" | ||
| + | |width="420pt"| | ||
| + | [[Image:run_31660_track12_3_residsdrift23.png|thumb|420px|Residuals vs drift time ch23, with interpolation]] | ||
| + | |width="420pt"| | ||
| + | [[Image:run_31660_track12_3_resids23over2100ns.png|thumb|420px|Residuals ch23 for drift time > 2100ns ch23, with interpolation]] | ||
|} | |} | ||
Revision as of 16:36, 18 May 2011
- Page still under edit, incomplete
Results using Gerard's leading edge algo to extract drift-time from CDC prototype cosmics data
- 50/50 mixed Ar, CO2
- Recalibrated MFCs
- Outer plenum added to CDC prototype to prevent leaks
- New modified HVB
- 2050V
- Modified preamp4 (terminating resistors removed)
- Prototype tilted at 45o to enclosing scintillators (which trigger daq)
The original algorithm is here LE algo It takes 50 samples from the ADC data, convolutes them with the filter which I think is the inverse of the filter which the fADC uses when it digitises its input, optionally interpolates 4 extra points between each sample, and returns the first value above a predefined threshold. There are a variety of implementations below.
Below: Events with ADC data going above pedestal mean + 5 sigma are selected. 50 samples starting 25 samples before the threshold crossing are sent to the LE algo after subtracting mean pedestal. The threshold for the LE algo is 2 sigma. Left: drift-time with interpolation; Right: no interpolation
Below: 20 samples starting 10 samples before the threshold crossing are sent to the LE algo after pedestal subtraction.
Below: Events with ADC data going above pedestal mean + 5 sigma are selected. Events with 5 or more hits are selected. 20 samples starting 10 samples before the threshold crossing are sent to the LE algo after subtracting mean pedestal. The threshold for the LE algo is 3 sigma. Left: drift-time with interpolation; Right: no interpolation
Below: As above, for the interpolated data, but only events where a track can be fitted are selected (Events with ADC data going above pedestal mean + 5 sigma are selected. Events with 5 or more hits are selected. 20 samples starting 10 samples before the threshold crossing are sent to the LE algo after subtracting mean pedestal. The threshold for the LE algo is 3 sigma.)
