SNR-Based Denoising Dynamic Statistical Threshold Detection of FBG Spectral Peaks

Abstract

This paper targets a Denoising Dynamic Statistical Threshold (DDST) detection algorithm of Fiber Bragg Grating (FBG) spectral peaks at the presence of changing Signal-to-Noise Ratio (SNR) in a sensing channel. Computing the DDST is based on statistical parameters of the background noise. The DDST is determined by adjusting it using the SNR via determining the targeted probability of false alarms (p FA ). The proposed algorithm implements background noise fluctuations, nonlinear signal attenuation of a Single-Mode Fiber (SMF), as well as influence of the short-term interference. The implemented sliding wavelength window technique in conjunction with the FBG spectral peaks power scaling allow automatic adjusting of p FA and the DDST. During the possible FBGs resonant wavelengths overlap resulting from approaching/colliding spectral power responses of FBGs, the proposed algorithm also improves the detection robustness and resolving of these overlaps. The DDST marginally takes into account spectral shapes of FBGs resonant wavelength peaks. Advantageously, DDST wavelength resolving is independent of FBG spectral peaks shapes. Our DDST algorithm is also simple to implement. Measurements done by two Optical Spectral Analyzers (OSAs) confirmed significant improvements in the background noise reduction (i.e. signal denoising), noisy FBG spectral peaks shapes smoothing and SNR, improved adjacent FBG spectral peaks detectability and resolving. Our experiments also confirmed usability of the DDST algorithm under severe network conditions (with low reflected FBG power below −75 dB and low SNR < 4 dB resulting a standard deviation of σ > 7 dB in the background noise fluctuations).

Impact of Reducing Statistically Small Population Sampling on 2 Threshold Detection in FBG Optical Sensing

“…In our recent [15] digital sliding window denoising technique, a number of discrete power spectral population samples were processed. In order to increase computational efficiency, it is possible in some cases to reduce population sampling while maintaining the success of statistical detection.…”

“…Previously [14,15], we reported on a denoising technique based on digital sliding window. In this case, the statistical detector has been introduced to detect spectral power of FBGs in an additive mixture of the signal and background noise.…”

“…The related decision-making process is shown in Figure 2. In a previous study [15], a fixed K-number of the discrete power spectral samples were processed using the sliding window technique where K-number corresponded to a number of discrete wavelength steps within the FBG bandwidth. As was shown, K can be smaller.…”

Impact of Reducing Statistically Small Population Sampling on Threshold Detection in FBG Optical Sensing

“…The goal has been to avoid complex mathematical procedures, fingerprinting databases, or artificial intelligence-based techniques [19][20][21]. In [22,23], we reported on a denoising technique based on a digital sliding window. A statistical detector was introduced to detect the spectral power of FBGs in an additive mixture of the signal and background noise.…”

“…As a consequence, optimal decisions without prior knowledge of likelihood functions can be constructed regardless of the prior knowledge of the Gaussian process regression method (or similar) [28]. In a previous study [23], a fixed K-number of the discrete power spectral samples was processed using the sliding window technique where the K-number corresponded to a number of discrete wavelength steps within the FBG bandwidth. As was shown, K can be smaller.…”