The effect of signal to noise ratio on accuracy of temperature measurements for Brillouin lidar in water. (24th August 2017)

• Record Type: Journal Article
• Authors: Liang, Kun; Niu, Qunjie; Wu, Xiangkui; Xu, Jiaqi; Peng, Li; Zhou, Bo
• Is Part Of: Laser physics. Volume 27:Number 9(2017:Sep.)
• Publication Date: 2017-08-24
• Subjects: Brillouin scattering -- lidar -- SNR -- temperature measurements; Lasers -- Periodicals; 621.36605
• DOI: 10.1088/1555-6611/aa760c
• Languages: English
• ISSNs: 1054-660X

Abstract

Abstract: A lidar system with Fabry–Pérot etalon and an intensified charge coupled device can be used to obtain the scattering spectrum of the ocean and retrieve oceanic temperature profiles. However, the spectrum would be polluted by noise and result in a measurement error. To analyze the effect of signal to noise ratio (SNR) on the accuracy of measurements for Brillouin lidar in water, the theory model and characteristics of SNR are researched. The noise spectrums with different SNR are repetitiously measured based on simulation and experiment. The results show that accuracy is related to SNR, and considering the balance of time consumption and quality, the average of five measurements is adapted for real remote sensing under the pulse laser conditions of wavelength 532 nm, pulse energy 650 mJ, repetition rate 10 Hz, pulse width 8 ns and linewidth 0.003 cm −1 (90 MHz). Measuring with the Brillouin linewidth has a better accuracy at a lower temperature (<15 °C), while measuring with the Brillouin shift is a more appropriate method at a higher temperature (>15 °C), based on the classical retrieval model we adopt. The experimental results show that the temperature error is 0.71 °C and 0.06 °C based on shift and linewidth respectively when the image SNR is at the range of 3.2 dB–3.9 dB.