1 2 /* ----------------------------------------------------------------------------------------------------------- 3 Software License for The Fraunhofer FDK AAC Codec Library for Android 4 5 Copyright 1995 - 2013 Fraunhofer-Gesellschaft zur Frderung der angewandten Forschung e.V. 6 All rights reserved. 7 8 1. INTRODUCTION 9 The Fraunhofer FDK AAC Codec Library for Android ("FDK AAC Codec") is software that implements 10 the MPEG Advanced Audio Coding ("AAC") encoding and decoding scheme for digital audio. 11 This FDK AAC Codec software is intended to be used on a wide variety of Android devices. 12 13 AAC's HE-AAC and HE-AAC v2 versions are regarded as today's most efficient general perceptual 14 audio codecs. AAC-ELD is considered the best-performing full-bandwidth communications codec by 15 independent studies and is widely deployed. AAC has been standardized by ISO and IEC as part 16 of the MPEG specifications. 17 18 Patent licenses for necessary patent claims for the FDK AAC Codec (including those of Fraunhofer) 19 may be obtained through Via Licensing (www.vialicensing.com) or through the respective patent owners 20 individually for the purpose of encoding or decoding bit streams in products that are compliant with 21 the ISO/IEC MPEG audio standards. Please note that most manufacturers of Android devices already license 22 these patent claims through Via Licensing or directly from the patent owners, and therefore FDK AAC Codec 23 software may already be covered under those patent licenses when it is used for those licensed purposes only. 24 25 Commercially-licensed AAC software libraries, including floating-point versions with enhanced sound quality, 26 are also available from Fraunhofer. Users are encouraged to check the Fraunhofer website for additional 27 applications information and documentation. 28 29 2. COPYRIGHT LICENSE 30 31 Redistribution and use in source and binary forms, with or without modification, are permitted without 32 payment of copyright license fees provided that you satisfy the following conditions: 33 34 You must retain the complete text of this software license in redistributions of the FDK AAC Codec or 35 your modifications thereto in source code form. 36 37 You must retain the complete text of this software license in the documentation and/or other materials 38 provided with redistributions of the FDK AAC Codec or your modifications thereto in binary form. 39 You must make available free of charge copies of the complete source code of the FDK AAC Codec and your 40 modifications thereto to recipients of copies in binary form. 41 42 The name of Fraunhofer may not be used to endorse or promote products derived from this library without 43 prior written permission. 44 45 You may not charge copyright license fees for anyone to use, copy or distribute the FDK AAC Codec 46 software or your modifications thereto. 47 48 Your modified versions of the FDK AAC Codec must carry prominent notices stating that you changed the software 49 and the date of any change. For modified versions of the FDK AAC Codec, the term 50 "Fraunhofer FDK AAC Codec Library for Android" must be replaced by the term 51 "Third-Party Modified Version of the Fraunhofer FDK AAC Codec Library for Android." 52 53 3. NO PATENT LICENSE 54 55 NO EXPRESS OR IMPLIED LICENSES TO ANY PATENT CLAIMS, including without limitation the patents of Fraunhofer, 56 ARE GRANTED BY THIS SOFTWARE LICENSE. Fraunhofer provides no warranty of patent non-infringement with 57 respect to this software. 58 59 You may use this FDK AAC Codec software or modifications thereto only for purposes that are authorized 60 by appropriate patent licenses. 61 62 4. DISCLAIMER 63 64 This FDK AAC Codec software is provided by Fraunhofer on behalf of the copyright holders and contributors 65 "AS IS" and WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, including but not limited to the implied warranties 66 of merchantability and fitness for a particular purpose. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR 67 CONTRIBUTORS BE LIABLE for any direct, indirect, incidental, special, exemplary, or consequential damages, 68 including but not limited to procurement of substitute goods or services; loss of use, data, or profits, 69 or business interruption, however caused and on any theory of liability, whether in contract, strict 70 liability, or tort (including negligence), arising in any way out of the use of this software, even if 71 advised of the possibility of such damage. 72 73 5. CONTACT INFORMATION 74 75 Fraunhofer Institute for Integrated Circuits IIS 76 Attention: Audio and Multimedia Departments - FDK AAC LL 77 Am Wolfsmantel 33 78 91058 Erlangen, Germany 79 80 www.iis.fraunhofer.de/amm 81 amm-info (at) iis.fraunhofer.de 82 ----------------------------------------------------------------------------------------------------------- */ 83 84 /*************************** Fraunhofer IIS FDK Tools ********************** 85 86 Author(s): Haricharan Lakshman, Manuel Jander 87 Description: Trigonometric functions fixed point fractional implementation. 88 89 ******************************************************************************/ 90 91 #include "FDK_trigFcts.h" 92 93 #include "fixpoint_math.h" 94 95 96 97 98 #define IMPROVE_ATAN2_ACCURACY 1 // 0 --> 59 dB SNR 1 --> 65 dB SNR 99 #define MINSFTAB 7 100 #define MAXSFTAB 25 101 102 #if IMPROVE_ATAN2_ACCURACY 103 static const FIXP_DBL f_atan_expand_range[MAXSFTAB-(MINSFTAB-1)] = 104 { 105 /***************************************************************************** 106 * 107 * Table holds fixp_atan() output values which are outside of input range 108 * of fixp_atan() to improve SNR of fixp_atan2(). 109 * 110 * This Table might also be used in fixp_atan() [todo] so there a wider input 111 * range can be covered, too. 112 * 113 * Matlab (generate table): 114 * for scl = 7:25 % MINSFTAB .. MAXSFTAB 115 * at=atan(0.5 *(2^scl)); % 0.5 because get in 'middle' area of current scale level 'scl' 116 * at/2 % div at by ATO_SCALE 117 * end 118 * 119 * Table divided by 2=ATO_SCALE <-- SF=ATO_SF 120 *****************************************************************************/ 121 FL2FXCONST_DBL(7.775862990872099e-001), FL2FXCONST_DBL(7.814919928673978e-001), FL2FXCONST_DBL(7.834450483314648e-001), 122 FL2FXCONST_DBL(7.844216021392089e-001), FL2FXCONST_DBL(7.849098823026687e-001), FL2FXCONST_DBL(7.851540227918509e-001), 123 FL2FXCONST_DBL(7.852760930873737e-001), FL2FXCONST_DBL(7.853371282415015e-001), FL2FXCONST_DBL(7.853676458193612e-001), 124 FL2FXCONST_DBL(7.853829046083906e-001), FL2FXCONST_DBL(7.853905340029177e-001), FL2FXCONST_DBL(7.853943487001828e-001), 125 FL2FXCONST_DBL(7.853962560488155e-001), FL2FXCONST_DBL(7.853972097231319e-001), FL2FXCONST_DBL(7.853976865602901e-001), 126 FL2FXCONST_DBL(7.853979249788692e-001), FL2FXCONST_DBL(7.853980441881587e-001), FL2FXCONST_DBL(7.853981037928035e-001), 127 FL2FXCONST_DBL(7.853981335951259e-001) 128 // pi/4 = 0.785398163397448 = pi/2/ATO_SCALE 129 }; 130 #endif 131 132 FIXP_DBL fixp_atan2(FIXP_DBL y, FIXP_DBL x) 133 { 134 FIXP_DBL q; 135 FIXP_DBL at; // atan out 136 FIXP_DBL at2; // atan2 out 137 FIXP_DBL ret = FL2FXCONST_DBL(-1.0f); 138 INT sf,sfo,stf; 139 140 // --- division 141 142 if (y > FL2FXCONST_DBL(0.0f)) 143 { 144 if (x > FL2FXCONST_DBL(0.0f)) { 145 q = fDivNormHighPrec( y, x, &sf); // both pos. 146 } 147 else if (x < FL2FXCONST_DBL(0.0f)) { 148 q = -fDivNormHighPrec( y,-x, &sf); // x neg. 149 } 150 else {//(x ==FL2FXCONST_DBL(0.0f)) 151 q = FL2FXCONST_DBL(+1.0f); // y/x = pos/zero = +Inf 152 sf = 0; 153 } 154 } 155 else if (y < FL2FXCONST_DBL(0.0f)) 156 { 157 if (x > FL2FXCONST_DBL(0.0f)) { 158 q = -fDivNormHighPrec(-y, x, &sf); // y neg. 159 } 160 else if (x < FL2FXCONST_DBL(0.0f)) { 161 q = fDivNormHighPrec(-y,-x, &sf); // both neg. 162 } 163 else {//(x ==FL2FXCONST_DBL(0.0f)) 164 q = FL2FXCONST_DBL(-1.0f); // y/x = neg/zero = -Inf 165 sf = 0; 166 } 167 } 168 else { // (y ==FL2FXCONST_DBL(0.0f)) 169 q = FL2FXCONST_DBL(0.0f); 170 sf = 0; 171 } 172 sfo = sf; 173 174 // --- atan() 175 176 if ( sfo > ATI_SF ) { 177 // --- could not calc fixp_atan() here bec of input data out of range 178 // ==> therefore give back boundary values 179 180 #if IMPROVE_ATAN2_ACCURACY 181 if (sfo > MAXSFTAB) sfo = MAXSFTAB; 182 #endif 183 184 if ( q > FL2FXCONST_DBL(0.0f) ) { 185 #if IMPROVE_ATAN2_ACCURACY 186 at = +f_atan_expand_range[sfo-ATI_SF-1]; 187 #else 188 at = FL2FXCONST_DBL( +M_PI/2 / ATO_SCALE); 189 #endif 190 } 191 else if ( q < FL2FXCONST_DBL(0.0f) ) { 192 #if IMPROVE_ATAN2_ACCURACY 193 at = -f_atan_expand_range[sfo-ATI_SF-1]; 194 #else 195 at = FL2FXCONST_DBL( -M_PI/2 / ATO_SCALE); 196 #endif 197 } 198 else { // q== FL2FXCONST_DBL(0.0f) 199 at = FL2FXCONST_DBL( 0.0f ); 200 } 201 }else{ 202 // --- calc of fixp_atan() is possible; input data within range 203 // ==> set q on fixed scale level as desired from fixp_atan() 204 stf = sfo - ATI_SF; 205 if (stf > 0) q = q << (INT)fMin( stf,DFRACT_BITS-1); 206 else q = q >> (INT)fMin(-stf,DFRACT_BITS-1); 207 at = fixp_atan(q); // ATO_SF 208 } 209 210 // --- atan2() 211 212 at2 = at >> (AT2O_SF - ATO_SF); // now AT2O_SF for atan2 213 if ( x > FL2FXCONST_DBL(0.0f) ) { 214 ret = at2; 215 } 216 else if ( x < FL2FXCONST_DBL(0.0f) ) { 217 if ( y >= FL2FXCONST_DBL(0.0f) ) { 218 ret = at2 + FL2FXCONST_DBL( M_PI / AT2O_SCALE); 219 } else { 220 ret = at2 - FL2FXCONST_DBL( M_PI / AT2O_SCALE); 221 } 222 } 223 else { 224 // x == 0 225 if ( y > FL2FXCONST_DBL(0.0f) ) { 226 ret = FL2FXCONST_DBL( +M_PI/2 / AT2O_SCALE); 227 } 228 else if ( y < FL2FXCONST_DBL(0.0f) ) { 229 ret = FL2FXCONST_DBL( -M_PI/2 / AT2O_SCALE); 230 } 231 else if ( y == FL2FXCONST_DBL(0.0f) ) { 232 ret = FL2FXCONST_DBL(0.0f); 233 } 234 } 235 return ret; 236 } 237 238 239 FIXP_DBL fixp_atan(FIXP_DBL x) 240 { 241 INT sign; 242 FIXP_DBL result, temp; 243 244 // SNR of fixp_atan() = 56 dB 245 FIXP_DBL ONEBY3P56 = (FIXP_DBL)0x26800000; // 1.0/3.56 in q31 246 FIXP_DBL P281 = (FIXP_DBL)0x00013000; // 0.281 in q18 247 FIXP_DBL ONEP571 = (FIXP_DBL)0x6487ef00; // 1.571 in q30 248 249 if (x < FIXP_DBL(0)) { 250 sign = 1; 251 x = - x ; 252 } else { 253 sign = 0; 254 } 255 256 /* calc of arctan */ 257 if(x < ( Q(Q_ATANINP)-FL2FXCONST_DBL(0.00395)) ) 258 { 259 INT res_e; 260 261 temp = fPow2(x); // q25 * q25 - (DFRACT_BITS-1) = q19 262 temp = fMult(temp, ONEBY3P56); // q19 * q31 - (DFRACT_BITS-1) = q19 263 temp = temp + Q(19); // q19 + q19 = q19 264 result = fDivNorm(x, temp, &res_e); 265 result = scaleValue(result, (Q_ATANOUT-Q_ATANINP+19-DFRACT_BITS+1) + res_e ); 266 } 267 else if( x < FL2FXCONST_DBL(1.28/64.0) ) 268 { 269 FIXP_DBL delta_fix; 270 FIXP_DBL PI_BY_4 = FL2FXCONST_DBL(3.1415926/4.0) >> 1; /* pi/4 in q30 */ 271 272 delta_fix = (x - FL2FXCONST_DBL(1.0/64.0)) << 5; /* q30 */ 273 result = PI_BY_4 + (delta_fix >> 1) - (fPow2Div2(delta_fix)); 274 } 275 else 276 { 277 INT res_e; 278 279 temp = fPow2Div2(x); // q25 * q25 - (DFRACT_BITS-1) - 1 = q18 280 temp = temp + P281; // q18 + q18 = q18 281 result = fDivNorm(x, temp, &res_e); 282 result = scaleValue(result, (Q_ATANOUT-Q_ATANINP+18-DFRACT_BITS+1) + res_e ); 283 result = ONEP571 - result; // q30 + q30 = q30 284 } 285 if (sign) { 286 result = -result; 287 } 288 289 return(result); 290 } 291 292 293 294 #include "FDK_tools_rom.h" 295 296 FIXP_DBL fixp_cos(FIXP_DBL x, int scale) 297 { 298 FIXP_DBL residual, error, sine, cosine; 299 300 residual = fixp_sin_cos_residual_inline(x, scale, &sine, &cosine); 301 error = fMult(sine, residual); 302 303 return cosine - error; 304 } 305 306 FIXP_DBL fixp_sin(FIXP_DBL x, int scale) 307 { 308 FIXP_DBL residual, error, sine, cosine; 309 310 residual = fixp_sin_cos_residual_inline(x, scale, &sine, &cosine); 311 error = fMult(cosine, residual); 312 313 return sine + error; 314 } 315 316 void fixp_cos_sin (FIXP_DBL x, int scale, FIXP_DBL *cos, FIXP_DBL *sin) 317 { 318 FIXP_DBL residual, error0, error1, sine, cosine; 319 320 residual = fixp_sin_cos_residual_inline(x, scale, &sine, &cosine); 321 error0 = fMult(sine, residual); 322 error1 = fMult(cosine, residual); 323 *cos = cosine - error0; 324 *sin = sine + error1; 325 } 326 327 328 329 330 331