FPGA based fiber optic gyroscope signal denoising using discrete wavelet transform

Abstract

This paper presents field programmable gate array (FPGA) implementation of the forward/inverse discrete wavelet transform for denoising Fiber Optic Gyroscope (FOG) signal. In this work an extensive study on the effect of different threshold techniques of DWT algorithm are carried out denoising the FOG signal. Different architectures such as multiply and accumulate (MAC), Distributed Arithmetic (DA) and Systolic MAC are used to implemented the DWT algorithm on XILINX Virtex-5FXT-1136 FPGA development board. The resource utilization of all the three architectures are compared and the experimental result concludes that the DA architecture is the optimum architecture that gives latency of 1057 clock cycles for processing 1024 samples with maximum operating frequency as 174 MHz. Further the DWT algorithm reduces the drift by 5 and 100 times for denoising the gyroscope and accelerometer signal of the considered FOG. These results confirmed a DWT shows good improvement of denoising for processing the noisy FOG and accelerometer signal. © 2006-2012 Asian Research Publishing Network (ARPN).