Raman Distributed Temperature Sensor Performance Analysis Using Wavelet Techniques

This paper reports a work on our study to enhance the performance of Raman distributed temperature sensors (RDTS) by utilizing wavelet techniques. Specifically, we employ wavelet transforms (WT) to analyze the temporal signal of the RDTS in both time and frequency domains, using finite support basis functions with varying resolutions. To conduct our analysis, we consider six common families of wavelet transforms, Daubechies (dbn, where n ∈ 1, 2, 3, 4), Coiflets (coifn, where n ∈ 1, 2, 3, 4), Symlets (symn, where n ∈ 3, 4, 6, 8), Fejer-Korovkin (fkn, where n = 4), and Biorthogonal spline wavelets (biorn, where n ∈ 1.1, 2.6, 3.3, 4.4, 5.5, 6.8), and interval-dependent denoising of the signal to verify the performance improvement of the designed Raman DTS. The results show that the Fejer-Korovkin4 and the interval-dependent denoising scheme perform better in denoising the resultant temperature measurement plot and SNR improvement.