High-temperature measurements above 1000 °C are critical in harsh environments such as aerospace, metallurgy, fossil fuel, and power production. Fiber-optic high-temperature sensors are gradually replacing traditional electronic sensors due to their small size, resistance to electromagnetic interference, remote detection, multiplexing, and distributed measurement advantages. This paper reviews the sensing principle, structural design, and temperature measurement performance of fiber-optic high-temperature sensors, as well as recent significant progress in the transition of sensing solutions from glass to crystal fiber. Finally, future prospects and challenges in developing fiber-optic high-temperature sensors are also discussed.

We propose a novel denoising method based on empirical mode decomposition (EMD) to improve the signal-to-noise ratio (SNR) for vibration sensing in phase-sensitive optical time domain reflectometry (φ-OTDR) systems. Raw Rayleigh backscattering traces are decomposed into a series of intrinsic mode functions (IMFs) and a residual component using an EMD algorithm. High frequency noise is eliminated by removing several IMFs at the position without vibration selected by the Pearson correlation coefficient (PCC). When the pulse width is 50 ns, the SNR of location information for the vibration events of 100 Hz and 1.2 kHz is increased to as high as 42.52 dB and 39.58 dB, respectively, with a 2 km sensing fiber, which demonstrates the excellent performance of this new method.

We have experimentally analyzed and compared the performance of optical frequency domain reflectometry (OFDR) assisted by image processing methods including wavelet transform and Gaussian filter arithmetic in terms of strain measurement range. The incomplete space similarity in the distributed strain measurement is used to recover the accurate strain gradient information by the image processing methods. The distributed spectrum shifts of the cross-correlation calculation along the sensing fiber between reference signal and measurement signal is regarded as a two-dimension (2D) image which is disposed by the proposed methods rather than extracting the spectrum shifts immediately. Compared with the traditional processing method, the proposed image processing method is able to remove and smoothen the fake peaks or outliers caused by the spatial mismatch of stretched fiber in OFDR system. In the experiments, strain gradient information up to 7000 μϵ can be identified by the proposed methods with a 4 mm spatial resolution. Furthermore, the wavelet transform is discussed in detail in order to determine the value of the suitable decomposition level in OFDR data processing scheme. The proposed methods provide a new idea to achieve the wide strain measurement range in OFDR system, which is a significant step toward a high-performance distributed sensing system, especially for occasions when a wide range of strain measurement is required.

Measurement accuracy of the optical frequency domain reflectometry (OFDR) will be deteriorated severely when the spatial resolution is improved due to the influence of the random noise, especially in long range distributed fiber optical sensing system. In order to eliminate the random noise, we have experimentally compared and analyzed the performance of OFDR system with and without the image processing methods in term of spatial resolution. The data arrays of the cross-correlation results are obtained after processing the collected distributed Rayleigh scattering signals along the sensing fiber, which can be used to construct a two-dimension (2D) image with the spectrum shift information induced by the strain. Thus, the image denoising methods can be applied to remove the random noise and improve the measurement accuracy. Based on this principle, two image denoising methods including the total variation method and the 2D Gaussian filter arithmetic are introduced to realize distributed sensing with high spatial resolution in long distance OFDR system. Experimental results show that the random noise can be suppressed and the outliers can be removed by the proposed methods. The strain gradients with high spatial resolution are identified effectively when the image denoising methods are used. Furthermore, spatial resolution with 1.3mm is achieved by the proposed methods with a wavelength sweeping rate of 20nm/s in 52m sensing fiber. The proposed methods provide a potential solution for OFDR system in long distance sensing with high spatial resolution, which is a significant step toward a high-performance OFDR system for practical applications.

Information sensor can realize the ubiquitous connection between objects and humans by the distributed fiber-optic sensor array, i.e., emerging infrastructure for Internet of Things, in which optical frequency domain reflectometry (OFDR) with its high spatial resolution plays an important role in achieving intelligent perception and identification of the objects. However, given a certain fiber length, there are tradeoffs among the processing speed, the spatial resolution, and the strain measurement range. In this article, a time optimization interpolation method and a distance domain compensation method are reported and experimentally verified to break the aforementioned tradeoffs for the first time. First, a full theoretical analysis on how to reduce the interpolation number and the reason why it is difficult to achieve high spatial resolution and large strain measurement and how to resolve it is made. Second in the proof-of-concept experiment, measurements of large strains up to 10 000 $\boldsymbol{\mu } \boldsymbol{\varepsilon }$ are realized along the sensing fiber with a spatial resolution of 2 mm using a conventional OFDR system. Also, the processing time is shortened by 76 times compared with the traditional processing method. This article makes a significant step toward high-performance OFDR system with fast processing, high spatial resolution, and wide strain measurement.

A curvelet denoising method has been proposed to reduce the time domain noise to improve the detection performance in the distributed fiber vibration sensing system based on phase-sensitive optical time domain reflectometry. The raw Rayleigh backscattering traces are regarded as a gray image and the random noise can be eliminated by the curvelet transform; hence, the amplitude difference induced by the external vibration can be extracted. The detection of a vibration event with 10 m spatial resolution along a 4 km single mode fiber is demonstrated. The signal-to-noise ratio (SNR) of location information for 50 Hz and 1 kHz vibration based on this new method increases to as high as 7.8 dB and 8.0 dB, respectively, compared to the conventional method, showing the remarkable denoising capability of this new approach.