Random Fiber Grating Characterization Based on OFDR and Transfer Matrix Method

Zichao Zhou, Chen Chen, Ping Lu, Stephen J. Mihailov, Xiaoyi Bao2020 · DOI

TL;DR

This paper tackles the challenge of quantitatively understanding how the "randomness" of random fiber gratings (RFGs) affects their optical properties. By combining experimental measurements with optical frequency domain reflectometry (OFDR) and theoretical simulations using the transfer matrix method, the authors establish a clear relationship between RFG randomness (quantified by entropy) and its reflectivity, paving the way for better design and application of these gratings in sensors and lasers.

What Problem Does This Paper Address?

Random fiber gratings (RFGs) are gaining traction for their ability to create broadband reflection spectra, making them useful in applications like fiber optic sensors and random fiber lasers. Unlike regular fiber Bragg gratings (FBGs) with precisely defined periodic structures, RFGs have intentionally introduced random variations in their grating periods. While this randomness is key to their functionality, there's been a lack of quantitative understanding of how different degrees of randomness impact their spectral characteristics and overall performance. This makes it difficult to optimize manufacturing processes and predict RFG behavior for specific applications. The paper aims to bridge this gap by establishing a quantifiable link between the degree of randomness in an RFG and its measurable optical properties.

Methods

The researchers employed a two-pronged approach: experimental characterization and theoretical simulation.

  • Experimental Characterization: They used Optical Frequency Domain Reflectometry (OFDR) to precisely measure the reflectivity of two RFGs with varying degrees of randomness. OFDR offers high spatial resolution, allowing them to analyze the grating's response along its length. This involved:
  • Measuring the backscattering strength and its fluctuations in the spatial domain.
  • Analyzing the local spectral response at different points along the grating.
  • Investigating the linewidth scaling behavior of spectral features with grating length.
  • Theoretical Simulation: To establish a quantitative relationship, they used the transfer matrix method. This method allows for simulating the optical properties of gratings by breaking them down into smaller sub-gratings and calculating how light propagates through them.
  • They introduced entropy as a mathematical measure for the degree of randomness, specifically related to the variation in sub-grating periods.
  • They simulated the average reflectivity and peak reflectivity of RFGs as a function of this entropy.

Datasets

The paper does not describe the use of pre-existing datasets. Instead, it focuses on the characterization of two custom-fabricated RFGs with different degrees of randomness. These RFGs served as the "datasets" for their experimental evaluation.

Key Insights and Ideas

  • Randomness impacts reflectivity: Higher degrees of randomness in RFGs lead to lower average backscattering strength and more pronounced intensity fluctuations in the spatial domain.
  • Local spectral complexity: RFGs with higher randomness exhibit more complex local spectral responses, featuring multiple peaks and a wider range of peak wavelengths.
  • Entropy as a randomness metric: The concept of entropy provides a robust mathematical framework to quantify the degree of randomness in RFGs, specifically linked to the variation in sub-grating periods.
  • Sub-grating entropy matters most: The average reflectivity of an RFG is primarily determined by the entropy of its individual sub-gratings, rather than the overall entropy of the entire grating. This highlights the localized nature of the randomness's impact.
  • New criteria for peak reflectivity: The paper proposes that the peak reflectivity of RFGs is determined by the term $\kappa^2L\Delta P$ (where $\kappa$ is the coupling coefficient, $L$ is grating length, and $\Delta P$ is the period variation range) rather than just $\kappa L$, especially when $\Delta P$ is significant (greater than 8 nm). This offers a new way to estimate RFG properties.

Key Results

  • Experimental:
    • A "high disordered RFG" showed an average backscattering strength of -54 dBV/mm, significantly lower than the "low disordered RFG" at -34 dBV/mm, despite identical refractive index modulation and grating length.
  • The high disordered RFG exhibited peak-to-peak reflectivity fluctuations of 35 dBV/mm, compared to 24 dBV/mm for the low disordered RFG.
  • Local spectral responses of high disordered RFGs showed multiple peaks even at short interrogation lengths (40 µm), whereas low disordered RFGs required longer lengths (200 µm) to show multiple peaks.
  • The standard deviation of distributed peak wavelength was 6.72 nm for low disordered RFGs and 26.65 nm for high disordered RFGs, indicating much larger wavelength fluctuations with higher randomness.
  • The linewidth of fine spectral features scales inversely with grating length.
  • Simulation:
    • A linear relationship was found between the average reflectivity (in dB scale) and the sub-grating's entropy: reflectivity decreases linearly as entropy increases, particularly when the measurement wavelength range is smaller than the sub-grating's central wavelength fluctuation range.
  • The simulation results for the average reflectivity difference between the two RFGs (~20 dB) closely matched experimental observations.

Limitations and Future Work

  • Limitations:
    • The simulation's wavelength range was limited to 1500-1600 nm, which influenced the observed reflectivity trends.
  • The assumption of uniform distribution for random sub-grating periods in simulations might slightly overestimate the entropy for the experimental RFGs, as their period variations might have been more focused.
  • The paper notes that the criteria for estimating RFG properties based on $\kappa^2L\Delta P$ is particularly relevant when $\Delta P$ is larger than 8 nm; for smaller $\Delta P$, the behavior is more akin to regular FBGs.
  • Future Work: The authors suggest that their findings provide a better understanding of RFG structures and can help optimize manufacturing processes for future applications in random fiber lasers and sensors. The established quantitative relationships are crucial for designing RFGs with tailored properties.

Summary updated

BibTeX
@misc{zhou2020random,
  author = {Zichao Zhou and Chen Chen and Ping Lu and Stephen J. Mihailov and Liang Chen and Xiaoyi Bao},
  title = {{Random Fiber Grating Characterization Based on OFDR and Transfer Matrix Method}},
  year = {2020},
  doi = {10.3390/s20216071},
}

Loading related content...

Figure