| 1 | /* ======================================================================== *\
|
|---|
| 2 | !
|
|---|
| 3 | ! *
|
|---|
| 4 | ! * This file is part of MARS, the MAGIC Analysis and Reconstruction
|
|---|
| 5 | ! * Software. It is distributed to you in the hope that it can be a useful
|
|---|
| 6 | ! * and timesaving tool in analyzing Data of imaging Cerenkov telescopes.
|
|---|
| 7 | ! * It is distributed WITHOUT ANY WARRANTY.
|
|---|
| 8 | ! *
|
|---|
| 9 | ! * Permission to use, copy, modify and distribute this software and its
|
|---|
| 10 | ! * documentation for any purpose is hereby granted without fee,
|
|---|
| 11 | ! * provided that the above copyright notice appear in all copies and
|
|---|
| 12 | ! * that both that copyright notice and this permission notice appear
|
|---|
| 13 | ! * in supporting documentation. It is provided "as is" without express
|
|---|
| 14 | ! * or implied warranty.
|
|---|
| 15 | ! *
|
|---|
| 16 | !
|
|---|
| 17 | ! Author(s): Markus Gaug 05/2004 <mailto:markus@ifae.es>
|
|---|
| 18 | !
|
|---|
| 19 | ! Copyright: MAGIC Software Development, 2002-2004
|
|---|
| 20 | !
|
|---|
| 21 | !
|
|---|
| 22 | \* ======================================================================== */
|
|---|
| 23 |
|
|---|
| 24 | //////////////////////////////////////////////////////////////////////////////
|
|---|
| 25 | //
|
|---|
| 26 | // MExtractTimeFastSpline
|
|---|
| 27 | //
|
|---|
| 28 | // Fast arrival Time extractor using a cubic spline algorithm of Numerical Recipes.
|
|---|
| 29 | // It returns the position of the half maximum between absolute maximum
|
|---|
| 30 | // and pedestal of the spline that interpolates the FADC slices.
|
|---|
| 31 | //
|
|---|
| 32 | // The precision of the half-maximum searches can be chosen by:
|
|---|
| 33 | // SetPrecision().
|
|---|
| 34 | //
|
|---|
| 35 | // The precision of the maximum-finder is fixed to 0.025 FADC units.
|
|---|
| 36 | //
|
|---|
| 37 | //////////////////////////////////////////////////////////////////////////////
|
|---|
| 38 | #include "MExtractTimeFastSpline.h"
|
|---|
| 39 |
|
|---|
| 40 | #include "MPedestalPix.h"
|
|---|
| 41 |
|
|---|
| 42 | #include "MLog.h"
|
|---|
| 43 | #include "MLogManip.h"
|
|---|
| 44 |
|
|---|
| 45 |
|
|---|
| 46 | ClassImp(MExtractTimeFastSpline);
|
|---|
| 47 |
|
|---|
| 48 | using namespace std;
|
|---|
| 49 |
|
|---|
| 50 | const Byte_t MExtractTimeFastSpline::fgHiGainFirst = 2;
|
|---|
| 51 | const Byte_t MExtractTimeFastSpline::fgHiGainLast = 14;
|
|---|
| 52 | const Byte_t MExtractTimeFastSpline::fgLoGainFirst = 3;
|
|---|
| 53 | const Byte_t MExtractTimeFastSpline::fgLoGainLast = 14;
|
|---|
| 54 | const Float_t MExtractTimeFastSpline::fgResolution = 0.003;
|
|---|
| 55 | const Float_t MExtractTimeFastSpline::fgRiseTime = 1.5;
|
|---|
| 56 | // --------------------------------------------------------------------------
|
|---|
| 57 | //
|
|---|
| 58 | // Default constructor.
|
|---|
| 59 | //
|
|---|
| 60 | // Calls:
|
|---|
| 61 | // - SetRange(fgHiGainFirst, fgHiGainLast, fgLoGainFirst, fgLoGainLast)
|
|---|
| 62 | //
|
|---|
| 63 | // Initializes:
|
|---|
| 64 | // - fResolution to fgResolution
|
|---|
| 65 | //
|
|---|
| 66 | MExtractTimeFastSpline::MExtractTimeFastSpline(const char *name, const char *title)
|
|---|
| 67 | : fHiGainFirstDeriv(NULL), fLoGainFirstDeriv(NULL),
|
|---|
| 68 | fHiGainSecondDeriv(NULL), fLoGainSecondDeriv(NULL)
|
|---|
| 69 | {
|
|---|
| 70 |
|
|---|
| 71 | fName = name ? name : "MExtractTimeFastSpline";
|
|---|
| 72 | fTitle = title ? title : "Calculate photons arrival time using a fast spline";
|
|---|
| 73 |
|
|---|
| 74 | SetResolution();
|
|---|
| 75 | SetRiseTime ();
|
|---|
| 76 | SetRange(fgHiGainFirst, fgHiGainLast, fgLoGainFirst, fgLoGainLast);
|
|---|
| 77 |
|
|---|
| 78 | }
|
|---|
| 79 |
|
|---|
| 80 | MExtractTimeFastSpline::~MExtractTimeFastSpline()
|
|---|
| 81 | {
|
|---|
| 82 |
|
|---|
| 83 | if (fHiGainFirstDeriv)
|
|---|
| 84 | delete fHiGainFirstDeriv;
|
|---|
| 85 | if (fLoGainFirstDeriv)
|
|---|
| 86 | delete fLoGainFirstDeriv;
|
|---|
| 87 | if (fHiGainSecondDeriv)
|
|---|
| 88 | delete fHiGainSecondDeriv;
|
|---|
| 89 | if (fLoGainSecondDeriv)
|
|---|
| 90 | delete fLoGainSecondDeriv;
|
|---|
| 91 |
|
|---|
| 92 | }
|
|---|
| 93 |
|
|---|
| 94 |
|
|---|
| 95 | // --------------------------------------------------------------------------
|
|---|
| 96 | //
|
|---|
| 97 | // SetRange:
|
|---|
| 98 | //
|
|---|
| 99 | // Calls:
|
|---|
| 100 | // - MExtractor::SetRange(hifirst,hilast,lofirst,lolast);
|
|---|
| 101 | // - Deletes x, if not NULL
|
|---|
| 102 | // - Creates x according to the range
|
|---|
| 103 | //
|
|---|
| 104 | void MExtractTimeFastSpline::SetRange(Byte_t hifirst, Byte_t hilast, Byte_t lofirst, Byte_t lolast)
|
|---|
| 105 | {
|
|---|
| 106 |
|
|---|
| 107 | MExtractor::SetRange(hifirst,hilast,lofirst,lolast);
|
|---|
| 108 |
|
|---|
| 109 | if (fHiGainFirstDeriv)
|
|---|
| 110 | delete fHiGainFirstDeriv;
|
|---|
| 111 | if (fLoGainFirstDeriv)
|
|---|
| 112 | delete fLoGainFirstDeriv;
|
|---|
| 113 | if (fHiGainSecondDeriv)
|
|---|
| 114 | delete fHiGainSecondDeriv;
|
|---|
| 115 | if (fLoGainSecondDeriv)
|
|---|
| 116 | delete fLoGainSecondDeriv;
|
|---|
| 117 |
|
|---|
| 118 | Int_t range = fHiGainLast - fHiGainFirst + 1;
|
|---|
| 119 |
|
|---|
| 120 | if (range < 2)
|
|---|
| 121 | {
|
|---|
| 122 | *fLog << warn << GetDescriptor()
|
|---|
| 123 | << Form("%s%2i%s%2i%s",": Hi-Gain Extraction range [",(int)fHiGainFirst,","
|
|---|
| 124 | ,fHiGainLast,"] too small, ") << endl;
|
|---|
| 125 | *fLog << warn << GetDescriptor()
|
|---|
| 126 | << " will move higher limit to obtain 4 slices " << endl;
|
|---|
| 127 | SetRange(fHiGainFirst, fHiGainLast+4-range,fLoGainFirst,fLoGainLast);
|
|---|
| 128 | range = fHiGainLast - fHiGainFirst + 1;
|
|---|
| 129 | }
|
|---|
| 130 |
|
|---|
| 131 |
|
|---|
| 132 | fHiGainFirstDeriv = new Float_t[range];
|
|---|
| 133 | memset(fHiGainFirstDeriv,0,range*sizeof(Float_t));
|
|---|
| 134 | fHiGainSecondDeriv = new Float_t[range];
|
|---|
| 135 | memset(fHiGainSecondDeriv,0,range*sizeof(Float_t));
|
|---|
| 136 |
|
|---|
| 137 | range = fLoGainLast - fLoGainFirst + 1;
|
|---|
| 138 |
|
|---|
| 139 | if (range >= 2)
|
|---|
| 140 | {
|
|---|
| 141 |
|
|---|
| 142 | fLoGainFirstDeriv = new Float_t[range];
|
|---|
| 143 | memset(fLoGainFirstDeriv,0,range*sizeof(Float_t));
|
|---|
| 144 | fLoGainSecondDeriv = new Float_t[range];
|
|---|
| 145 | memset(fLoGainSecondDeriv,0,range*sizeof(Float_t));
|
|---|
| 146 |
|
|---|
| 147 | }
|
|---|
| 148 |
|
|---|
| 149 | }
|
|---|
| 150 |
|
|---|
| 151 |
|
|---|
| 152 | // --------------------------------------------------------------------------
|
|---|
| 153 | //
|
|---|
| 154 | // Calculates the arrival time for each pixel
|
|---|
| 155 | //
|
|---|
| 156 | void MExtractTimeFastSpline::FindTimeHiGain(Byte_t *first, Float_t &time, Float_t &dtime,
|
|---|
| 157 | Byte_t &sat, const MPedestalPix &ped) const
|
|---|
| 158 | {
|
|---|
| 159 |
|
|---|
| 160 | const Int_t range = fHiGainLast - fHiGainFirst + 1;
|
|---|
| 161 | const Byte_t *end = first + range;
|
|---|
| 162 | Byte_t *p = first;
|
|---|
| 163 | Byte_t max = 0;
|
|---|
| 164 | Byte_t maxpos = 0;
|
|---|
| 165 |
|
|---|
| 166 | //
|
|---|
| 167 | // Check for saturation in all other slices
|
|---|
| 168 | //
|
|---|
| 169 | while (++p<end)
|
|---|
| 170 | {
|
|---|
| 171 | if (*p > max)
|
|---|
| 172 | {
|
|---|
| 173 | max = *p;
|
|---|
| 174 | maxpos = p-first;
|
|---|
| 175 | }
|
|---|
| 176 |
|
|---|
| 177 | if (*p >= fSaturationLimit)
|
|---|
| 178 | {
|
|---|
| 179 | sat++;
|
|---|
| 180 | break;
|
|---|
| 181 | }
|
|---|
| 182 | }
|
|---|
| 183 |
|
|---|
| 184 | if (sat)
|
|---|
| 185 | return;
|
|---|
| 186 |
|
|---|
| 187 | if (maxpos < 2)
|
|---|
| 188 | return;
|
|---|
| 189 |
|
|---|
| 190 | Float_t pp;
|
|---|
| 191 |
|
|---|
| 192 | p = first;
|
|---|
| 193 | fHiGainSecondDeriv[0] = 0.;
|
|---|
| 194 | fHiGainFirstDeriv[0] = 0.;
|
|---|
| 195 |
|
|---|
| 196 | for (Int_t i=1;i<range-1;i++)
|
|---|
| 197 | {
|
|---|
| 198 | p++;
|
|---|
| 199 | pp = fHiGainSecondDeriv[i-1] + 4.;
|
|---|
| 200 | fHiGainSecondDeriv[i] = -1.0/pp;
|
|---|
| 201 | fHiGainFirstDeriv [i] = *(p+1) - 2.* *(p) + *(p-1);
|
|---|
| 202 | fHiGainFirstDeriv [i] = (6.0*fHiGainFirstDeriv[i]-fHiGainFirstDeriv[i-1])/pp;
|
|---|
| 203 | }
|
|---|
| 204 |
|
|---|
| 205 | fHiGainSecondDeriv[range-1] = 0.;
|
|---|
| 206 |
|
|---|
| 207 | for (Int_t k=range-2;k>0;k--)
|
|---|
| 208 | fHiGainSecondDeriv[k] = fHiGainSecondDeriv[k]*fHiGainSecondDeriv[k+1] + fHiGainFirstDeriv[k];
|
|---|
| 209 | for (Int_t k=range-2;k>0;k--)
|
|---|
| 210 | fHiGainSecondDeriv[k] /= 6.;
|
|---|
| 211 |
|
|---|
| 212 | //
|
|---|
| 213 | // Now find the maximum
|
|---|
| 214 | //
|
|---|
| 215 | Float_t step = 0.2; // start with step size of 1ns and loop again with the smaller one
|
|---|
| 216 | Float_t lower = (Float_t)maxpos-1.;
|
|---|
| 217 | Float_t upper = (Float_t)maxpos;
|
|---|
| 218 | Float_t x = lower;
|
|---|
| 219 | Float_t y = 0.;
|
|---|
| 220 | Float_t a = 1.;
|
|---|
| 221 | Float_t b = 0.;
|
|---|
| 222 | Int_t klo = maxpos-1;
|
|---|
| 223 | Int_t khi = maxpos;
|
|---|
| 224 | Float_t klocont = (Float_t)*(first+klo);
|
|---|
| 225 | Float_t khicont = (Float_t)*(first+khi);
|
|---|
| 226 | time = upper;
|
|---|
| 227 | Float_t abmax = khicont;
|
|---|
| 228 |
|
|---|
| 229 | //
|
|---|
| 230 | // Search for the maximum, starting in interval maxpos-1. If no maximum is found, go to
|
|---|
| 231 | // interval maxpos+1.
|
|---|
| 232 | //
|
|---|
| 233 | while (x<upper-0.3)
|
|---|
| 234 | {
|
|---|
| 235 |
|
|---|
| 236 | x += step;
|
|---|
| 237 | a -= step;
|
|---|
| 238 | b += step;
|
|---|
| 239 |
|
|---|
| 240 | y = a*klocont
|
|---|
| 241 | + b*khicont
|
|---|
| 242 | + (a*a*a-a)*fHiGainSecondDeriv[klo]
|
|---|
| 243 | + (b*b*b-b)*fHiGainSecondDeriv[khi];
|
|---|
| 244 |
|
|---|
| 245 | if (y > abmax)
|
|---|
| 246 | {
|
|---|
| 247 | abmax = y;
|
|---|
| 248 | time = x;
|
|---|
| 249 | }
|
|---|
| 250 | }
|
|---|
| 251 |
|
|---|
| 252 |
|
|---|
| 253 | if (time > upper-0.1)
|
|---|
| 254 | {
|
|---|
| 255 |
|
|---|
| 256 | upper = (Float_t)maxpos+1.;
|
|---|
| 257 | lower = (Float_t)maxpos;
|
|---|
| 258 | x = lower;
|
|---|
| 259 | a = 1.;
|
|---|
| 260 | b = 0.;
|
|---|
| 261 | khi = maxpos+1;
|
|---|
| 262 | klo = maxpos;
|
|---|
| 263 | klocont = (Float_t)*(first+klo);
|
|---|
| 264 | khicont = (Float_t)*(first+khi);
|
|---|
| 265 |
|
|---|
| 266 | while (x<upper-0.3)
|
|---|
| 267 | {
|
|---|
| 268 |
|
|---|
| 269 | x += step;
|
|---|
| 270 | a -= step;
|
|---|
| 271 | b += step;
|
|---|
| 272 |
|
|---|
| 273 | y = a* klocont
|
|---|
| 274 | + b* khicont
|
|---|
| 275 | + (a*a*a-a)*fHiGainSecondDeriv[klo]
|
|---|
| 276 | + (b*b*b-b)*fHiGainSecondDeriv[khi];
|
|---|
| 277 |
|
|---|
| 278 | if (y > abmax)
|
|---|
| 279 | {
|
|---|
| 280 | abmax = y;
|
|---|
| 281 | time = x;
|
|---|
| 282 | }
|
|---|
| 283 | }
|
|---|
| 284 | }
|
|---|
| 285 |
|
|---|
| 286 | const Float_t up = time+step-0.055;
|
|---|
| 287 | const Float_t lo = time-step+0.055;
|
|---|
| 288 | const Float_t maxpossave = time;
|
|---|
| 289 |
|
|---|
| 290 | x = time;
|
|---|
| 291 | a = upper - x;
|
|---|
| 292 | b = x - lower;
|
|---|
| 293 |
|
|---|
| 294 | step = 0.04; // step size of 83 ps
|
|---|
| 295 |
|
|---|
| 296 | while (x<up)
|
|---|
| 297 | {
|
|---|
| 298 |
|
|---|
| 299 | x += step;
|
|---|
| 300 | a -= step;
|
|---|
| 301 | b += step;
|
|---|
| 302 |
|
|---|
| 303 | y = a* klocont
|
|---|
| 304 | + b* khicont
|
|---|
| 305 | + (a*a*a-a)*fHiGainSecondDeriv[klo]
|
|---|
| 306 | + (b*b*b-b)*fHiGainSecondDeriv[khi];
|
|---|
| 307 |
|
|---|
| 308 | if (y > abmax)
|
|---|
| 309 | {
|
|---|
| 310 | abmax = y;
|
|---|
| 311 | time = x;
|
|---|
| 312 | }
|
|---|
| 313 |
|
|---|
| 314 | }
|
|---|
| 315 |
|
|---|
| 316 | if (time < klo + 0.02)
|
|---|
| 317 | {
|
|---|
| 318 | klo--;
|
|---|
| 319 | khi--;
|
|---|
| 320 | klocont = (Float_t)*(first+klo);
|
|---|
| 321 | khicont = (Float_t)*(first+khi);
|
|---|
| 322 | upper--;
|
|---|
| 323 | lower--;
|
|---|
| 324 | }
|
|---|
| 325 |
|
|---|
| 326 | x = maxpossave;
|
|---|
| 327 | a = upper - x;
|
|---|
| 328 | b = x - lower;
|
|---|
| 329 |
|
|---|
| 330 | while (x>lo)
|
|---|
| 331 | {
|
|---|
| 332 |
|
|---|
| 333 | x -= step;
|
|---|
| 334 | a += step;
|
|---|
| 335 | b -= step;
|
|---|
| 336 |
|
|---|
| 337 | y = a* klocont
|
|---|
| 338 | + b* khicont
|
|---|
| 339 | + (a*a*a-a)*fHiGainSecondDeriv[klo]
|
|---|
| 340 | + (b*b*b-b)*fHiGainSecondDeriv[khi];
|
|---|
| 341 |
|
|---|
| 342 | if (y > abmax)
|
|---|
| 343 | {
|
|---|
| 344 | abmax = y;
|
|---|
| 345 | time = x;
|
|---|
| 346 | }
|
|---|
| 347 | }
|
|---|
| 348 |
|
|---|
| 349 | const Float_t pedes = ped.GetPedestal();
|
|---|
| 350 | const Float_t halfmax = pedes + (abmax - pedes)/2.;
|
|---|
| 351 |
|
|---|
| 352 | //
|
|---|
| 353 | // Now, loop from the maximum bin leftward down in order to find the position of the half maximum.
|
|---|
| 354 | // First, find the right FADC slice:
|
|---|
| 355 | //
|
|---|
| 356 | klo = maxpos;
|
|---|
| 357 | while (klo > maxpos-fStartBeforeMax)
|
|---|
| 358 | {
|
|---|
| 359 | if (*(first+klo) < (Byte_t)halfmax)
|
|---|
| 360 | break;
|
|---|
| 361 | klo--;
|
|---|
| 362 | }
|
|---|
| 363 |
|
|---|
| 364 | //
|
|---|
| 365 | // Loop from the beginning of the slice upwards to reach the halfmax:
|
|---|
| 366 | // With means of bisection:
|
|---|
| 367 | //
|
|---|
| 368 | x = (Float_t)klo;
|
|---|
| 369 | a = 1.;
|
|---|
| 370 | b = 0.;
|
|---|
| 371 | klocont = (Float_t)*(first+klo);
|
|---|
| 372 | khicont = (Float_t)*(first+klo+1);
|
|---|
| 373 | time = x;
|
|---|
| 374 |
|
|---|
| 375 | step = 0.5;
|
|---|
| 376 | Bool_t back = kFALSE;
|
|---|
| 377 |
|
|---|
| 378 | while (step > fResolution)
|
|---|
| 379 | {
|
|---|
| 380 |
|
|---|
| 381 | if (back)
|
|---|
| 382 | {
|
|---|
| 383 | x -= step;
|
|---|
| 384 | a += step;
|
|---|
| 385 | b -= step;
|
|---|
| 386 | }
|
|---|
| 387 | else
|
|---|
| 388 | {
|
|---|
| 389 | x += step;
|
|---|
| 390 | a -= step;
|
|---|
| 391 | b += step;
|
|---|
| 392 | }
|
|---|
| 393 |
|
|---|
| 394 | y = a*klocont
|
|---|
| 395 | + b*khicont
|
|---|
| 396 | + (a*a*a-a)*fHiGainSecondDeriv[klo]
|
|---|
| 397 | + (b*b*b-b)*fHiGainSecondDeriv[khi];
|
|---|
| 398 |
|
|---|
| 399 | if (y >= halfmax)
|
|---|
| 400 | back = kTRUE;
|
|---|
| 401 | else
|
|---|
| 402 | back = kFALSE;
|
|---|
| 403 |
|
|---|
| 404 | step /= 2.;
|
|---|
| 405 |
|
|---|
| 406 | }
|
|---|
| 407 |
|
|---|
| 408 | time = (Float_t)fHiGainFirst + x;
|
|---|
| 409 | dtime = fResolution;
|
|---|
| 410 | }
|
|---|
| 411 |
|
|---|
| 412 |
|
|---|
| 413 | // --------------------------------------------------------------------------
|
|---|
| 414 | //
|
|---|
| 415 | // Calculates the arrival time for each pixel
|
|---|
| 416 | //
|
|---|
| 417 | void MExtractTimeFastSpline::FindTimeLoGain(Byte_t *first, Float_t &time, Float_t &dtime,
|
|---|
| 418 | Byte_t &sat, const MPedestalPix &ped) const
|
|---|
| 419 | {
|
|---|
| 420 |
|
|---|
| 421 | const Int_t range = fLoGainLast - fLoGainFirst + 1;
|
|---|
| 422 | const Byte_t *end = first + range;
|
|---|
| 423 | Byte_t *p = first;
|
|---|
| 424 | Byte_t max = 0;
|
|---|
| 425 | Byte_t maxpos = 0;
|
|---|
| 426 |
|
|---|
| 427 | //
|
|---|
| 428 | // Check for saturation in all other slices
|
|---|
| 429 | //
|
|---|
| 430 | while (++p<end)
|
|---|
| 431 | {
|
|---|
| 432 | if (*p > max)
|
|---|
| 433 | {
|
|---|
| 434 | max = *p;
|
|---|
| 435 | maxpos = p-first;
|
|---|
| 436 | }
|
|---|
| 437 |
|
|---|
| 438 | if (*p >= fSaturationLimit)
|
|---|
| 439 | {
|
|---|
| 440 | sat++;
|
|---|
| 441 | break;
|
|---|
| 442 | }
|
|---|
| 443 | }
|
|---|
| 444 |
|
|---|
| 445 | if (sat)
|
|---|
| 446 | return;
|
|---|
| 447 |
|
|---|
| 448 | if (maxpos < 2)
|
|---|
| 449 | return;
|
|---|
| 450 |
|
|---|
| 451 | Float_t pp;
|
|---|
| 452 |
|
|---|
| 453 | p = first;
|
|---|
| 454 | fLoGainSecondDeriv[0] = 0.;
|
|---|
| 455 | fLoGainFirstDeriv[0] = 0.;
|
|---|
| 456 |
|
|---|
| 457 | for (Int_t i=1;i<range-1;i++)
|
|---|
| 458 | {
|
|---|
| 459 | p++;
|
|---|
| 460 | pp = fLoGainSecondDeriv[i-1] + 4.;
|
|---|
| 461 | fLoGainSecondDeriv[i] = -1.0/pp;
|
|---|
| 462 | fLoGainFirstDeriv [i] = *(p+1) - 2.* *(p) + *(p-1);
|
|---|
| 463 | fLoGainFirstDeriv [i] = (6.0*fLoGainFirstDeriv[i]-fLoGainFirstDeriv[i-1])/pp;
|
|---|
| 464 | }
|
|---|
| 465 |
|
|---|
| 466 | fLoGainSecondDeriv[range-1] = 0.;
|
|---|
| 467 |
|
|---|
| 468 | for (Int_t k=range-2;k>0;k--)
|
|---|
| 469 | fLoGainSecondDeriv[k] = fLoGainSecondDeriv[k]*fLoGainSecondDeriv[k+1] + fLoGainFirstDeriv[k];
|
|---|
| 470 | for (Int_t k=range-2;k>0;k--)
|
|---|
| 471 | fLoGainSecondDeriv[k] /= 6.;
|
|---|
| 472 |
|
|---|
| 473 | //
|
|---|
| 474 | // Now find the maximum
|
|---|
| 475 | //
|
|---|
| 476 | Float_t step = 0.2; // start with step size of 1ns and loop again with the smaller one
|
|---|
| 477 | Float_t lower = (Float_t)maxpos-1.;
|
|---|
| 478 | Float_t upper = (Float_t)maxpos;
|
|---|
| 479 | Float_t x = lower;
|
|---|
| 480 | Float_t y = 0.;
|
|---|
| 481 | Float_t a = 1.;
|
|---|
| 482 | Float_t b = 0.;
|
|---|
| 483 | Int_t klo = maxpos-1;
|
|---|
| 484 | Int_t khi = maxpos;
|
|---|
| 485 | Float_t klocont = (Float_t)*(first+klo);
|
|---|
| 486 | Float_t khicont = (Float_t)*(first+khi);
|
|---|
| 487 | time = upper;
|
|---|
| 488 | Float_t abmax = khicont;
|
|---|
| 489 |
|
|---|
| 490 | //
|
|---|
| 491 | // Search for the maximum, starting in interval maxpos-1. If no maximum is found, go to
|
|---|
| 492 | // interval maxpos+1.
|
|---|
| 493 | //
|
|---|
| 494 | while (x<upper-0.3)
|
|---|
| 495 | {
|
|---|
| 496 |
|
|---|
| 497 | x += step;
|
|---|
| 498 | a -= step;
|
|---|
| 499 | b += step;
|
|---|
| 500 |
|
|---|
| 501 | y = a*klocont
|
|---|
| 502 | + b*khicont
|
|---|
| 503 | + (a*a*a-a)*fLoGainSecondDeriv[klo]
|
|---|
| 504 | + (b*b*b-b)*fLoGainSecondDeriv[khi];
|
|---|
| 505 |
|
|---|
| 506 | if (y > abmax)
|
|---|
| 507 | {
|
|---|
| 508 | abmax = y;
|
|---|
| 509 | time = x;
|
|---|
| 510 | }
|
|---|
| 511 |
|
|---|
| 512 | }
|
|---|
| 513 |
|
|---|
| 514 | if (time > upper-0.1)
|
|---|
| 515 | {
|
|---|
| 516 |
|
|---|
| 517 | upper = (Float_t)maxpos+1.;
|
|---|
| 518 | lower = (Float_t)maxpos;
|
|---|
| 519 | x = lower;
|
|---|
| 520 | a = 1.;
|
|---|
| 521 | b = 0.;
|
|---|
| 522 | khi = maxpos+1;
|
|---|
| 523 | klo = maxpos;
|
|---|
| 524 | klocont = (Float_t)*(first+klo);
|
|---|
| 525 | khicont = (Float_t)*(first+khi);
|
|---|
| 526 |
|
|---|
| 527 | while (x<upper-0.3)
|
|---|
| 528 | {
|
|---|
| 529 |
|
|---|
| 530 | x += step;
|
|---|
| 531 | a -= step;
|
|---|
| 532 | b += step;
|
|---|
| 533 |
|
|---|
| 534 | y = a* klocont
|
|---|
| 535 | + b* khicont
|
|---|
| 536 | + (a*a*a-a)*fLoGainSecondDeriv[klo]
|
|---|
| 537 | + (b*b*b-b)*fLoGainSecondDeriv[khi];
|
|---|
| 538 |
|
|---|
| 539 | if (y > abmax)
|
|---|
| 540 | {
|
|---|
| 541 | abmax = y;
|
|---|
| 542 | time = x;
|
|---|
| 543 | }
|
|---|
| 544 | }
|
|---|
| 545 | }
|
|---|
| 546 |
|
|---|
| 547 | const Float_t up = time+step-0.055;
|
|---|
| 548 | const Float_t lo = time-step+0.055;
|
|---|
| 549 | const Float_t maxpossave = time;
|
|---|
| 550 |
|
|---|
| 551 | x = time;
|
|---|
| 552 | a = upper - x;
|
|---|
| 553 | b = x - lower;
|
|---|
| 554 |
|
|---|
| 555 | step = 0.025; // step size of 165 ps
|
|---|
| 556 |
|
|---|
| 557 | while (x<up)
|
|---|
| 558 | {
|
|---|
| 559 |
|
|---|
| 560 | x += step;
|
|---|
| 561 | a -= step;
|
|---|
| 562 | b += step;
|
|---|
| 563 |
|
|---|
| 564 | y = a* klocont
|
|---|
| 565 | + b* khicont
|
|---|
| 566 | + (a*a*a-a)*fLoGainSecondDeriv[klo]
|
|---|
| 567 | + (b*b*b-b)*fLoGainSecondDeriv[khi];
|
|---|
| 568 |
|
|---|
| 569 | if (y > abmax)
|
|---|
| 570 | {
|
|---|
| 571 | abmax = y;
|
|---|
| 572 | time = x;
|
|---|
| 573 | }
|
|---|
| 574 |
|
|---|
| 575 | }
|
|---|
| 576 |
|
|---|
| 577 | if (time < klo + 0.01)
|
|---|
| 578 | {
|
|---|
| 579 | klo--;
|
|---|
| 580 | khi--;
|
|---|
| 581 | klocont = (Float_t)*(first+klo);
|
|---|
| 582 | khicont = (Float_t)*(first+khi);
|
|---|
| 583 | upper--;
|
|---|
| 584 | lower--;
|
|---|
| 585 | }
|
|---|
| 586 |
|
|---|
| 587 | x = maxpossave;
|
|---|
| 588 | a = upper - x;
|
|---|
| 589 | b = x - lower;
|
|---|
| 590 |
|
|---|
| 591 | while (x>lo)
|
|---|
| 592 | {
|
|---|
| 593 |
|
|---|
| 594 | x -= step;
|
|---|
| 595 | a += step;
|
|---|
| 596 | b -= step;
|
|---|
| 597 |
|
|---|
| 598 | y = a* klocont
|
|---|
| 599 | + b* khicont
|
|---|
| 600 | + (a*a*a-a)*fLoGainSecondDeriv[klo]
|
|---|
| 601 | + (b*b*b-b)*fLoGainSecondDeriv[khi];
|
|---|
| 602 |
|
|---|
| 603 | if (y > abmax)
|
|---|
| 604 | {
|
|---|
| 605 | abmax = y;
|
|---|
| 606 | time = x;
|
|---|
| 607 | }
|
|---|
| 608 |
|
|---|
| 609 | }
|
|---|
| 610 |
|
|---|
| 611 | const Float_t pedes = ped.GetPedestal();
|
|---|
| 612 | const Float_t halfmax = pedes + (abmax - pedes)/2.;
|
|---|
| 613 |
|
|---|
| 614 | //
|
|---|
| 615 | // Now, loop from the maximum bin leftward down in order to find the position of the half maximum.
|
|---|
| 616 | // First, find the right FADC slice:
|
|---|
| 617 | //
|
|---|
| 618 | klo = maxpos;
|
|---|
| 619 | while (klo > maxpos-4)
|
|---|
| 620 | {
|
|---|
| 621 | if (*(first+klo) < (Byte_t)halfmax)
|
|---|
| 622 | break;
|
|---|
| 623 | klo--;
|
|---|
| 624 | }
|
|---|
| 625 |
|
|---|
| 626 | //
|
|---|
| 627 | // Loop from the beginning of the slice upwards to reach the halfmax:
|
|---|
| 628 | // With means of bisection:
|
|---|
| 629 | //
|
|---|
| 630 | x = (Float_t)klo;
|
|---|
| 631 | a = 1.;
|
|---|
| 632 | b = 0.;
|
|---|
| 633 | klocont = (Float_t)*(first+klo);
|
|---|
| 634 | khicont = (Float_t)*(first+klo+1);
|
|---|
| 635 | time = x;
|
|---|
| 636 |
|
|---|
| 637 | step = 0.5;
|
|---|
| 638 | Bool_t back = kFALSE;
|
|---|
| 639 |
|
|---|
| 640 | while (step > fResolution)
|
|---|
| 641 | {
|
|---|
| 642 |
|
|---|
| 643 | if (back)
|
|---|
| 644 | {
|
|---|
| 645 | x -= step;
|
|---|
| 646 | a += step;
|
|---|
| 647 | b -= step;
|
|---|
| 648 | }
|
|---|
| 649 | else
|
|---|
| 650 | {
|
|---|
| 651 | x += step;
|
|---|
| 652 | a -= step;
|
|---|
| 653 | b += step;
|
|---|
| 654 | }
|
|---|
| 655 |
|
|---|
| 656 | y = a*klocont
|
|---|
| 657 | + b*khicont
|
|---|
| 658 | + (a*a*a-a)*fLoGainSecondDeriv[klo]
|
|---|
| 659 | + (b*b*b-b)*fLoGainSecondDeriv[khi];
|
|---|
| 660 |
|
|---|
| 661 | if (y >= halfmax)
|
|---|
| 662 | back = kTRUE;
|
|---|
| 663 | else
|
|---|
| 664 | back = kFALSE;
|
|---|
| 665 |
|
|---|
| 666 | step /= 2.;
|
|---|
| 667 |
|
|---|
| 668 | }
|
|---|
| 669 |
|
|---|
| 670 | time = (Float_t)fLoGainFirst + x;
|
|---|
| 671 | dtime = fResolution;
|
|---|
| 672 | }
|
|---|
| 673 |
|
|---|
| 674 |
|
|---|