Home

This document is a cache from http://notes.ump.edu.my/fkee/e-Books/C%20Programming%20&%20PC%20interfacing/(ebook-pdf)%20DSP%20-%20Real%20Time%20Digital%20Signal%20Processing.pdf


Real-time digital signal processing: implementations, ... changes in the input signal is limited by its internal clock rate, so that it may be slow to

Document source : notes.ump.edu.my

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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 All Pages

3. Build the fixed-point FFT project and verify the results. Comparing the results
with the floating-point complex radix-2 FFT results obtained by running it on
the PC.
4. The FFT output samples are squared and placed in a data buffer named
spectrum []. Use CCS to plot the results by displaying the spectrum []and
re1 []buffer.
5. Profile the DSP run-time clock cycles for 128-point and 1024-point FFTs. Record
the memory usage of the fixed-point functions bit_rev()and fft().
7.5.2 Experiment 7B ± Radix-2 Complex FFT Using Assembly
Language
Although using intrinsics can improve the DSP performance, the assembly language
implementation has been proven to have the fastest execution speed and memory
efficiency for most applications, especially for computational intensive algorithms
such as FFT. The development time for assembly code, however, will be much
longer than that of C code. In addition, the maintenance and upgrade of assembly
code are usually more difficult. In this experiment, we will use C55x assembly routines
for computing the same radix-2 FFT algorithm as the fixed-point C function used
in Experiment 7A. The assembly FFT routine is listed in Table 7.5. This routine is
written based on the C function used for Experiment 7A, and it follows the C55x
C calling convention. For readability, the assembly code has been written to mimic
the C function of Experiment 7A closely. It optimizes the memory usage but not
the run-time efficiency. By unrolling the loop and taking advantage of the FFT butterfly
characteristics, the FFT execution speed can be further improved with the expense of
the memory space, see the exercise problems at the end of this chapter.
In fft.asm, the local variables are defined as structure using the stack relative
addressing mode when the assembly routine is called. The last memory location con-
tains the return address of the caller function. Since the status registers ST1and ST3 will
be modified, we use two stack locations to store the contents of these status registers at
entry. The status registers will be restored upon returning to the caller function. The
complex temporary variable is stored in two consecutive memory locations by using a
bracket with the numerical number to indicate the number of memory locations for the
integer data type.
The FFT implementation is carried out in three nested loops. The butterfly computa-
tion is implemented in the inner loop and the group loop is in the middle, while the
stages are managed by the outer loop. Among these three loops, the butterfly loop is
repeated most often. We use the local block repeat instruction, rptblocal, for the
butterfly loop and the middle loop to minimize the loop overhead. We also use parallel
instructions, modulo addressing, and dual memory access instructions to further
improve the efficiency of butterfly computation. By limiting the size of the loop, we
can place the middle loop inside the DSP instruction buffer queue (IBQ) as well. The
FFT computation is improved since the two inner loops are only fetched once from the
program memory each time we compute the groups and butterflies. The twiddle-factor
EXPERIMENTS USING THE TMS320C55X
341

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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 All Pages






Summary :

7.5.2 Experiment 7B ± Radix-2 Complex FFT Using Assembly Language Although using intrinsics can improve the DSP performance, the assembly language implementation has been proven to have the fastest execution speed and memory efficiency for most applications, especially for computational intensive algorithms such as FFT. The butterfly computa- tion is implemented in the inner loop and the group loop is in the middle, while the stages are managed by the outer loop. We use the local block repeat instruction, rptblocal, for the butterfly loop and the middle loop to minimize the loop overhead.


Tags : memory,assembly,use,function,butterfly,experiment,using,code,results,routine,dsp,radix2,efficiency





Terms    |    Link pdf-search-files.com    |    Site Map
   |    Content Removal Notice   
   |    Contact   

All books are the property of their respective owners.
Please respect the publisher and the author for their creations if their books copyrighted