Dynamic Kernel Function Fast Fourier Transform with Variable Truncation Scheme for Wideband Signal Detection

• Published in: 2008 IEEE Instrumentation and Measurement Technology Conference pp. 1354 - 1359
• Main Authors: Lee, Y.-H.G., Sarathy, V., Chen, C.-I.H.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.05.2008
• Subjects: Analog-digital conversion; Bandwidth; dynamic kernel function; Dynamic range; Fast Fourier transforms; Field programmable gate arrays; high throughput rate; instantaneous dynamic range; Kernel; Signal design; Signal detection; spurious-free dynamic range; Throughput; variable truncation scheme for fixed-point fixed-precision FFT; Wideband
• ISBN: 9781424415403; 1424415403
• ISSN: 1091-5281
• DOI: 10.1109/IMTC.2008.4547253

A 2.56 giga sample per second (GSPS) fixed-point fixed-precision dynamic kernel function fast Fourier transform (FFT) with variable truncation scheme is designed for real-time wideband signal detection. This design is verified for two-tone signal detection in a bandwidth of 1.14 GHz at 20 MHz intervals with output throughput rates of 50 ns. The dynamic kernel FFT uses about 32 % of the slices available on the Xilinx Virtex 4 FPGA model XC4VSX55 onboard the Delphi ADC3255 with an Atmel 10-bit analog to digital converter (ADC). The proposed design has an averaged single signal spurious-free dynamic range (SFDR) of 40.6 dB and the ability to detect a weak input signal at -45 dBm. This is possible for a fixed-precision FFT design due to the embedded variable truncation scheme. In addition, with an ideal 8-bit ADC, averaged single signal SFDR performance of 47.60 dB and two-tone signal instantaneous dynamic range (IDR) of 32.60 dB are achieved. The two-tone signal IDR is the ratio between the strong and weak signal inputs in the time domain. The high throughput rates and single signal SFDR have performed better than Xilinx LogicCOREtrade 256-point IP FFT's throughput rate of 4.827 to 5.119 mus and single-tone SFDR of 25- 30 dB for an 8-bit input.