fftw는 plan을 세우고 계산하는데, plan을 export 하고 import 할 수 있다고 한다.
그러면 초기 구동 속도를 더 올릴수 있으려나?
| Importing and Exporting Wisdom void fftw_export_wisdom_to_file(FILE *output_file); fftw_status fftw_import_wisdom_from_file(FILE *input_file); |
| fftw @ 22Hz sine파 대충 돌려보니 (0) | 2022.11.16 |
|---|---|
| real to complex (0) | 2022.11.04 |
| FFT (0) | 2022.10.24 |
| fftw 예제 와 복소수 처리 (0) | 2022.10.19 |
| fftw 테스트(tests/bench) (0) | 2022.10.19 |
FFT Size - 몇 번의 샘플을 분석할 것인지
Window - 사용하지 않을때 Rectangular / Flat / Uniform
- Hannning Window(cosine?)
- Blackman 윈도우, Kaiser 윈도우(최근 제안, 사이드 로브가 적은 편)
사이드 로브(side libe, leakage) - 중심 주파수 하나가 아니라 주변부 주파수가 나오는 문제
대역폭(Bandwidth) - Hz 단위
[링크 : https://m.blog.naver.com/suya309/221467948212]
| real to complex (0) | 2022.11.04 |
|---|---|
| fftw wisdom (0) | 2022.11.04 |
| fftw 예제 와 복소수 처리 (0) | 2022.10.19 |
| fftw 테스트(tests/bench) (0) | 2022.10.19 |
| fftw cross compile (0) | 2022.10.19 |
| fftw wisdom (0) | 2022.11.04 |
|---|---|
| FFT (0) | 2022.10.24 |
| fftw 테스트(tests/bench) (0) | 2022.10.19 |
| fftw cross compile (0) | 2022.10.19 |
| fft (0) | 2020.11.25 |
도대체 저 옵션들은 먼지 모르겠다.
| fftw-3.3.4/tests/bench -o nthreads=2 --verbose=1 --verify 'ok10bx6bx6e11x13b' --verify 'ik10bx6bx6e11x13b' --verify 'obrd7x13v16' --verify 'ibrd7x13v16' --verify 'ofrd7x13v16' --verify 'ifrd7x13v16' --verify '//obcd7x13v16' --verify '//ibcd7x13v16' --verify '//ofcd7x13v16' --verify '//ifcd7x13v16' --verify 'obcd7x13v16' --verify 'ibcd7x13v16' --verify 'ofcd7x13v16' --verify 'ifcd7x13v16' --verify 'okd10bv127' --verify 'ikd10bv127' --verify '//obr240' --verify '//ibr240' --verify '//ofr240' --verify '//ifr240' --verify 'obr240' --verify 'ibr240' --verify 'ofr240' --verify 'ifr240' --verify '//obc240' --verify '//ibc240' --verify '//ofc240' --verify '//ifc240' --verify 'obc240' --verify 'ibc240' --verify 'ofc240' --verify 'ifc240' --verify 'ok11760e00' --verify 'ik11760e00' --verify 'obr33v31' --verify 'ibr33v31' --verify 'ofr33v31' --verify 'ifr33v31' --verify '//obc33v31' --verify '//ibc33v31' --verify '//ofc33v31' --verify '//ifc33v31' --verify 'obc33v31' --verify 'ibc33v31' |
[링크 : https://unix.stackexchange.com/questions/209753/how-do-i-check-if-fftw-installed-correctly]
+
[링크 : https://people.sc.fsu.edu/~jburkardt/c_src/fftw_test/]
[링크 : https://people.sc.fsu.edu/~jburkardt/c_src/fftw_test/fftw_test.html]
| FFT (0) | 2022.10.24 |
|---|---|
| fftw 예제 와 복소수 처리 (0) | 2022.10.19 |
| fftw cross compile (0) | 2022.10.19 |
| fft (0) | 2020.11.25 |
| fftw (0) | 2020.11.25 |
[링크 : https://www.fftw.org/download.html]
| ./configure --prefix=/home/zhouxiaoyong/fftw3_test --disable-fortran --with-slow-timer --host=arm-none-linux-gnueabi --enable-single --enable-neon --enable-shared CC=arm-none-linux-gnueabi-gcc CFLAGS="-march=armv7-a -mfpu=neon -fPIC -ldl -mfloat-abi=softfp" |
[링크 : https://codeantenna.com/a/ztGG77F10Z]
| The basic usage of FFTW is simple. A typical call to FFTW looks like: #include <fftw.h> ... { fftw_complex in[N], out[N]; fftw_plan p; ... p = fftw_create_plan(N, FFTW_FORWARD, FFTW_ESTIMATE); ... fftw_one(p, in, out); ... fftw_destroy_plan(p); } For example, code to perform an in-place FFT of a three-dimensional array might look like: #include <fftw.h> ... { fftw_complex in[L][M][N]; fftwnd_plan p; ... p = fftw3d_create_plan(L, M, N, FFTW_FORWARD, FFTW_MEASURE | FFTW_IN_PLACE); ... fftwnd_one(p, &in[0][0][0], NULL); ... fftwnd_destroy_plan(p); } The following is a brief example in which the wisdom is read from a file, a plan is created (possibly generating more wisdom), and then the wisdom is exported to a string and printed to stdout. { fftw_plan plan; char *wisdom_string; FILE *input_file; /* open file to read wisdom from */ input_file = fopen("sample.wisdom", "r"); if (FFTW_FAILURE == fftw_import_wisdom_from_file(input_file)) printf("Error reading wisdom!\n"); fclose(input_file); /* be sure to close the file! */ /* create a plan for N=64, possibly creating and/or using wisdom */ plan = fftw_create_plan(64,FFTW_FORWARD, FFTW_MEASURE | FFTW_USE_WISDOM); /* ... do some computations with the plan ... */ /* always destroy plans when you are done */ fftw_destroy_plan(plan); /* write the wisdom to a string */ wisdom_string = fftw_export_wisdom_to_string(); if (wisdom_string != NULL) { printf("Accumulated wisdom: %s\n",wisdom_string); /* Just for fun, destroy and restore the wisdom */ fftw_forget_wisdom(); /* all gone! */ fftw_import_wisdom_from_string(wisdom_string); /* wisdom is back! */ fftw_free(wisdom_string); /* deallocate it since we're done */ } } |
| fftw 예제 와 복소수 처리 (0) | 2022.10.19 |
|---|---|
| fftw 테스트(tests/bench) (0) | 2022.10.19 |
| fft (0) | 2020.11.25 |
| fftw (0) | 2020.11.25 |
| ffmpeg fft 분석 예제 (0) | 2020.11.25 |
이산 코사인 변환 DCT(discrete cosine transform)
이산 푸리에 변환 DFT(discrete Fourier transform)
type-II DCT DCT
type-III DCT 역DCT, IDCT
실수 신호에 대하여, 변환 결과물이 복소수로 나오는 DFT
DCT는 실수로만 결과물이 나오기 때문에, 처리하기가 간편하여 신호처리 및 영상처리에 널리 사용
| fftw 테스트(tests/bench) (0) | 2022.10.19 |
|---|---|
| fftw cross compile (0) | 2022.10.19 |
| fftw (0) | 2020.11.25 |
| ffmpeg fft 분석 예제 (0) | 2020.11.25 |
| FFT 라이브러리 (0) | 2017.07.07 |
예전에 찾은적이 있네..
fft를 위한 라이브러리인데 도대체 어떤 함수를 써서 분석을 하고
주파수 대역은 어떤식으로 맵핑이 가능한건질 모르겠다.
[링크 : http://www.fftw.org/]
[링크 : https://github.com/FFTW/fftw3]
[링크 : https://iamaman.tistory.com/652]
+
2020.11.26
[링크 : http://www.fftw.org/fftw2_doc/fftw_2.html]
[링크 : http://www.fftw.org/fftw3_doc/Complex-One_002dDimensional-DFTs.html]
| fftw 테스트(tests/bench) (0) | 2022.10.19 |
|---|---|
| fftw cross compile (0) | 2022.10.19 |
| fft (0) | 2020.11.25 |
| ffmpeg fft 분석 예제 (0) | 2020.11.25 |
| FFT 라이브러리 (0) | 2017.07.07 |
일단은 마이크로 raw audio나 wav를 받아서 분석하는걸 찾아봐야겠다.
[링크 : https://www.ffmpeg.org/doxygen/trunk/fft-test_8c-source.html]
[링크 : https://www.ffmpeg.org/doxygen/trunk/fft_8c_source.html]
[링크 : https://www.ffmpeg.org/doxygen/3.3/group__lavc__fft.html]
| fftw 테스트(tests/bench) (0) | 2022.10.19 |
|---|---|
| fftw cross compile (0) | 2022.10.19 |
| fft (0) | 2020.11.25 |
| fftw (0) | 2020.11.25 |
| FFT 라이브러리 (0) | 2017.07.07 |
[링크 : http://fftw.org/]
[링크 : https://stackoverflow.com/questions/5685765/computing-fft-and-if]
일단 3.3.5 에서 NEON 지원을 하는데 64비트 ARM일 경우에만 되는 듯?
FFTW 3.3.5 Jul 31, 2016
New SIMD support:
Power8 VSX instructions in single and double precision. To use, add --enable-vsx to configure.
Support for AVX2 (256-bit FMA instructions). To use, add --enable-avx2 to configure.
Experimental support for AVX512 and KCVI. (--enable-avx512, --enable-kcvi) This code is expected to work but the FFTW maintainers do not have hardware to test it.
Support for AVX128/FMA (for some AMD machines) (--enable-avx128-fma)
Double precision Neon SIMD for aarch64. This code is expected to work but the FFTW maintainers do not have hardware to test it.ft-with-fftw-library-c
[링크 : http://www.fftw.org/release-notes.html]
AArch64 was introduced in ARMv8-A and is included in subsequent versions of ARMV8-A
[링크 : https://en.wikipedia.org/wiki/ARM_architecture#AArch64_features]
[링크 : https://libav.org/]
[링크 : https://libav.org/documentation/doxygen/master/group__lavc__fft.html]
[링크 : https://www.ffmpeg.org/doxygen/trunk/fft-test_8c-source.html]
| fftw 테스트(tests/bench) (0) | 2022.10.19 |
|---|---|
| fftw cross compile (0) | 2022.10.19 |
| fft (0) | 2020.11.25 |
| fftw (0) | 2020.11.25 |
| ffmpeg fft 분석 예제 (0) | 2020.11.25 |
|
