1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457
|
/*!
* \file
* \brief Transforms test program
* \author Tony Ottosson, Thomas Eriksson, Simon Wood, Adam Piatyszek, Andy Panov and Bogdan Cristea
*
* -------------------------------------------------------------------------
*
* Copyright (C) 1995-2013 (see AUTHORS file for a list of contributors)
*
* This file is part of IT++ - a C++ library of mathematical, signal
* processing, speech processing, and communications classes and functions.
*
* IT++ is free software: you can redistribute it and/or modify it under the
* terms of the GNU General Public License as published by the Free Software
* Foundation, either version 3 of the License, or (at your option) any
* later version.
*
* IT++ is distributed in the hope that it will be useful, but WITHOUT ANY
* WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
* FOR A PARTICULAR PURPOSE. See the GNU General Public License for more
* details.
*
* You should have received a copy of the GNU General Public License along
* with IT++. If not, see <http://www.gnu.org/licenses/>.
*
* -------------------------------------------------------------------------
*/
#include <vector>
#include <itpp/itsignal.h>
#include "gtest/gtest.h"
#ifdef _OPENMP
#include <omp.h>
#endif
using namespace itpp;
//set test tolerance (measure of relative and absolute error)
const double max_rel_error = 1e-6;
const double max_abs_error = 1e-6;
//transform results tester
template<typename T>
inline void test_result(const Vec<T>& in, const Vec<T>& ref)
{
int n = size(in);
it_assert(n == ref.size(), "compute_rel_error(): input and reference sizes must be equal.");
for(int i = 0; i < n ; ++i) {
if(abs(in(i) - ref(i)) < max_abs_error) continue; //handle numbers with absolute value close to zero (relative error can be huge for them)
double rel_error = abs(in(i) - ref(i)) / abs(in(i));
ASSERT_LE(rel_error, max_rel_error);
}
}
//tester for multiple results collected in std::vector
template<typename T>
inline void test_result(const std::vector<Vec<T> >& in, const Vec<T>& ref)
{
typename std::vector<Vec<T> >::size_type i;
for(i = 0; i < in.size(); ++i) test_result(in[i], ref);
}
//Reference transform implementations. These functions are not intended to be fast.
//They just strictly follow the transform definitions
//reference DFT implementation
template<typename T>
inline cvec ref_dft(const Vec<T>& in)
{
int n = size(in);
it_assert(n > 0, "ref_dft(): zero-sized input detected.");
cvec ret(n);
for(int i = 0; i < n; ++i) {
std::complex<double> res = 0.0;
for(int j = 0; j < n; ++j) {
res += std::complex<double>(cos(2 * pi * i * j / n), -sin(2 * pi * i * j / n)) * in(j);
}
ret(i) = res;
}
return ret;
}
//reference IDFT implementation
inline cvec ref_idft(const cvec& in)
{
int n = size(in);
it_assert(n > 0, "ref_idft(): zero-sized input detected.");
cvec ret(n);
for(int i = 0; i < n; ++i) {
std::complex<double> res = 0.0;
for(int j = 0; j < n; ++j) {
res += std::complex<double>(cos(2 * pi * i * j / n), sin(2 * pi * i * j / n)) * in(j);
}
ret(i) = res;
}
ret *= 1.0 / n;
return ret;
}
//Type-II DCT reference implementation
inline vec ref_dct(const vec& in)
{
int n = size(in);
it_assert(n > 0, "ref_dct(): zero-sized input detected.");
vec ret(n);
for(int i = 0; i < n; ++i) {
double res = 0.0;
for(int j = 0; j < n; ++j) {
res += cos(pi * (j + 0.5) * i / n) * in(j);
}
ret(i) = 2 * res;
}
// Scale to matlab definition format
ret /= std::sqrt(2.0 * n);
ret(0) /= std::sqrt(2.0);
return ret;
}
//Type-III DCT (IDCT) reference implementation
inline vec ref_idct(const vec& in)
{
int n = size(in);
it_assert(n > 0, "ref_dct(): zero-sized input detected.");
vec tmp = in;
tmp(0) *= std::sqrt(2.0);
tmp /= std::sqrt(2.0 * n);
vec ret(n);
for(int i = 0; i < n; ++i) {
double res = 0.0;
for(int j = 1; j < n; ++j) {
res += cos(pi * (i + 0.5) * j / n) * tmp(j);
}
ret(i) = 2 * res + tmp(0);
}
return ret;
}
//Transform testers run the same transform function in single or multiple threads (if OMP is enabled).
template<typename InType, typename OutType>
std::vector<OutType> run_transform_test(OutType(*transform_function)(const InType&), const InType& test_input)
{
//select number of results. Multiple threads run the same code if OMP is enabled
#ifdef _OPENMP
static const int threads_cnt = omp_get_max_threads();
omp_set_num_threads(threads_cnt);
#else
static const int threads_cnt = 1;
#endif
std::vector<OutType> ret(threads_cnt);
#pragma omp parallel
{
//parallel region start. Spawn the threads.
#pragma omp for
for(int j = 0; j < threads_cnt; ++j) {
ret[j] = transform_function(test_input);
}
//parallel region end. Join the threads.
}
return ret;
}
template<typename InType, typename OutType>
std::vector<OutType> run_transform_test(OutType(*transform_function)(const InType&, int), const InType& test_input, int N)
{
//select number of results. Multiple threads run the same code if OMP is enabled
#ifdef _OPENMP
static const int threads_cnt = omp_get_max_threads();
omp_set_num_threads(threads_cnt);
#else
static const int threads_cnt = 1;
#endif
std::vector<OutType> ret(threads_cnt);
#pragma omp parallel
{
//parallel region start. Spawn the threads.
#pragma omp for
for(int j = 0; j < threads_cnt; ++j) {
ret[j] = transform_function(test_input, N);
}
//parallel region end. Join the threads.
}
return ret;
}
//----------------------------------------------
//Gtest test cases
//----------------------------------------------
TEST(Transforms, FFTReal)
{
int N = 16;
vec x = randn(N);
if(!have_fourier_transforms()) FAIL() << "Fourier Transforms are not supported with this library build.";
//vector processing test
{
SCOPED_TRACE("y = fft_real(x) test");
//cvec y = fft_real(x);
std::vector<cvec> y = run_transform_test(fft_real, x);
test_result(y, ref_dft(x));
}
//subvector processing test
{
SCOPED_TRACE("y = fft_real(x, N) test, N < length(x)");
int N_sub = 11; //odd subvector length
//cvec y = fft_real(x, N_sub);
std::vector<cvec> y = run_transform_test(fft_real, x, N_sub);
test_result(y, ref_dft(x(0, N_sub - 1)));
}
//zero-padded vector processing test
{
SCOPED_TRACE("y = fft_real(x, N) test, N > length(x)");
int N_zp = 8; //zero-padding length
//cvec y = fft_real(x, N + N_zp);
std::vector<cvec> y = run_transform_test(fft_real, x, N + N_zp);
x.set_size(N + N_zp, true);
test_result(y, ref_dft(x));
}
}
TEST(Transforms, IFFTReal)
{
if(!have_fourier_transforms()) FAIL() << "Fourier Transforms are not supported with this library build.";
//vector processing test
{
SCOPED_TRACE("y = ifft_real(x) test");
int N = 16;
cvec t = randn_c(N - 1);
cvec x(N);
//generate test complex sequence with real spectra
x.set_subvector(1, 0.5 * (t + conj(reverse(t))));
x(0) = randn();
//run transform & test results
//vec y = ifft_real(x);
std::vector<vec> y = run_transform_test(ifft_real, x);
test_result(y, real(ref_idft(x)));
}
//subvector processing test
{
SCOPED_TRACE("y = ifft_real(x, N) test, N < length(x)");
int N = 16, N_sub = 11; //define odd subvector length to test odd-length transform
cvec t = randn_c(N - 1);
cvec x(N);
//fill subvector samples with Hermitian sequence
x.set_subvector(1, 0.5 * (t(1, N_sub - 1) + conj(reverse(t(1, N_sub - 1)))));
x(0) = randn();
//fill the rest of x with random data (these data should be ignored by IFFT implementation)
x.set_subvector(N_sub, randn_c(N - N_sub));
//run transform & test results
//vec y = ifft_real(x, N_sub);
std::vector<vec> y = run_transform_test(ifft_real, x, N_sub);
test_result(y, real(ref_idft(x(0, N_sub - 1))));
}
//zero-padded vector processing test
{
SCOPED_TRACE("y = ifft_real(x, N) test, N > length(x)");
int N_data = 32, N_zp = 8; //define data and zero-padding length
cvec t = randn_c(N_data);
cvec x(N_data + N_zp + 1);
//generate the test data. sequence posesses Hermitian symmetry after zero-padding with N_zp zeros.
x(0) = randn();
x.set_subvector(1, N_zp, std::complex<double>(0));
x.set_subvector(N_zp + 1, 0.5 * (t + conj(reverse(t))));
//run transform & test results
//vec y = ifft_real(x, N_data + 2*N_zp + 1);
std::vector<vec> y = run_transform_test(ifft_real, x, N_data + 2 * N_zp + 1);
x.set_size(N_data + 2 * N_zp + 1, true);
test_result(y, real(ref_idft(x)));
}
}
TEST(Transforms, FFTCplx)
{
if(!have_fourier_transforms()) FAIL() << "Fourier Transforms are not supported with this library build.";
int N = 16;
cvec x = randn_c(N);
//vector processing test
{
SCOPED_TRACE("y = fft(x) test");
//cvec y = fft(x);
std::vector<cvec> y = run_transform_test(fft, x);
test_result(y, ref_dft(x));
}
//subvector processing test
{
SCOPED_TRACE("y = fft(x, N) test, N < length(x)");
int N_sub = 11; //odd subvector length
//cvec y = fft(x, N_sub);
std::vector<cvec> y = run_transform_test(fft, x, N_sub);
test_result(y, ref_dft(x(0, N_sub - 1)));
}
//zero-padded vector processing test
{
SCOPED_TRACE("y = fft(x, N) test, N > length(x)");
int N_zp = 8; //zero-padding length
//cvec y = fft(x, N + N_zp);
std::vector<cvec> y = run_transform_test(fft, x, N + N_zp);
x.set_size(N + N_zp, true);
test_result(y, ref_dft(x));
}
}
TEST(Transforms, IFFTCplx)
{
if(!have_fourier_transforms()) FAIL() << "Fourier Transforms are not supported with this library build.";
int N = 16;
cvec x = randn_c(N), y;
//vector processing test
{
SCOPED_TRACE("y = ifft(x) test");
//cvec y = ifft(x);
std::vector<cvec> y = run_transform_test(ifft, x);
test_result(y, ref_idft(x));
}
//subvector processing test
{
SCOPED_TRACE("y = ifft(x, N) test, N < length(x)");
int N_sub = 11; //odd subvector length
//cvec y = ifft(x, N_sub);
std::vector<cvec> y = run_transform_test(ifft, x, N_sub);
test_result(y, ref_idft(x(0, N_sub - 1)));
}
//zero-padded vector processing test
{
SCOPED_TRACE("y = ifft(x, N) test, N > length(x)");
int N_zp = 8; //zero-padding length
//cvec y = ifft(x, N + N_zp);
std::vector<cvec> y = run_transform_test(ifft, x, N + N_zp);
x.set_size(N + N_zp, true);
test_result(y, ref_idft(x));
}
}
TEST(Transforms, DCT)
{
if(!have_cosine_transforms()) FAIL() << "Cosine Transforms are not supported with this library build.";
int N = 16;
vec x = randn(N);
//vector processing test
{
SCOPED_TRACE("y = dct(x) test");
//vec y = dct(x);
std::vector<vec> y = run_transform_test(dct, x);
test_result(y, ref_dct(x));
}
//subvector processing test
{
SCOPED_TRACE("y = dct(x, N) test, N < length(x)");
int N_sub = 11; //odd subvector length
//vec y = dct(x, N_sub);
std::vector<vec> y = run_transform_test(dct, x, N_sub);
test_result(y, ref_dct(x(0, N_sub - 1)));
}
//zero-padded vector processing test
{
SCOPED_TRACE("y = dct(x, N) test, N > length(x)");
int N_zp = 8; //zero-padding length
//vec y = dct(x, N + N_zp);
std::vector<vec> y = run_transform_test(dct, x, N + N_zp);
x.set_size(N + N_zp, true);
test_result(y, ref_dct(x));
}
}
TEST(Transforms, IDCT)
{
if(!have_cosine_transforms()) FAIL() << "Cosine Transforms are not supported with this library build.";
int N = 16;
vec x = randn(N);
//vector processing test
{
SCOPED_TRACE("y = idct(x) test");
//vec y = idct(x);
std::vector<vec> y = run_transform_test(idct, x);
test_result(y, ref_idct(x));
}
//subvector processing test
{
SCOPED_TRACE("y = idct(x, N) test, N < length(x)");
int N_sub = 11; //odd subvector length
//vec y = idct(x, N_sub);
std::vector<vec> y = run_transform_test(idct, x, N_sub);
test_result(y, ref_idct(x(0, N_sub - 1)));
}
//zero-padded vector processing test
{
SCOPED_TRACE("y = idct(x, N) test, N > length(x)");
int N_zp = 8; //zero-padding length
//vec y = idct(x, N + N_zp);
std::vector<vec> y = run_transform_test(idct, x, N + N_zp);
x.set_size(N + N_zp, true);
test_result(y, ref_idct(x));
}
}
//Run several transforms sequentially. This test runs three FFT transforms sequentially in each thread.
//The test is intended to verify FFT operation on larger data sets and test for possible clashes on shared
//data in multithreaded environment.
cvec seq_transforms_test(const cvec& test_input)
{
//run 128,256,512-point FFT on test dataset and store results in output vector
cvec output(128 + 256 + 512);
output.set_subvector(0, fft(test_input, 128));
output.set_subvector(128, fft(test_input, 256));
output.set_subvector(128 + 256, fft(test_input));
return output;
}
cvec seq_transforms_ref(const cvec& test_input)
{
//compute reference transforms to verify test results
cvec output(128 + 256 + 512);
output.set_subvector(0, ref_dft(test_input(0, 127)));
output.set_subvector(128, ref_dft(test_input(0, 255)));
output.set_subvector(128 + 256, ref_dft(test_input));
return output;
}
TEST(Transforms, FFT_128_256_512)
{
if(!have_fourier_transforms()) FAIL() << "Fourier Transforms are not supported with this library build.";
cvec test_input = randn_c(512);
std::vector<cvec> y = run_transform_test(seq_transforms_test, test_input);
test_result(y, seq_transforms_ref(test_input));
}
|