Open Access
Add to BibSonomy
Abstract
In this work, we develop experimentally a Fabry-Perot fiber optic interferometer applied to the measurement of autocorrelation of complex dynamic pulses generated by a figure-eight fiber laser. The principle is based in the superposition of multiple pulses, which requires two partially reflecting flat surfaces in parallel, resulting in a simple and compact autocorrelator design. The autocorrelation trace obtained exhibits a typical double-scaled structure for noise-like pulses (NLPs), with an ultrashort coherence spur on the order of 100 fs riding upon a broad pedestal of 120 ps. Finally, we show experimentally that the developed Fabry-Perot device is able to measure accurately the autocorrelation of NLPs, as confirmed by comparing the measurement with that of a conventional autocorrelator scheme based on a Michelson interferometer, with the additional advantages of a more compact setup and a much easier alignment procedure compared to the latter.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Extrinsic Fabry-Perot Interferometers (FPI) have been investigated for several decades in applications related to sensors (for measuring temperature, curvature, refractive index, etc.) [1–3]. Recently, its applications of interest have focused on methods based on the self-referencing technique such as autocorrelation to characterize ultrashort pulses [4]. Advanced techniques exist, such as frequency-resolved optical gating [5], spectral phase interferometry for direct electric-field reconstruction [6] or modified-spectrum auto interferometric correlation [7] that can fully reconstruct the electric field of an optical pulse, however their implementation is quite complex. For this reason, the autocorrelator is still a convenient tool for estimating a pulse duration because of its simplicity, in particular it does not require advanced computing to retrieve the information.
In a Michelson interferometer-based autocorrelator, the pulse is split into two replicas and each of them is sent to two orthogonal delay arms. The pulses are reflected back at the end of each arm by normal incident mirrors or retroreflectors. Scanning is performed through translation of one of the mirrors. The two arms should be aligned such that two optical beams are recombined in parallel. On the other hand, an autocorrelator using a modified Mach-Zehnder interferometer scheme also requires two separate delay arms [8]. Previously, an in-line scanning autocorrelator was reported by swinging a birefringent plate rather than using translating mirrors [9]. Recently, an autocorrelator using double-wedge interferometer was proposed, which realizes the in-line configuration as a scanning type autocorrelator, however it cannot be used with a lens which is necessary for low power measurement, because angular dispersion is introduced by the wedge pair [10].
These techniques are of wide interest for the study of complex pulses, specifically pulses such as the so-called noise-like pulse (NLP), which is a ∼ns-long wave packet containing thousands of pico- or femtoseconds sub-pulses, displaying a very rich and complex internal dynamics at short time scale but exhibiting a stable overall behavior on large time scales [11]. Since 1997, the study of NLPs has been receiving significant research attention because of their extraordinary and distinctive features contrasted with the conventional mode-locked pulses [12]. In terms of fundamental research, these complex pulses constitute and ideal platform to study either fairly stable or very chaotic regimes simply by tuning polarization in the laser cavities [13]. In addition, NLPs allow energies per pulse up to 100s of nJ [14] or even µJ, which are thousands of times larger than soliton energy, because a NLP is not a simple pulse but a large bunch of radiation. NLPs are ubiquitous, as they are obtained in a wide variety of passively mode-locked laser architectures, operating at different wavelengths, in both normal and anomalous dispersion regimes [15,16], and using different kinds of fibers including standard and doped fibers, double-clad, and high-nonlinearity fibers (HNLF). Due to the distinctive features of NLPs presented above, several applications have been developed based on the fundamental mode locking operation, such as low-coherence spectral interferometry [17], micromachining [18], nonlinear frequency conversion [19], optical coherence tomography (OCT) [20], and supercontinuum generation [21].
In this work, a new type of autocorrelator using a Fabry-Perot interferometer is proposed that allows the detection of minimal variations, characterized by high portability and a more compact setup with a much easier alignment than a conventional autocorrelator. The Fabry-Perot Interferometer is fabricated by a silica capillary tube and cured UV resin, with the advance in cured UV polymers, the fabrication of these devices turned cost-effective and straightforward. The Fabry-Perot interferometer was applied to the measurement of autocorrelation of complex pulses generated by a figure eight fiber laser. We show that the Fabry-Perot device is able to measure accurately the autocorrelation of NLPs, with the advantages of a more compact setup and a much easier alignment procedure than a conventional autocorrelator based on a Michelson interferometer.
2. Fabrication of Fabry-Perot interferometer
The fabrication process of the Fabry-Perot cavity is not intricated. In this procedure, a single-mode fiber (SMF-28) is cleaned, cleaved, and aligned with a 1 µl capillary. At this point, a 3D translation stage (MBT610D) is used to align these elements. It is essential to mention that the capillary has a 200-µm inner diameter and a total length of 3.2 cm. The SMF-28 fiber and the capillary are aligned, the optical fiber is inserted into the capillary using the translation stage. Once the optical fiber reaches the middle of the capillary, a sensitive UV glue polymer is applied at the end of the capillary, where the buffer of the SMF-28 is exposed, then the glue polymer is cured by UV led. Afterwards, a second SMF-28 fiber is introduced at the other end of the capillary. Now, when a suitable cavity length is achieved, the sensitive Norland Optical Adhesive 68 (NOA68) is added and cured. The final optical fiber structure is presented in Fig. 1. The cavity length (L) is controlled by the “Z” plane of the 3D translation stage. Moreover, by using the “XY” positions, the cavity losses are minimized.
When the polymer is cured, an air quantity is encapsulated. As a result, an air microcavity is generated between the two flat fibers ends; here, refractive index interfaces are composed by silica fiber (nf = 1.4570) and air (n = 1.00). Then, when the input light arrives at the end of the fiber, a reflection R1 occurs; the transmitted light travels through the air cavity and arrives at the second SMF-28 fiber; here, a second reflection R2 is generated. The Fresnel relation can express these reflections as [22]:
Fig. 2. Spectrum response of the capillary low-coherence extrinsic Fabry-Perot Interferometer considering the estimated cavity length.
Figure 2 shows the experimental and computational responses of the capillary low-coherence extrinsic Fabry-Perot Interferometer considering the estimated cavity length. The difference between the computed and experimental traces presented in this figure is mainly due to the small variation of the FPI length and a misalignment due to the dimensions used in the experimental manufacturing process (order of µm), such is the case because the diameter of the capillary is much larger than the optical fiber used. In contrast, the numerical simulation assumes ideal conditions in order to estimate the behavior of the FPI.
3. Principle of operation of model FR-103MN autocorrelator
The FR-103MN was used as a reference to validate the results obtained in this work. This measuring equipment utilizes the second-harmonic generation (SHG) method of the 1st kind in a conventional Michelson Interferometer set-up to perform autocorrelation measurement. In the standard configuration, noncollinear beam interaction leads to background-free autocorrelation measurement. Periodic variation of the linear delay in one arm of the Michelson arrangement is introduced by a pair of parallel mirrors centered about a rotating axis. In the geometry of Fig. 3, the rotation of the mirror assembly leads to an increase (or decrease) of path length for the traversing beam. Thus, the transmitted pulse train is delayed (or advanced) with respect to its reference (zero delay) position. This delay varies with time as a function of the shaft’s rotation, which is linear and given by
where D is the distance between the mirrors, f is the frequency of rotation, and c is the speed of light. Rotation of the mirror assembly leads to a repetitive generation of linear delay which, used in the described SHG configuration, provides a continuous display of the autocorrelation function of the pulses on a conventional high impedance oscilloscope synchronized with this rotation.Fig. 3. Periodic variation of the linear delay in one arm of the Michelson arrangement in autocorrelator FR-103MN.
The FR-103MN can readily be used to cross correlate ultrashort pulses from two separate beams. The second beam for cross correlation (CC) is introduced in the center of the opening in the right- side plate.
Figure 4 shows a top and side schematic of the autocorrelator FR-103MN in order to understand the experimental arrangement used of the measurement equipment. The autocorrelator can also be used to cross-correlate ultrashort pulses from two separate beams, the scheme is shown in Fig. 5. The first beam is incident at the front aperture with the unit aligned as for autocorrelation (AC). The second beam is introduced through the aperture on the right-side panel in a direction opposite but parallel to the first, such that it is translated by the comer mirror to traverse the first beam’s path in autocorrelation. Set the integration time constant switch to 1 fs, and the motor speed to lowest (2.5 Hz). Fringe resolved autocorrelation will now be observed when the retroreflecting mirror on the mirrors side is properly aligned using the control knobs on the right-side panel, The beams can be given a small vertical deflection off the perfect retroreflection condition. This prevents feedback to the laser without affecting interferometric operation. The beams going back towards the laser form secondary spots on the semi-closed variable input aperture of the FR-103MN. These beams (from the two Michelson arm) should be set to overlap above or below the main incident beam which enters the center of the input aperture. Since the 1 fs position has low gain and integration, sufficiently high beam power may be needed, depending on the pulse width, wavelength, and crystal type. A small percentage of the output of a typical mode-locked Ti-Sapphire laser is usually sufficient using the KDP crystal. When the comer-minor is installed back for noncollinear background-free) operation, it should be positioned such that the edge of its mount lines up with a scratch marked on the T-stage.
4. Autocorrelation using a Fabry-Perot interferometer
The Fabry-Perot autocorrelator requires only two partially reflecting flat surfaces in parallel (Fig. 6). Thus, the optical layout is simpler and more compact than that of a Michelson autocorrelator: in the latter case, two arms perpendicular to each other are needed, whereas all optical components are aligned along a single optical axis in the case of a Fabry-Perot autocorrelator. An ideal Fabry-Perot interferometer produces an infinite number of copies of the pulses with multiple delays due to the two parallel reflecting surfaces. All those replicas are sent to the nonlinear medium such as a BBO crystal to generate the second harmonic like a conventional autocorrelator.
There are works where it is assumed that with non-dispersive plane-wave propagation, the autocorrelation trace of a Fabry-Perot autocorrelator looks similar to that of Michelson autocorrelator. In [25], the authors mention that a Fabry-Perot autocorrelator signal usually exhibits shorter duration than a Michelson autocorrelator signal for a given test pulse. Indeed, the phase mismatch among multiple pulses can be quickly increased as delay is getting away from zero, this response results from the coherent superposition of multiple pulses, in spite of these multiple interferences, the proportionality between the FWHM of a test pulse, and that of its Fabry-Perot autocorrelation is maintained.
In order to understand the mechanism for obtaining the autocorrelation trace, theoretical description of signal of FPI is used, exhibiting a shorter signal duration of FPI compared to a Michelson autocorrelator, while the phase miss-matching among multiple pulses can be rapidly increased as delay is getting far away from zero. Thus, the signal resulting from the coherent superposition of multiple pulses decays much faster than a Michelson autocorrelator signal. However, in spite of multiple interferences, the proportionality between FWHM of a test pulse is maintained, making it possible estimate the pulse duration from the autocorrelation duration by using the ratio, under the assumption of a certain pulse shape. When the reflectance approaches zero, the shape of a signal and the shape factor of FPI become similar to those of Michelson autocorrelator, because the effective number of transmitted pulses through FPI decreases drastically for the lower reflectance [25]. Finally, in order to obtain useful information in experimental situations, the chirped pulses due to the different propagation length of the material (e. g. fused silica) are compared. These statements will be validated in the results and conclusions section by using the autocorrelator FR-103MN.
5. Autocorrelation of noise-like pulses
Because of the complex nature and fast dynamics of NLPs, we can get a precise idea of their internal structure only by means of numerical simulations, because a precise experimental measurement able to unveil its internal details is not possible with the current optoelectronics; even with ultrafast oscilloscopes we can usually measure only the envelope (see Fig. 7). The autocorrelation trace (Fig. 7(b)) displays a double-scaled profile, with a broad pedestal and a narrow central coherence spur. The extension of the pedestal reflects the total duration of the waveform, whereas that of the narrow central peak reflects the duration of the solitons. Figure 7(c) shows the optical spectrum averaged over 10 pulses.
With respect to this last point, an appropriate characterization of NLPs is essential because in practice, due to the lack of information provided by conventional measurements, it can be confused with other pulsed regimes. Due to the duration of the NLPs (∼ns) and the absence or quasi-absence of temporal coherence through the packet, it has not been possible to characterize well its instantaneous frequency.
6. Experimental setup
The experimental setup is a F8L of about 215 m of length, formed by a ring cavity and a NOLM. In the ring section, a 980/1550 nm beam combiner was used to launch the pump power (6.5 W) from a 976 nm laser diode (Focus Light DLS03-FCMSE55-I-25-976-5) into the cavity. The active fiber is a piece of 1.6-m-long erbium/ytterbium double-clad fiber (EYDCF) with a core diameter of 12 µm (NA = 0.20) and an inner cladding diameter (flat-to-flat) of 130 µm (70-dB/m core absorption at 1530 nm). A polarization-dependent optical isolator (PD-ISO) is inserted to ensure unidirectional laser operation. The ring cavity also includes a section of single-mode fiber (SMF1, D = 18 ps/nm/km) of about 200 m of length and a polarization controller (PC) composed by QWR-HWR-QWR plates. The output port is provided by a 90/10 coupler, with the 90% output port connected to the combiner, closing the loop. A power-symmetric, polarization-imbalanced NOLM scheme is used as the SA, consisting of a 50/50 coupler, a 10-m-long SMF2 twisted at a rate of 5 turns per meter, and a QWR inserted asymmetrically in the loop in order to break the polarization symmetry. Figure 8 also shows the Fabry-Perot fiber optic interferometer inserted at the laser output, and the micro-curvature mechanism to control the multiple reflections in the interferometer, as well as the measurement equipment for visualizing the autocorrelation of NLPs.
For proper wave plate adjustments, self-starting mode locking takes place, resulting in the generation of NLPs with an average power of ∼40 mW at the fundamental cavity frequency. The FWHM of the NLPs generated by the F8L changes in the range shown in Fig. 9 when the HWR position is varied over the whole range of mode locking. The positioning of the QWR plate in the NOLM allows the self-starting of the pulsed regime in the F8L, while the adjustment of the QWR, HWR, QWR plates of the polarization controller (PC) allows to adjust the temporal duration of the pulses, as well as their spectral width at the laser output.
We observe NLPs envelopes with FWHM pulse durations between 5 ns to 19 ns and an estimated output peak power of 5 W. The pulse was measured in the time-domain using a 2-GHz photodetector and a 2-GHz oscilloscope, with a period of T = 1.1 µs, which corresponds to a fundamental repetition rate of 909 kHz, consistent with the ∼226-m cavity length. In order to further confirm the NLP regime, the optical spectrum and the autocorrelation trace were obtained. A typical NLP double-scaled autocorrelation trace was obtained, with a narrow coherence peak riding a wide and smooth pedestal of the order of ps, limited by the range of the autocorrelator. In spite of this, the pedestal presents a marked slope, and its extension is only of a few tens of ps (much narrower than the duration of the whole bunch), which is consistent with the existence of sub-ns substructures within the NLPs, as previously studied in [15]. The existence of these sub-packets (at an intermediate scale between the NLP coherence time and the total bunch duration) is confirmed by a slight modulation of the NLP envelopes in Fig. 9 (although the photodetector and scope bandwidth are not sufficient to clearly visualize them). These results will be discussed in detail in the following.
7. Results and discussions
In this section, we discuss the effect of a chirp on the autocorrelation signal of the FPI. In the case of FPI, there is also a similar feature on the correlation signal with respect to the propagation, the FWHM of the correlation signal increases significantly with amount of dispersion as one can see in Fig. 10. Thus, one needs to examine more carefully on the tail of correlation signal when dealing with a chirped pulse.
The great merit of FP autocorrelator is its simplicity, allowing for easy alignment. Based in autocorrelation signals of a chirped pulse propagating through a Fabry-Perot interferometer (see Fig. 6), we were able to calculate the dimensions of the manufactured interferometer obtained a length on the order of ∼99.15 µm, which coincides with the manufacturing dimensions presented in section 2.
Fig. 11. Comparison of autocorrelation trace between Fabry-Perot Interferometer and the Autocorrelator FR-103MN.
Figure 12 shows the outputs of F8L used as primary pulse profiles to test the scheme developed in this paper. Data from the autocorrelation traces and sampling oscilloscope measurements were also used to estimate the peak power of the pulses at Output 1. We estimated that the pulse peak power varies from ∼1.5 W for the narrow-bandwidth pulses to ∼9 W for wide-bandwidth pulses. In this case, we show the global autocorrelation traces of the pulses (Fig. 12 (a)) and a close-up view of the coherence peaks (Fig. 12 (b)). These results show that, the shorter the NLP and the higher its peak power, the narrower its coherence peak, which is consistent with common observations.
The operation specifications for the FPI used as autocorrelator were obtained based on the characteristics of the equipment used to validate the measurements made in this work. Obtaining different autocorrelation measurements with the FPI and comparing them with the FR-103MN, we obtain the following specifications: pulsewidth resolution: < 1 fs, minimum pulsewidth: ∼4 fs, maximum pulsewidth: ∼100 ps, scan range: > 195 ps, which are limited by the measuring equipment. Finally, we show that the noise-like pulses were successfully measured by using a Fabry-Perot interferometer for both transform limited pulse and chirped pulse demonstrating experimentally that a FPI can be used as an autocorrelator. The Fabry-Perot developed is able to measure the autocorrelation of NLPs, achieving a more compact setup with a much easier alignment than a conventional autocorrelator based on a Michelson interferometer.
8. Conclusion
In conclusion, we show that the Fabry-Perot developed is able to measure the autocorrelation of noise-like pulses. The autocorrelation signal shows that there is a linear relationship between the width of the autocorrelation signal and the real pulse width. The noise-like pulses were successfully measured by using a Fabry-Perot interferometer for both transform limited pulse and chirped pulse demonstrating experimentally that a Fabry-Perot interferometer can be used as an autocorrelator. With use of the measurements obtained through of an autocorrelator model FR-103XL, we can determinate some measurement specifications for the measurement of autocorrelation by the Fabry-Perot Interferometer. Finally, we obtained a more compact and accessible setup with a much easier alignment than a conventional autocorrelator based on a Michelson interferometer.
Funding
Consejo Nacional de Ciencia y Tecnología (CF-2023-G-109, Investigadores por Mexico (project 3155)).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. C. E. Lee, A. M. Markus, E. Udd, and H. F. Taylor, “Optical-fiber Fabry–Perot embedded sensor,” Opt. Lett. 14(21), 1225–1227 (1989). [CrossRef]
2. T. Li, R. G. May, A. Wang, and R. O. Claus, “Optical scanning extrinsic Fabry–Perot interferometer for absolute microdisplacement measurement,” Appl. Opt. 36(34), 8858–8861 (1997). [CrossRef]
3. Y.-J. Rao, “Recent progress in fiber-optic extrinsic Fabry–Perot interferometric sensors,” Opt. Fiber Technol. 12(3), 227–237 (2006). [CrossRef]
4. M. Schultze, E. Goulielmakis, M. Uiberacker, M. Hofstetter, J. Kim, D. Kim, F. Krausz, and U. Kleineberg, “Powerful 170-attosecond XUV pulses generated with few-cycle laser pulses and broadband multilayer optics,” New J. Phys. 9(7), 243 (2007). [CrossRef]
5. R. Trebino and D. J. Kane, “Using phase retrieval to measure the intensity and phase of ultrashort pulses: frequency-resolved optical gating,” J. Opt. Soc. Am. A 10(5), 1101–1111 (1993). [CrossRef]
6. C. Iaconis, V. Wong, and I. A. Walmsley, “Direct interferometric techniques for characterizing ultrashort optical pulses,” IEEE J. Sel. Top. Quantum Electron. 4(2), 285–294 (1998). [CrossRef]
7. D. A. Bender, J. W. Nicholson, and M. Sheik-Bahae, “Ultrashort laser pulse characterization using modified spectrum auto-interferometric correlation (MOSAIC),” Opt. Express 16(16), 11782–11794 (2008). [CrossRef]
8. P. Wasylczyk, “Ultracompact autocorrelator for femtosecond laser pulses,” Rev. Sci. Instrum. 72(4), 2221–2223 (2001). [CrossRef]
9. S. M. Kobtsev, S. V. Smirnov, S. V. Kukarin, and V. B. Sorokin, “Femtosecond autocorrelator based on a swinging birefringent plate,” Quantum Electron. 31(9), 829–833 (2001). [CrossRef]
10. J. An and D. E. Kim, “Compact in-line autocorrelator using double wedge,” Opt. Express 20(3), 3325–3330 (2012). [CrossRef]
11. J. C. Hernandez-Garcia, O. Pottiez, R. Grajales-Coutiño, B. Ibarra-Escamilla, E. A. Kuzin, J. M. Estudillo-Ayala, and J. Gutierrez-Gutierrez, “Generation of Long Broadband Pulses with a Figure-Eight Fiber Laser,” Laser Phys. 21(8), 1518–1524 (2011). [CrossRef]
12. C. Lecaplain and P. Grelu, “Rogue waves among noiselike-pulse laser emission: An experimental investigation,” Phys. Rev. A 90(1), 013805 (2014). [CrossRef]
13. Y. Jeong, L. A. Vazquez-Zuniga, S. Lee, and Y. Kwon, “On the formation of noise-like pulses in fiber ring cavity configurations,” Opt. Fiber Technol. 20(6), 575–592 (2014). [CrossRef]
14. J. P. Lauterio-Cruz, J. C. Hernandez-Garcia, O. Pottiez, J. M. Estudillo-Ayala, E. A. Kuzin, R. Rojas-Laguna, H. Santiago-Hernandez, and D. Jauregui-Vazquez, “High energy noise-like pulsing in a double-clad Er/Yb figure-of-eight fiber laser,” Opt. Express 24(13), 13778–13787 (2016). [CrossRef]
15. H. Santiago-Hernandez, O. Pottiez, M. Duran-Sanchez, R. I. Alvarez-Tamayo, J. P. Lauterio-Cruz, J. C. Hernandez-Garcia, B. Ibarra-Escamilla, and E. A. Kuzin, “Dynamics of noise-like pulsing at sub-ns scale in a passively mode-locked fiber laser,” Opt. Express 23(15), 18840–18849 (2015). [CrossRef]
16. E. Garcia-Sanchez, O. Pottiez, Y. Bracamontes-Rodriguez, J. P. Lauterio-Cruz, H. E. Ibarra-Villalon, J. C. Hernandez-Garcia, M. Bello-Jimenez, and E. A. Kuzin, “Complex dynamics of a fiber laser in non-stationary pulsed operation,” Opt. Express 24(17), 18917–18930 (2016). [CrossRef]
17. S. Keren and M. Horowitz, “Interrogation of fiber gratings by use of low-coherence spectral interferometry of noiselike pulses,” Opt. Lett. 26(6), 328–330 (2001). [CrossRef]
18. K. Özgören, B. Öktem, S. Yilmaz, FÖ Ilday, and K. Eken, “83 W, 31 MHz, square-shaped, 1 ns-pulsed all-fiber-integrated laser for micromachining,” Opt. Express 19(18), 17647–17652 (2011). [CrossRef]
19. S. V. Smirnov, S. M. Kobtsev, and S. V. Kukarin, “Efficiency of non-linear frequency conversion of double-scale pico-femtosecond pulses of passively mode-locked fiber laser,” Opt. Express 22(1), 1058–1064 (2014). [CrossRef]
20. Y. J. You, C. Wang, Y. L. Lin, A. Zaytsev, P. Xue, and C. L. Pan, “Ultrahigh-resolution optical coherence tomography at 1.3 µm central wavelength by using a supercontinuum source pumped by noise-like pulses,” Laser Phys. Lett. 13(2), 025101 (2016). [CrossRef]
21. J. P. Lauterio-Cruz, J. D. Filoteo-Razo, O. Pottiez, J. C. Hernandez-Garcia, Y. E. Bracamontes-Rodriguez, and H. Rostro-Gonzalez, “Numerical Comparative Study of Supercontinuum Generation in Photonic Crystal Fibers Using Noise-Like Pulses and Ultrashort Pulses,” IEEE Photonics J. 11(5), 1–12 (2019). [CrossRef]
22. K.C. Yee and R.R. Chance, “Synthesis and properties of a new polydiacetylene: Poly[1,6-di(N-carbazolyl)-2,4-hexadiyne],” J. Polym. Sci. Polym. Phys. Ed. 16(3), 431–441 (1978). [CrossRef]
23. Y. Zhao, Y. Yuan, W. Gan, and M. Yang, “Optical fiber Fabry–Perot humidity sensor based on polyimide membrane: Sensitivity and adsorption kinetics,” Sens. Actuators, A 281(1), 48–54 (2018). [CrossRef]
24. Y. Yu, X. Chen, Q. Huang, C. Du, S. Ruan, and H. Wei, “Enhancing the pressure sensitivity of a Fabry–Perot interferometer using a simplified hollow-core photonic crystal fiber with a microchannel,” Appl. Phys. B 120(3), 461–467 (2015). [CrossRef]
25. J. An, K. Pyun, O. Kwon, and D. E. Kim, “An autocorrelator based on a Fabry-Perot interferometer,” Opt. Express 21(1), 70–78 (2013). [CrossRef]
26. X. Zhang, H. Shao, H. Pan, Y. Yang, H. Bai, F. Pang, and T. Wang, “Simple capillary-based extrinsic Fabry–Perot interferometer for strain sensing,” Chin. Opt. Lett. 15(7), 070601 (2017). [CrossRef]
