Reduced Precision Redundancy in a Radix-4 FFT implementation on a Field Programmable Gate Array

Reduced Precision Redundancy (RPR) is demonstrated as a new method for improving fault tolerance in Field Programmable Gate Arrays (FPGAs) replacing Triple Modular Redundancy (TMR) to protect against the Single Event Effects due to radiation in arithmetic processes.12 As a test of this approach, the RPR technique was used to implement a Radix-4 Fast Fourier Transform (FFT). This design was implemented in a Xilinx Virtex 2® FPGA in order to find the possible gain in speed and reduction in power and resources as compared to the TMR method. Simulation of different degrees of RPR explore the impact on speed and power on the FPGA performance at various levels of precision reduction.

[1]  J. Tukey,et al.  An algorithm for the machine calculation of complex Fourier series , 1965 .

[2]  Zhang Qin,et al.  Design of a high performance FFT processor based on FPGA , 2005, Proceedings of the ASP-DAC 2005. Asia and South Pacific Design Automation Conference, 2005..

[3]  Xuemei Liu,et al.  The Design of Radix-4 FFT by FPGA , 2008, 2008 International Symposium on Intelligent Information Technology Application Workshops.

[4]  Athanasios Gavros.  Use of the Reduced Precision Redundancy (RPR) method in a radix4 FFT implementation , 2010 .

[5]  M.N.S. Swamy,et al.  An improved radix-16 FFT algorithm , 2004, Canadian Conference on Electrical and Computer Engineering 2004 (IEEE Cat. No.04CH37513).

[6]  M. Omair Ahmad,et al.  Improved radix-4 and radix-8 FFT algorithms , 2004, 2004 IEEE International Symposium on Circuits and Systems (IEEE Cat. No.04CH37512).

[7]  Margaret A. Sullivan.  Reduced precision redundancy applied to arithmetic operations in field programmable gate arrays for satellite control and sensor systems , 2008 .