This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: UCLA Electrical Engineering Professor Jain EE 113D TA Delbert Huang Experiment 5 Fast Fourier Transform Purpose 1. To understand how subroutines and function calls are used in the TMS320C542 assembly code. 2. To understand how the Fast Fourier Transform (FFT) is implemented on the TMS320C542 DSKPlus board. 3. To experiment with signal capture and spectral analysis using the data input/output and graphing tools provided with the Code Explorer software for the DSKPlus board. Pre-lab 1. Review the theory and implementation of the FFT (refer to Proakis and Manolakis or your EE113 notes). 2. In the lab, we will be implementing a 2N-point, radix-2, FFT using the RFFT.ASM originally from Texas Instruments. The data sequence can range from 16 to 1024 (in power of 2 increments, of course). Fortunately, you will not have to program the algorithm for accomplishing this task by yourself! You will nd the source code on the class web site. Unzip it into C: \ EE113L \ SOURCE \ EXP_5 directory. The main program makes use of a number of subroutine programs or macros. The FFT computation itself is carried out in four phases, and the macro les are designed to implement each of these phases. Initially, the program assumes that the data sequence to be analyzed is already available in the data memory of the processor. In the actual experiment, you will be using a modi cation of an earlier program to capture data sequences from real world signals and making this data available for the RFFT.ASM program. Phase one involves bit-reversal of the input data. Phase two performs the ( logN )-stage complex FFT (using pre-computed twiddle factors from sine and cosine tables, in xed point integer format). Phase three separates the FFT output in order to compute four independent real sequences which are the real even part, odd real part, even imaginary part, and the odd imaginary part, respectively. The fourth and nal phase performs one more set of butter ies in order to create the complex output which corresponds to the 2N-point complex FFT of the original 2N-point real input sequence. Additionalcorresponds to the 2N-point complex FFT of the original 2N-point real input sequence....
View Full Document
- Spring '09
- Electrical Engineering