readme.txt

TigerSharc TS201 32-bit floating point FFT routine

the file fft16 contains Assembly and C source code for performing
Fast Fourier Transforms in fixed point format

Use at your own risk.

 ------------------------------------------------------------------------------


1. fixed point FFT


The file fft_fxp16.zip contains Assembly and C source code for performing
optimized Complex Fixed Point Fast Fourier Transform.

-fft_fxp16.asm : Optimized complex C-callable fixed point FFT.
-main.c : main program to call fft16.asm file.
-variables.asm : Twiddle factor table
-FFTDef.h : Define
-ini_cache.h: contains the program that preloads the cache with the twiddless
-macros_rev0.8.h: contains the program used to initialize the cache
-ADSP-TS201_C.ldf : Linker Description File for TS201
-ADSP-TS101_C.ldf : Linker Description File for TS101
-generate_number.m: matlab program that produces input files to FFT
-generate_twiddles.m: matlab program that produces the twiddles factors for various FFTs
-cvt2hex.m, cvt2real.m, save_fixed.m, twiddles.m: matlab utilities
-FFT_fxp16.m: matlab program that tests the outputs of the assembly program
-fft_N_TS201.tcl, fft_N_Ts101.tcl with N=64, 128, ...,32768: tcl scripts used to
    build, execute and save the fft results
-inputs folder: contains the input points files: inputN.dat, N=64, 128,...,32768
-twiddles folder: contains the twiddle factor specific for each number of
    points N: twidN.dat
-outputs folder: contains the fft output files output_N.dat obtained from the DSP and
    the files output_N_matlab.dat obtained by performing the fft in Matlab.

Please see the file FFT_fxp16_Project_rev01.doc for more details.