FPGA Architecture for 2D Fast Fourier Transform Based on 2D Decomposition for Large-Sized Data

Jung Sub Kim1,  Chi-Li Yu2,  Lanping Deng2,  Srinidhi Kestur1,  Vijaykrishnan Narayanan1,  Chaitali Chakrabarti2
1Pennsylvania State University, USA, 2Arizona State University, USA


Abstract

In this paper, we propose an efficient architecture to implement the 2D FFT for large-sized input data based on a novel 2D decomposition algorithm. This architecture achieves very high throughput by exploiting the inherent parallelism and the row-wise burst access pattern of the external memory. A custom-designed high throughput memory interface block enables maximum utilization of the memory bandwidth. In addition, an automatic system generator is provided for mapping this architecture onto a reconfigurable platform of Xilinx Virtex4 or Virtex5 devices. For a 2K*2K input size, the proposed architecture is 1.96x times faster than RC decomposition based implementation under the same memory constraints, and also outperforms existing 2D FFT implementations.