关键词: 分布式测温, 拉曼信号, 自适应滤波, 变步长, 最小均方根

Abstract:

The core of a distributed optical fiber temperature measurement system (DTS) is signal collection and data processing.The measuring accuracy,spatial resolution,collecting speed and final response time are all affected by the noise in signal.The anti-Stokes signal contains very weak temperature information and is often completely overwhelmed by noise.Therefore,getting the useful temperature information from the signal is particularly important.A variable step size least-mean-square (LMS) adaptive filtering algorithm is researched for the characteristic of the noise signal in distributed optical fiber temperature measurement system and used to denoise the Raman scattering temperature signal.Matlab simulation shows that this algorithm has good denoising performance and improves the system measurement accuracy and spatial resolution.

Key words: distributed temperature;Raman signal;adaptive filtering;variable step size;minimum rms