Analysis of radix-2 decimation in time algorithm for FPGA co-processors
Muhammad Ibrahim, Mohsin Kamal, Omar Khan, Khalil Ullah · 2016
The Fast Fourier Transform (FFT) is a widely used algorithm for spectral analysis of signals and is widely used in applications of communications, biomedical signal processing, industrial process control, etc. Its wide usage and application has resulted in the implementation of a number of FFT algorithms for inputs of various sizes, dimensions, and platform types such as desktop computers, graphic cards, and special micro-processors. In this paper, a common but important FFT algorithm known as the radix-2 decimation in time (DIT) algorithm is implemented on a Spartan-6 FPGA kit using Xilinx ISE 14.2. The implementation is compared with various pre-installed IP-core modules of Xilinx for complex input of various sample sizes. The results show that our implementation of the radix-2 DIT algorithm for FPGA systems has not only a lower computation cost, less memory and resources utilization, but also provides better absolute error compared to the IP-core modules. A hardware synthesis is also provided.