This paper presents a new approach to exploit the symmetry of twiddle factors of Real-valued Fast Fourier Transform (RFFT) in a Data-Flow model. By following Data-Flow principles, we organize the operations of the FFT in order to reduce the execution time. In particular, we show that by using a Data-Flow specification in a FFT application, we can have a better execution time in comparison to the currently used libraries and algorithms such as FFTW.