Journal: Sensors (Basel, Switzerland)

Article Title: Review and Analysis of Peak Tracking Techniques for Fiber Bragg Grating Sensors

doi: 10.3390/s17102368

Figure Lengend Snippet: DWT denoising of an FBG spectrum. The original noisy spectrum (blue) is filtered through a DWT→thresholding→IDWT cycle once (red) and twice (green) for denoising. Daubechies wavelets db1 ( left ) and db3 ( right ) have been used.

Article Snippet: It is possible to notice that, apart from a delay introduced by the db3 wavelet, multiple DWT filtering reduces the noise on each FBG spectrum, as the FBG peak stands progressively over the noise threshold.The wavelet process of denoising, using a db3 wavelet, is shown in as implemented with the Matlab wavelet analysis [ ].

Techniques:

Journal: Sensors (Basel, Switzerland)

Article Title: Review and Analysis of Peak Tracking Techniques for Fiber Bragg Grating Sensors

doi: 10.3390/s17102368

Figure Lengend Snippet: Stationary wavelet analysis of an input spectrum. The upper part shows the noisy FBG spectrum (length N = 416), and its denoised estimation through a single db3 operation. The left column shows decimated wavelet waveforms, each having length ( N /2): the chart shows the approximation a 3 , and the detail d 1 , d 2 , d 3 . The right chart shows the upsampled ( Q = 2) replicas of the decimated coefficients, to restore the original spectral length.

Article Snippet: It is possible to notice that, apart from a delay introduced by the db3 wavelet, multiple DWT filtering reduces the noise on each FBG spectrum, as the FBG peak stands progressively over the noise threshold.The wavelet process of denoising, using a db3 wavelet, is shown in as implemented with the Matlab wavelet analysis [ ].

Techniques:

Journal: Sensors (Basel, Switzerland)

Article Title: Review and Analysis of Peak Tracking Techniques for Fiber Bragg Grating Sensors

doi: 10.3390/s17102368

Figure Lengend Snippet: Effect of wavelet denoising on a FBG spectrum sampled over 1 pm wavelength grid. ( left chart ) SNR = 3 dB; ( right chart ) = 23 dB. Denoising has been performed with db3 wavelet.

Article Snippet: It is possible to notice that, apart from a delay introduced by the db3 wavelet, multiple DWT filtering reduces the noise on each FBG spectrum, as the FBG peak stands progressively over the noise threshold.The wavelet process of denoising, using a db3 wavelet, is shown in as implemented with the Matlab wavelet analysis [ ].

Techniques: