Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

High-sensitivity optical fiber temperature sensor based on a dual-loop optoelectronic oscillator with the Vernier effect

Open Access Open Access

Abstract

In this paper, a high-sensitivity optical fiber temperature sensor based on a dual-loop optoelectronic oscillator (OEO) with the Vernier effect has been proposed and experimentally demonstrated. Different from the traditional dual-loop OEOs which comprise a very long loop and a short loop to achieve low-phase noise and single-mode selection, the proposed OEO scheme has two loops with slightly different lengths and does not use any RF filters. A part of the fiber in one of the loops is used as a temperature sensing element as well as the delaying component. An obvious Vernier effect has been generated in the frequency response of the OEO. By detecting the frequency shift of the envelope peak of the measured frequency response curve, the temperature sensing interrogation of the dual-loop OEO based sensor is conducted. The experimental results show that the sensitivity of the proposed dual-loop OEO based temperature sensor can be improved from 6.625 KHz/°C for a single-loop OEO to 210.25KHz/°C by employing the Vernier effect.

© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement

1. Introduction

Fiber-optic sensors have attracted extensive attention for their unique advantages such as light weight, small size, high sensitivity, anti-electromagnetic interference, good stability in harsh environments and so on [1]. In the past few decades, people have proposed various fiber optic sensors, such as fiber grating [2], Mach-Zehnder interferometers (MZIs) [3], Sagnac interferometer [4], and Fabry-Perot interferometers (FPIs) [5]. However, the demodulation of most optical fiber sensors is realized by monitoring the optical domain parameters such as optical power and wavelength via optical spectrum analyzer (OSA), which has not only low resolution but also slow demodulation speed. In recent years, optical fiber sensing system using microwave photonic technology has been widely studied. In this field, the sensing information of the fiber sensor can be demodulated in microwave domain to achieve better performance [67]. Among them, optoelectronic oscillators (OEOs) have attracted great interest for their ability to generate high-frequency microwave signals with high purity and stability, as well as their potential applications in optical fiber sensing [8]. A range of sensing applications based on OEO have been reported [910]. Due to the huge frequency gap between the light wave and the microwave, tiny variations in the optical domain will cause significant changes in the microwave domain. Meanwhile, because the electronic spectrum analyzer has much higher resolution than the OSA, the OEO based optical sensors would have higher sensitivity and resolution.

Vernier effect is a commonly used method to enhance the measurement accuracy. It uses two different scales, of which one slides along the other one. The measurement is based on the overlaps between lines on the two scales. It has been widely used in various types of optical fiber interferometer sensors, such as Sagnac interferometer [11] microfiber resonators [12] and Fabry-Perot sensors [1314], which have been successfully exploited to effectively improve the sensitivity of the sensor. The interrogation principle of this type of sensor is to track the spectrum envelope of the cascade interferometer sensors rather than the spectral response of a single sensor. The extreme value points of the measured optical spectrum response curve are fitted. Taking advantage of the Vernier effect, the frequency shift of the envelope peak is more obvious than that of a single sensor [15].

In this paper, an optical fiber temperature sensor system combines the OEO with the Vernier effect is firstly proposed and experimentally demonstrated. Firstly, an optical fiber temperature sensor based on single-loop OEO is analyzed and experimentally verified. To improve the sensitivity, we proposed a structure of the dual-loop OEO with Vernier effect, which uses a polarization beam splitter (PBS) to divide the light wave into two beams whose polarization directions are perpendicular to each other. The two beams then travel along two sections of optical fibers with different lengths. Each channel in the dual loop can be oscillated separately, however, since the length of the two channels is different, the mode interval of the oscillating signal is different. Finally, only oscillating modes that matches the oscillation frequency of the two channels can be enhanced [16], thus producing the Vernier effect. Unlike the optical fiber sensor system that uses the Vernier effect in the optical domain, the Vernier effect in our sensing scheme is generated in the electronic domain. The envelope curve of microwave frequency response is obtained by fitting the extreme value point of the measured frequency response curve. In our proposed optical fiber temperature sensing system, the temperature information can be determined by detecting the frequency shift of the envelope peak of the OEO’s microwave signal. With the advantages of the Vernier effect, the frequency shift of the envelope peak is more significant than that of a single-loop OEO, and thus the sensitivity has been greatly improved. Moreover, thanks to the high resolution of the electrical spectrum analyzing technique, the performance in terms of the resolution can be significantly increased. In general, the proposed optical fiber temperature sensor combines the advantages of microwave photonics and Vernier effect. The theoretical analysis of the proposed approach is presented and discussed. Theoretical and experimental results demonstrate that the proposed method is of high sensitivity and resolution.

2. Principle

The schematic diagram of the dual-loop OEO based temperature sensor system is shown in Fig. 1. The light wave emitted from a laser source is coupled into a Mach-Zehnder modulator (MZM). The modulated optical signal is then passed through an erbium-doped fiber amplifier (EDFA) to provide enough optical power for the photoelectric oscillator system. The optical microwave output is then sent to a fiber PBS and is divided into two orthogonal linear polarization components with approximately the same power by tuning the polarization controller (PC1). The two signals then travel along separate fiber paths with different lengths. One of the two paths is used as sensing fiber and the other path is used as reference fiber. The sensing fiber is placed in a thermostat while the reference fiber is isolated from the temperature variations. The two signals finally come together in a polarization beam combiner (PBC) via two PCs (PC2 and PC3) respectively and are finally converted into a corresponding microwave signal by using a high-speed photodetector (PD, Finisar XPDV2120RA). The microwave signals are amplified with a microwave amplifier (EA) and one part of the microwave signals is sent to the electrical spectrum analyzer (ESA, R&S FSV30) for detecting via a power divider (Div). The other part of the microwave signals is fed back to the MZM to form the OEO loop. It should be highlighted that for a standard OEO, a narrow bandpass filter is essential to effectively select the oscillation frequency of interest. However, for sensing application, the use of narrow bandpass filter will seriously limit the temperature measurement range due to the limitation of the filter bandwidth. Moreover, to realizing Vernier effect, the narrow bandpass filter cannot be used.

 figure: Fig. 1.

Fig. 1. Schematic diagram of the dual-loop OEO based sensors

Download Full Size | PDF

The working principle of the proposed temperature sensing system is described as follows. First, we consider the case of single loop OEO system whose schematic diagram is shown in Fig. 2. According to the working principle of the OEO, when the net gain of the OEO loop exceed 0 dB, microwave signals oscillation of the OEO will occur. The oscillation frequency of OEO depends on the overall delay of the loop, which is the sum of the electrical delay and the optical delay. As the length of the electrical links is assumed to be constant, therefore the change in optical lengths can be measured by detecting the frequency shift of the RF oscillation signals. The overall delay of OEO loop can be expressed as:

$${\tau _g} = {\tau _e} + {\tau _\textrm{o}}$$
where τg is the total time delay of the loop, τe is the delay time of the electrical delay line, τo is the delay time of the optical delay line. The free spectrum range (FSR) of the OEO signal and the frequency of the Nth harmonic (f0) can be expressed as:
$$FSR = \frac{1}{{{\tau _\textrm{g}}}} = \frac{1}{{{\tau _e} + {\tau _o}}} = \frac{c}{{n{L_0}}}$$
$$ f_0 = N \times FSR = N\displaystyle{c \over {nL_0}}$$
where c is the propagation speed of light in vacuum, n is the effective refractive index of the single mode fiber. and L0 is the global equivalent length of the oscillator loop which is the sum of those of the electronic part with a delay τe, and optical section with a delay τo. When the sensing fiber is heated, the global equivalent length of the OEO loop will change due to the physical length change and the refractive index change. Assume the length of the sensing fiber is Ls. At this moment, the FSR will be changed to:
$$\begin{array}{l} FS{R_s} = \frac{c}{{n{L_0}\textrm{ + }\Delta n{L_s}\textrm{ + }n\Delta {L_s}}}\\ \textrm{ = }\frac{c}{{n{L_0}\textrm{ + }n \cdot \mathrm{\xi} \cdot \Delta T \cdot {L_s} + n \cdot \mathrm{\alpha} \cdot \Delta T \cdot {L_s}}}\\ \textrm{ = }\frac{c}{{n{L_0}\textrm{ + (}\mathrm{\xi} \textrm{ + }\mathrm{\alpha} \textrm{)} \cdot \Delta T \cdot n \cdot {L_s}}} \end{array}$$
where α is the thermal expansion of silica, ξ is the thermo-optic coefficient and ΔT is the temperature variation. Correspondingly, by using Eqs. (3) and (4), the oscillating frequency shift of the Nth oscillation signal can be expressed as:
$$\Delta f = N \cdot ({FS{R_s} - FSR} )\approx N \cdot FS{R_s} \cdot \left( {\frac{{\textrm{(}\mathrm{\xi} \textrm{ + }\mathrm{\alpha} \textrm{)} \cdot \Delta T \cdot {L_1}}}{{{L_0}}}} \right)$$

In Eq. (5), since the temperature induced optical path change is much less than the global equivalent length of the oscillator loop, the oscillating frequency shift of the OEO can be simplified as linearly proportional to the temperature change. Thus, by measuring the frequency change, the temperature change can be estimated. However, the sensitivity of the single loop OEO sensing system is very low due to the extremely low thermal expansion (∼0.55×10−6/°C) and thermo-optical (∼6.45×10−6/°C) coefficients of silica. From Eq. (5), it is easy to find out that the sensitivity is also proportional to the FSR of the OEO. Therefore, to improve the sensitivity, an OEO architecture based on two loops employing the Vernier effect which can greatly increase the FSR is proposed. As shown in Fig. 1, a reference loop with slightly different length is added to the single loop OEO. The overlap between the two oscillating modes corresponding to the two loops is employed to perform the temperature measurement. The total oscillating modes of the dual-loop OEO is the product of the two individual ones, which exhibits peaks at frequencies where two resonant modes partially overlap. The reference loop is well isolated from the temperature variations, which acts as the fixed part of the Vernier-scale. The sensing loop is more like the sliding part of the Vernier scale, as the temperature changes will cause a shift of the resonant modes. Finally, according to the principle of Vernier effect [1214], the FSR of the dual-loop OEO (FSRd) can be derivation as follows:

$$FS{R_d} = \frac{{FS{R_r} \times FS{R_s}}}{{|{FS{R_r} - FS{R_s}} |}}$$
where FSRr denote the FSR of the reference loop. From Eq. (6). it is obvious that the FSR of the dual-loop OEO can be amplified if the values of the FSRr and FSRs are very close to each other. Thus, by choosing suitable length of reference loop, the FSR of the dual-loop OEO can be greatly amplified and accordingly, the sensitivity will also be greatly improved.

 figure: Fig. 2.

Fig. 2. Schematic diagram of the single-loop OEO based sensors

Download Full Size | PDF

3. Experimental investigation and results

Firstly, an experiment based on a single loop structure is carried out based on the setup shown in Fig. 2. A 240 meters long fiber is used as the sensing head and is placed in a thermostat for temperature change test. The initial temperature of the thermostat is set to 20°C. As shown in Fig. 3, a number of oscillation modes are generated at the output of the OEO. When the temperature is increased, it is known from Eq. (3) and Eq. (4) that all microwave oscillation modes will experience frequency shift. In addition, each mode will have a different frequency shift, namely, the different sensing sensitivity. In our experiment, an oscillation mode around 521.7 MHz is chosen for temperature measurement. During the temperature test, the temperature of the sensing fiber is increased from 20°C to 100°C with a step of 10°C by controlling the temperature of the thermostat, and the frequency of the traced microwave signal at different temperature is recorded. The waiting time at each temperature is about 20 minutes to ensure the temperature of the thermostat become stable.

 figure: Fig. 3.

Fig. 3. Measured spectrum of microwave oscillation modes in a single loop OEO

Download Full Size | PDF

Figure 4(a) shows the superposition spectrum of the tracked microwave signals at different temperatures. As can be seen in Fig. 4(a), as the temperature of the sensing fiber increases from 20°C to 100°C, the tracked microwave frequency decreases from 521.7 MHz to 521.2 MHz. This phenomenon indicates that the length of the sensing fiber increases with the increase of the temperature, which leads to the increase of the time delay of the OEO loop, thus the frequency of the tracking microwave signal will shift to the lower frequency range. Figure 4(b) shows the frequency of the measured microwave signal as a function of the temperature applied to the sensing fiber. Consequently, an excellent linear relationship between the frequency shift and the temperature is observed and the sensing sensitivity is derived as −6.625 KHz/°C. In our experiment, the FSR is measured to be 676KHz, as a result, the measurement range is restricted to 100°C.

 figure: Fig. 4.

Fig. 4. (a) Superimposed electrical spectra of the generated microwave signal at different temperatures. (b) Measured oscillation frequency shift as a function of the applied temperature to the sensing fiber from 20°C to 100°C.

Download Full Size | PDF

Secondly, to compare with the temperature sensor based on single-loop OEO, the temperature sensing characteristics of the dual-loop OEO are tested. In this case, a 231meters long reference single mode fiber is added to the single-loop OEO to form a dual-loop OEO, as shown in Fig. 1. In the experiment, the frequency response is measured when the temperature of the sensor fiber is changed while the temperature of the reference fiber is kept constant. Figure 5 shows the measured frequency response of the two single-loop OEO and the dual-loop OEO at 20°C respectively. From Fig. 5(a) and Fig. 5(b), we can find that the values of the FSR of the sensing loop (676KHz) is very close to that of the reference loop (698KHz). To obtain the envelope curve of the frequency response of the dual-loop OEO, the extreme value points of the measured frequency response curve are fitted. As shown in Fig. 5(c), the frequency response curve of the optical fiber temperature sensor based on the dual-loop OEO shows obvious Vernier effect. The FSR of the dual-loop OEO is extended to 21.4 MHz, which agrees well with the theoretical estimate about 21.447 MHz and is much larger than that of the single-loop OEO.

 figure: Fig. 5.

Fig. 5. Frequency response of the OEO (a) only sensing loop (b) only reference loop (c) dual-loop structure and fitted envelope

Download Full Size | PDF

Similarly, an oscillation mode with the same harmonic order as that of the single loop OEO is chosen for temperature measurement. At this time, the frequency of the microwave signal tracked for temperature measurement is shifted to 542.5 MHz. The frequency responses of the traced microwave signal at different temperature are recorded.

Figure 6 shows the relationship between the frequency of the tracked microwave signal and the temperature of the sensing fiber. Figure 6(a) shows the superposition spectra of the envelope on the spectral response curve at different temperatures. As can be seen from Fig. 6(a), as the temperature of the sensing fiber increases from 20°C to 100°C, the peak frequency of the tracked spectral response envelope decreases from 542.5 MHz to 525.2 MHz. Figure 6(b) shows the frequency of the envelope peak is a function of the temperature applied to the sensing fiber. By linear fitting of the measurement data in Fig. 6(b), the sensitivity of the sensor can be reached to −201.25KHz/°C. Compared with the sensor based on single-loop OEO, the sensitivity of the sensor based on dual-loop OEO is improved by 30 times, which demonstrates that the Vernier effect produced by the dual loop is an effective way to improve the sensing sensitivity.

 figure: Fig. 6.

Fig. 6. (a) Measured superimposed electrical spectra of the frequency response of the sensor under different temperature (envelope of the curve) (b) Measured envelope peak frequency shift as a function of the applied temperature to the sensing fiber from 20°C to 100°C.

Download Full Size | PDF

The resolution of the proposed method is limited by the frequency measurement resolution (∼1 Hz) of the ESA for a fixed sensitivity and the measurable range of the method is limited by the FSR. In our experiment, the FSR of the dual-loop OEO is about 21.4 MHz. As a result, the measurement range is restricted to be 106°C. Therefore, there usually exist a trade-off between the sensitivity and the measurable range. However, if we detect the change of the FSR value, the measurable range of this proposed method can be considered as unlimited because there is a one-to-one correspondence between the value of the FSR and the temperature. The stability of the generated signal for sensing is then studied by recording the central frequencies of the envelope every 10 minutes for 2 hours at 40°C and 60°C respectively. The variation of the frequency is within 10KHz for both temperatures, showing good frequency stability. During the process, slight amplitude variations less than 1dBm of the envelope can be detected. However, the interrogation scheme is immune to the power fluctuation because of the characteristic of the frequency encoding.

It's important to point out that although other optical fiber sensors by using Vernier effect have been studied before. Our proposed method still has advantages over them. First, for fiber-optic interferometers based sensors by using Vernier effect, the detection is usually realized by monitoring the optical domain parameters such as optical power and wavelength via OSA, which has not only low resolution but also slow demodulation speed. In comparison, our proposed method is performed in the electrical domain. Thanks to the high speed and high resolution of the electrical spectrum analyzing technique, the interrogation performance has been greatly improved. Second, for microwave photonic filter (MPF) based fiber-optic sensor with Vernier effect, the sensing interrogation is conducted by detecting the frequency shift of the upper envelope of the measured frequency response curve, which is similar to the method proposed in our work. However, the envelope of the measured frequency response curve in our work has significantly sharper spectrum compared with the MPF based method, which ensures sharper resolution and higher accuracy. Moreover, a complicated Vector Network Analyzer (VNA) is essential for measuring the transmission frequency response of the MPF (S21 character). In contrast, our solution requires only simple spectrum analysis and can be realized through FFT based on a DSP unit.

4. Conclusion

In summarize, an optical fiber temperature sensor based on a dual-loop OEO with Vernier effect has been proposed and experimentally demonstrated. The fundamental principle of the work is to translate the temperature change to the oscillating frequency shift of the OEO. By employing two optical fiber loops with slightly different lengths, the Vernier effect can be generated in the frequency response of the OEO. The experimental results show that, with the help of the Vernier effect, the temperature sensitivity has been greatly improved compared to the temperature sensor system based on single-loop OEO. Moreover, the sensitivity amplification factor can be easily adjusted by appropriately setting the length difference between the two loops. It should be noted that since the basic concept of our proposed method is to convert the optical length change of the OEO loop into the frequency shift of the generated microwave signal. Besides temperature, this method can also be used to measure other physical parameters such as strain, transverse load, refractive index since they can also cause the length change of the OEO loop.

Funding

National Natural Science Foundation of China (61975082); Major Project of Nature Science Research for Colleges and Universities in Jiangsu Province (16KJA510001).

Disclosures

The authors declare no conflicts of interest.

References

1. T. G. Giallorenzi, J. Bucaro, A. Dandridge, G. Sigel, J. Cole, S. Rashleigh, and R. Priest, “Optical fiber sensor technology,” IEEE Trans. Microwave Theory Tech. 30(4), 472–511 (1982). [CrossRef]  

2. A. D. Kersey, M. A. Davis, H. J. Patrick, and M. LeBlanc, “Fiber grating sensors,” J. Lightwave Technol. 15(8), 1442–1463 (1997). [CrossRef]  

3. M. Sun, Y. Jin, and X. Dong, “All-Fiber Mach–Zehnder Interferometer for Liquid Level Measurement,” IEEE Sens. J. 15(7), 3984–3988 (2015). [CrossRef]  

4. X. He, C. Ma, and X. Wang, “Pressure vector sensor based on an orthogonal optical path Sagnac interferometer,” Opt. Express 28(6), 7969–7979 (2020). [CrossRef]  

5. B. Zhou, C. Lu, and B. Mao, “Magnetic field sensor of enhanced sensitivity and temperature self-calibration based on silica fiber Fabry-Perot resonator with silicone cavity,” Opt. Express 25(7), 8108–8114 (2017). [CrossRef]  

6. Y. Wang, J. Zhang, and J. Yao, “An optoelectronic oscillator for high sensitivity temperature sensing,” IEEE Photonics Technol. Lett. 28(13), 1458–1461 (2016). [CrossRef]  

7. Y. Wang, X. Ni, M. Wang, Y. Cui, and Q. Shi, “Demodulation of an optical fiber MEMS pressure sensor based on single bandpass microwave photonic filter,” Opt. Express 25(2), 644–653 (2017). [CrossRef]  

8. X. Yao and L. Maleki, “Optoelectronic microwave oscillator,” IEEE J. Quantum Electron. 32(7), 1141–1149 (1996). [CrossRef]  

9. L. D. Nguyen, K. Nakatani, and B. Journet, “Refractive index measurement by using an optoelectronic oscillator,” IEEE Photonics Technol. Lett. 22(12), 857–859 (2010). [CrossRef]  

10. Y. Zhu, X. Jin, and H. Chi, “High-sensitivity temperature sensor based on an optoelectronic oscillator,” Appl. Opt. 53(22), 5084–5087 (2014). [CrossRef]  

11. L. Shao, Y. Luo, Z. Zhang, X. Zou, and B. Luo, “Sensitivity-enhanced temperature sensor with cascaded fiber optic Sagnac interferometers based on Vernier-effect,” Opt. Commun. 336, 73–76 (2015). [CrossRef]  

12. Z. Xu, Q. Sun, B. Li, Y. Luo, W. Lu, and D. Liu, “Highly sensitive refractive index sensor based on cascaded microfiber knots with Vernier effect,” Opt. Express 23(5), 6662–6672 (2015). [CrossRef]  

13. P. Zhang, M. Tang, F. Gao, B. Zhu, and S. Fu, “Cascaded fiber-optic Fabry-Perot interferometers with Vernier effect for highly sensitive measurement of axial strain and magnetic field,” Opt. Express 22(16), 19581–19588 (2014). [CrossRef]  

14. Y. Zhao, P. Wang, R. Lv, and X. Liu, “Highly Sensitive Airflow Sensor Based on Fabry–Perot Interferometer and Vernier Effect,” J. Lightwave Technol. 34(23), 5351–5356 (2016). [CrossRef]  

15. T. Claes, W. Bogaerts, and P. Bienstman, “Experimental characterization of a silicon photonic biosensor consisting of two cascaded ring resonators based on the Vernier-effect and introduction of a curve fitting method for an improved detection limit,” Opt. Express 18(22), 22747–22761 (2010). [CrossRef]  

16. X. S. Yao and L. Maleki, “Multiloop optoelectronic oscillator,” IEEE J. Quantum Electron. 36(1), 79–84 (2000). [CrossRef]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (6)

Fig. 1.
Fig. 1. Schematic diagram of the dual-loop OEO based sensors
Fig. 2.
Fig. 2. Schematic diagram of the single-loop OEO based sensors
Fig. 3.
Fig. 3. Measured spectrum of microwave oscillation modes in a single loop OEO
Fig. 4.
Fig. 4. (a) Superimposed electrical spectra of the generated microwave signal at different temperatures. (b) Measured oscillation frequency shift as a function of the applied temperature to the sensing fiber from 20°C to 100°C.
Fig. 5.
Fig. 5. Frequency response of the OEO (a) only sensing loop (b) only reference loop (c) dual-loop structure and fitted envelope
Fig. 6.
Fig. 6. (a) Measured superimposed electrical spectra of the frequency response of the sensor under different temperature (envelope of the curve) (b) Measured envelope peak frequency shift as a function of the applied temperature to the sensing fiber from 20°C to 100°C.

Equations (6)

Equations on this page are rendered with MathJax. Learn more.

τ g = τ e + τ o
F S R = 1 τ g = 1 τ e + τ o = c n L 0
f 0 = N × F S R = N c n L 0
F S R s = c n L 0  +  Δ n L s  +  n Δ L s  =  c n L 0  +  n ξ Δ T L s + n α Δ T L s  =  c n L 0  + ( ξ  +  α ) Δ T n L s
Δ f = N ( F S R s F S R ) N F S R s ( ( ξ  +  α ) Δ T L 1 L 0 )
F S R d = F S R r × F S R s | F S R r F S R s |
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.