Tosi Daniele
Nazarbayev University, School of Engineering, 010000 Astana, Kazakhstan.
Sensors (Basel). 2015 Oct 29;15(11):27470-92. doi: 10.3390/s151127470.
The Karhunen-Loeve Transform (KLT) is applied to accurate detection of optical fiber sensors in the spectral domain. By processing an optical spectrum, although coarsely sampled, through the KLT, and subsequently processing the obtained eigenvalues, it is possible to decode a plurality of optical sensor results. The KLT returns higher accuracy than other demodulation techniques, despite coarse sampling, and exhibits higher resilience to noise. Three case studies of KLT-based processing are presented, representing most of the current challenges in optical fiber sensing: (1) demodulation of individual sensors, such as Fiber Bragg Gratings (FBGs) and Fabry-Perot Interferometers (FPIs); (2) demodulation of dual (FBG/FPI) sensors; (3) application of reverse KLT to isolate different sensors operating on the same spectrum. A simulative outline is provided to demonstrate the KLT operation and estimate performance; a brief experimental section is also provided to validate accurate FBG and FPI decoding.
卡尔胡宁-勒夫变换（KLT）用于在光谱域中精确检测光纤传感器。通过对光谱（尽管采样粗糙）进行KLT处理，随后对得到的特征值进行处理，就有可能解码多个光纤传感器的结果。尽管采样粗糙，但KLT比其他解调技术返回的精度更高，并且对噪声具有更高的抗性。本文给出了三个基于KLT处理的案例研究，代表了光纤传感领域当前的大部分挑战：（1）单个传感器的解调，如光纤布拉格光栅（FBG）和法布里-珀罗干涉仪（FPI）；（2）双（FBG/FPI）传感器的解调；（3）应用逆KLT来分离在同一光谱上工作的不同传感器。提供了一个模拟概要以演示KLT操作并估计性能；还提供了一个简短的实验部分以验证对FBG和FPI的精确解码。
