An FPGA-based single-precision floating-point 2048-points FFT implementation based on adaptive angle recoding CORDIC

Abstract

In this paper, an FPGA-based single-precision floating-point 2048-point FFT implementation is proposed, based on an adaptive angle recoding CORDIC algorithm. The design is built and verified on Altera Stratix V FPGA chip. The implementation had 102.55 MHz maximum frequency, throughput result of 8424.382 FFTs/s, and resources utilization of 76,282 ALUTs and 15,687 registers. The accuracy results were 5.889E-06 (Mean-Square-Error (MSE).