ABSTRACT
This paper analyzes and compares existing variety of LDV technologies. According to existing LDV's deficiencies such
as limited signals processing ability, poor adaptability and low precision, a new design scheme of LDV based on digital
signal processing is introduced in this article.
Doppler signal processing systems processes the spectrum by DSP. The system consists of a filter, a wave detector, an
A/D converter and buffer. DSP system samples digital signals and performs Fourier Transform, and the information of
velocity is obtained. To achieve a high signal to noise ratio (SNR), many signal processing techniques are applied, such
as inserting zero, signal window, the algorithm of periodic diagrams based on FFT and the algorithm of approaching
peak value, all of this highly improve the noise adaptivity. In recent years, the spectrum zooming technology is
developed quickly, the algorithm of Zoom-FFT based on complex modulation improves the analysis precision,
calculation efficiency, resolution and flexibility effectively. In this paper, the time domain sampled signals are windowed
firstly, then the zooming analysis via Zoom-FFT algorithm is implemented, and at last,the energy Centrobaric rectifying
algorithm is applied to do further estimation. Simulation results indicated that the algorithm has many merits, such as
anti-jamming, high accuracy and a small amount of computation.
Keywords: laser Doppler velocimetry; Digital signals processing; Signal to noise ratio; Average periodic diagrams;
Spectrum zooming.
1. INTRODUCTION
The Laser Doppler Velocimeter (LDV) is a kind of precise instrument. It is used to measure velocity of liquid. It is based
on Doppler phenomenon of light. It uses particles put in liquid to scatter the laser that shines on it, then detects the
change of the scattered light's frequency using photoelectric detector, finally gets the velocity of liquid according to the
velocity info contained in the scattered light, because the change of frequency of the scattered light and particle's velocity
is in direct ratio. The LDV has many merits, such as rapid dynamic response, good space differentiating ability, wide
measure range and high precision.
LDV is based on Doppler Effect, in Doppler signal processing, estimates Doppler frequency firstly, and then gets the
velocity according to the direct ratio of Doppler frequency to the liquid velocity. Errors of estimated frequency influence
the result of velocity measurement seriously, so a algorithm with high precision and short consumed time is needed.
Digital FFT technology pursues Doppler frequency in frequency domain, shows its merits in dynamic response, antijamming,
analysis precision, and measure range compared with the processing in time domain or in analog frequency
domain. With digital signal processing and digital storage capacity further improved, compared with the other signal
processing technologies (such as frequency tracking technology, Counting technology, etc.) ,digital FFT has advantages
in the ability to track changes in the speed and noise adaptivity will be increasingly obvious. Thus traditional digital FFT
technology considers the result frequency corresponding to the peak value as the Doppler frequency, the problem of low
precision is presented. In this paper, through theoretical analysis, windowing the time-domain signal samples with
Hanning function, then the small size Zoom-FFT algorithm, and energy centrobaric rectifying algorithm to make further
estimates, thus the laser Doppler frequency is precisely located.

Download full report
http://infooptics.hrbeu.edu.cn/Optics/Up...582804.pdf