How to implement a DCT method in c++ using MKL?

[Jacob_L_1]

I am trying to use MKL to implement a DCT method. I cannot find any methods or examples that handle the logic. I found some documentation on http://software.intel.com/sites/products/documentation/hpc/ipp/ipps/ipps_ch7/ch7_DCT.html#ch7_DCT

These methods don't appear to be part of MKL, at least I am not able to find them. Can someone point me in the right direction?

[Ying_H_Intel]

Hi Jacob,

The documentation is from Intel IPP http://software.intel.com/en-us/intel-ipp, it is sister library of MKL . And it provide DCT method directly. You may try it and it's forum : . http://software.intel.com/en-us/forums/intel-integrated-performance-primitives/

Best Regards,

Ying

[Jacob_L_1]

Ying,

Thanks for the feedback. By what I can tell, that library has another license and purchase agreement, or am I missing something and it is included in my companies purchase of MKL?

Jacob

[Jacob_L_1]

What I am looking for is how to implement DCT using MKL. I am inheritanting the implementation from a form employee who left our company. He wrote a wrapper dll around the MKL library and did not check the code into the repository. What I can tell I don't think he purchased IPP.

[Jacob_L_1]

I found some examples in the Intel\Composer XE 2013 SP1\mkl\examples\examples_core\fftw3xc\source\sp_plan_r2r_1d.c

[SergeyKostrov]

Jacob, Let me know if you need an example for application of IPP's DCT functions.

[Ying_H_Intel]

Hi Jacob,

Right, IPP need another license. If you have only MKL, MKL provide DFT and FFTW wrapper, you can see example under MKL install directory either DFTC ( DFT c example)or FFTW as you found, then do the transform for DCT if needed.

Best Regards,

Ying