今天晚上把整个二维FFT的计算终于写完了，明天把rapidIO一连就大功告成了。这个工程我打个包放到这里2维FFT图像处理大家有需要的可以自行下载查看
最后也用matlba与modelsim进行了对比，精度还是令人满意的。下面大家可以自行比较。
modelsim仿真：
这个是Matlba仿真代码：
for i=1:16;
for j=1:16;
A1(i,j)=16*(i-1)+j;
end
end
A2=fft2(A1)MATLAB结果如下：
A2 =
1.0e+04 *
Columns 1 through 5
3.2896 -0.0128 + 0.0643i -0.0128 + 0.0309i -0.0128 + 0.0192i -0.0128 + 0.0128i
-0.2048 + 1.0296i 0 0 0 0
-0.2048 + 0.4944i 0 0 0 0
-0.2048 + 0.3065i 0 0 0 0
-0.2048 + 0.2048i 0 0 0 0
-0.2048 + 0.1368i 0 0 0 0
-0.2048 + 0.0848i 0 0 0 0
-0.2048 + 0.0407i 0 0 0 0
-0.2048 0 0 0 0
-0.2048 - 0.0407i 0 0 0 0
-0.2048 - 0.0848i 0 0 0 0
-0.2048 - 0.1368i 0 0 0 0
-0.2048 - 0.2048i 0 0 0 0
-0.2048 - 0.3065i 0 0 0 0
-0.2048 - 0.4944i 0 0 0 0
-0.2048 - 1.0296i 0 0 0 0
Columns 6 through 10
-0.0128 + 0.0086i -0.0128 + 0.0053i -0.0128 + 0.0025i -0.0128 -0.0128 - 0.0025i
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
Columns 11 through 15
-0.0128 - 0.0053i -0.0128 - 0.0086i -0.0128 - 0.0128i -0.0128 - 0.0192i -0.0128 - 0.0309i
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
Column 16
-0.0128 - 0.0643i
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
原文：http://blog.csdn.net/xz_rabbit/article/details/19348519