Open Access
Add to BibSonomyAbstract
A high-sensitivity optical time-domain reflectometry based on Brillouin dynamic grating (BDG) is proposed and experimentally demonstrated in polarization-maintaining fibers, where a single-end access to a fiber under test is applied with co-propagation of pump and probe pulses for the operation of BDG. Distributed measurements of the BDG spectra are presented with 80 cm spatial resolution in 935 m range, showing strain and temperature sensitivities of 1.37 MHz/με and −57.48 MHz/°C, respectively.
©2012 Optical Society of America
1. Introduction
Brillouin dynamic grating (BDG) has recently attracted considerable attention as an emerging technology for fiber-optic sensing and signal processing [1–12], and extensive researches have been conducted on its fundamental properties and potential applications using polarization maintaining fibers (PMF’s) [1–8, 10, 11] and conventional single-mode fiber’s (SMF’s) [9, 12]. In particular distributed fiber sensors based on BDG can provide the advantages of high-sensitivity [2, 3, 5] or high-resolution [6, 7] in the monitoring of strain or temperature variations through the spectral analysis of the BDG reflection. In all the former experiments of BDG, bidirectional access to a fiber under test (FUT) has been applied, where Brillouin pump waves need to be counter-propagated from both ends of the FUT to locally generate BDG’s by way of stimulated Brillouin scattering (SBS) between the pump waves along with the propagation of a probe wave for the acquisition of the reflection spectra. In this paper, a novel type of optical time-domain reflectometry based on BDG (BDG-OTDR) is proposed with single-end access to a FUT in which the generation of BDG is achieved only by a single pump pulse co-propagated with a probe pulse for measuring the local BDG spectra. In the experimental confirmation, several advantageous features such as low operating power, high sensitivity, and simple process for data acquisition are demonstrated with 80 cm spatial resolution in 935 m range, together with the description of the operation principle and some limiting factors in applications.
2. Principle
Figure 1(b) shows the schematic of BDG-OTDR in comparison with that of ordinary Brillouin OTDR (BOTDR) in Fig. 1(a). In BOTDR, spontaneous Brillouin scattering of a pump pulse is measured in time-domain, and a microwave detection system is used for high-speed detection of local Brillouin frequencies. The spatial resolution and the signal amplitude are necessarily determined by the width and the amplitude of the pump pulse, respectively. Meanwhile, a relatively long pump pulse and a short probe pulse are co-propagated with orthogonal polarizations in the proposed BDG-OTDR. Acoustic waves (BDG’s) are generated by the amplified spontaneous Brillouin scattering (ASBS) of the pump pulse, and local spectra of the BDG’s are analyzed by the probe pulse.
Fig. 1 Schematics of (a) ordinary BOTDR, and (b) BDG-OTDR: PBS, polarization beam splitter; PD, photo detector.
The reflectance of the BDG linearly depends on the product of the powers of counter-propagating pump1 and pump2 waves based on a moving FBG approximation [8]. In BDG-OTDR, the role of pump2 is performed by the spontaneous Brillouin scattering of a single pump, the power of which linearly depends on the product of the power and the width of the pump. Therefore, the signal amplitude (S) of BDG-OTDR is determined by the following relation (Ppump: pump power, Pprobe: probe power, wpump: width of pump pulse, wprobe: width of probe pulse):
While the spatial resolution is determined by the width of the probe pulse, the signal amplitude is controlled by not only the amplitude of the probe but also the width and the amplitude of the pump in Eq. (1). This feature provides an easy and indirect way to increase the signal amplitude without using a high power pulse, enhancing a signal to noise ratio of the measurement.When compared to the bidirectional schemes, BDG-OTDR can provide a remarkable advantage (besides the single-end access to a FUT) that the analysis of the BDG spectra does not require the information on the Brillouin frequency (νB) map along a FUT. In the bidirectional approaches [2, 4], the map of νB is prerequisite to the BDG measurement to set the frequency offset between counter-propagating pump waves to the νB of the FUT for BDG generation by the SBS of the pump waves. The system could suffer the fluctuation of the signal amplitude when large variation of νB exists in the fiber. If the change of νB within the fiber exceeds the Brillouin gain bandwidth (~30 MHz), one need to perform multiple measurements to avoid the extinction of signal [2]. On the contrary, such a problem is possibly avoided in BDG-OTDR by the use of the ASBS of a single pump, which helps reducing the complexity of measurement process and eventually enhancing the practicality.
3. Experiments
Figure 2 depicts the experimental setup for BDG-OTDR, in which two 1550 nm laser diodes (LD1, LD2) were used as light sources for the pump and the probe, respectively.
Fig. 2 Experimental setup of the BDG-OTDR: LD, laser diode; EDFA, Er-doped fiber amplifier; EOM, electro-optic modulator; PBS, polarization beam splitter; FBG, fiber Bragg grating; DAQ, data acquisition module.
A 200 ns square pulse generated by an electro-optic modulator (EOM1) and a pulse generator was used for the pump, which was amplified by an Er-doped fiber amplifier (EDFA1) before launched to a FUT through the slow axis by a polarization beam splitter (PBS). For the probe, a bell-shaped pulse with a full-width at half maximum (FWHM) of 8 ns was generated by an EOM (EOM2) and the pulse generator, which was amplified by an EDFA (EDFA2) before launched to the FUT through the fast axis by an optical circulator and the PBS. The pulse repetition rate was 50 kHz, and the pulse peak power at the front end of the FUT was about 0.5 W for the pump and 1 W for the probe, respectively. The reflection signal of the probe was received by a 125 MHz photo detector (PD) and a 500 MSa/s data acquisition module (DAQ), after pre-amplified by an EDFA (EDFA3) and filtered by a fiber Bragg grating (FBG).
The structure of the FUT is shown in the inset of Fig. 2, which is a 952 m PMF (bow-tie fiber) with the cladding diameter of 80 μm and the νB of ~10.4 GHz. A 1 m section near the end of the fiber (z = 922 m) was strain- or temperature- controlled for test measurements, the length of which was determined considering the nominal spatial resolution of ~80 cm. The propagation of the pump and the probe pulses was synchronized as shown in Fig. 3(a) to maximize the reflection from BDG [7]. Examples of the reflection spectra from the FUT are depicted in Fig. 3(b) measured by an optical spectrum analyzer after EDFA3, which are two different cases (black and red) of the frequency offset (Δf) between the pump and the probe. Besides strong Rayleigh scattering of the pump and the probe, Stokes wave of the pump is seen in both cases. It should be noted that the pump power was set near but slightly below the SBS threshold which was is estimated to be ~1 W by the following equation [13]:
where gB = 5x10−11 m/W, Aeff = 50 μm2, and Leff = 20 m (covered by the pump pulse). The optimum power level of the pump pulse needs to be experimentally confirmed as depicted in Fig. 3(b), where one can see the reflection spectrum of the pump wave with the Rayleigh scattering about 8 dB stronger than the Stokes reflection by the Brillouin scattering. The system suffers pump depletion if the power is too high, and weak signal if the power is too low. The BDG reflection of the probe is observed in one (red) of the cases, where Δf is set to the BDG frequency (Δν) of ~63.4 GHz. The birefringence Δn is calculated to be ~4.74x10−4 by the BDG equation [1, 4]:where ν is the optical frequency of the pump and ng is the group index of the fiber. It is remarkable that there is a large difference (~10 dB) in the amplitude of the BDG reflection depending on whether the probe is within (red) or out of (black) the BDG spectrum. It is confirmed that the peak power of the probe is also below the threshold of SBS, considering similar amplitudes of its Stokes and anti-Stokes waves when the probe is out of the BDG spectrum (black).Fig. 3 (a) Oscilloscope traces of the pump and the probe pulses measured at the end of the FUT (b) Reflection spectra from the FUT measured by an optical spectrum analyzer located after EDFA3. The black and the gray curves correspond to the cases of the probe in and out of the BDG spectrum, respectively.
Distributed measurements of the BDG spectra were carried out sweeping Δf in the vicinity of 63.4 GHz by tuning LD1 with LD2 fixed. This control was helpful in obtaining clearer BDG spectra through the FBG filter than tuning LD2 with LD1 fixed. The sweeping span of Δf was 1.67 GHz with a step of 3.7 MHz, and the traces obtained by the DAQ were averaged 1024 times for noise suppression. Figure 4(a) shows two examples of the measured traces at different Δf’s, where large fluctuations are observed according to position and Δf, similar to those observed in the measurement of the bidirectional scheme [2], that could be attributed to local variations of the birefringence in the fiber. The obtained time traces are merged to construct the 3D distribution of the BDG spectra along the FUT as depicted in Fig. 4(b), in which the BDG reflection signal strongly attenuates according to the propagation distance. The inset shows the normalized BDG spectra near the front (26 m, blue) and the rear (888 m, red) ends of the FUT for comparison, where the spectral width (FWHM) of ~110 MHz is maintained. This result confirms that the spectral broadening of the probe pulse by self-phase modulation was negligible in the propagation.
Fig. 4 (a) Trace examples at different Δf’s in distributed measurements of the BDG spectra by BDG-OTDR. (b) 3D-distribution of the measured BDG spectra along the FUT. The inset is the BDG spectra near the front (26 m) and the rear (888 m) ends. Note z-axis corresponds to the signal amplitude. (c) The maximum amplitude of the BDG spectrum according to position. Note that the red curve is the result of exponential fit.
Figure 4(c) is the plot of the maximum amplitude of the traces as a function of position, which is constructed from Fig. 4(b) by collecting only the peak values of local BDG spectra. The graph shows a regular decrease of the signal amplitude which fits well with an exponential decay (red). The length for 3 dB attenuation is estimated to be ~211.4 m, much shorter than those (several kilometers at least) of ordinary BOTDR systems, which corresponds to ~14 dB/km. The FUT was a small-core PMF specially designed for sensing application, and the measured propagation loss by Rayleigh scattering was 2.3 dB/km. Considering Eq. (1) and the roundtrip of the signal, the propagation loss of 2.3 dB/km provides the signal attenuation of 9.2 dB (4 x 2.3)/km, to which the observed large signal loss can be mainly attributed. The additional 5 dB loss seems to have come from the effect of pump depletion induced purely by the BDG-operation in the portion overlapped with the probe, and further research is needed for the quantitative explanation [8].
The measurement results of the Δν distribution along the FUT are shown in Fig. 5(a) with a zoomed view around the test section (dashed box) depicted in Fig. 5(b), which are plots of 5 consecutive measurements with different strain (0 to 200 με) applied to the test section. One can see fluctuations of Δν within the fiber spool (~918 m), drop in the loose section of 3 m (918 m ~921 m), and gradual increase in the test section (921 ~922) by the applied strain. The differences of Δν from the first measurement (i.e. 0 με) are plotted in Fig. 5(c) showing the relative shift of Δν. The maximum deviation between the consecutive measurements (i.e. measurement error) is ± 5 MHz, almost uniform along the position in spite of the strong attenuation of the signal along the position. This feature shows that the current measurement error is limited by the frequency drifts of the LD’s. Figure 5(d) is a zoomed view of the dashed box in Fig. 5(c) showing the positive shift of Δν at the test section by the strain, which clearly confirms successful implementation of distributed measurements based on the BDG-OTDR.
Fig. 5 (a) Distribution map of Δν obtained by the BDG-OTDR. (b) Zoomed view near the 1 m test section (dashed box in (a)) with different strain applied (0 ~200 με). (c) Shifts of Δν from the first measurement (0 με). (d) Zoomed view of the dashed box in (c).
Figure 6 depicts the measured BDG spectra of the test section (top) and the shift of Δν (bottom) according to temperature (a) and strain (b). A Gaussian-like BDG spectrum is commonly seen with a FWHM of about 140 MHz. One can see the shifts of the BDG spectra in opposite directions for temperature and strain, and the sensitivities of which are measured to be −57.48 MHz/°C and 1.37 MHz/με, respectively, as shown by a linear fit (bottom) for each. It is also noticeable that the strain sensitivity is about 40% higher than the values of former works [5] while the temperature sensitivity is similar [2, 5], which seems to originate from different cladding diameter (80 μm and 125 μm) or difference type (bow-tie and PANDA) of the FUT. From the sensitivities, the accuracies of temperature and strain measurements are calculated to be ± 0.086 °C and ± 3.65 με, respectively.
Fig. 6 The BDG spectra (top) and the shift of Δν (bottom) of the 1 m test section measured by BDG-OTDR according to (a) temperature, and (b) strain. Note that the red line is the result of linear fit for each.
Since the proposed scheme takes advantage of ASBS accumulated by a long pump pulse for the generation of the BDG, a certain length of ‘blind zone’ necessarily appear at the end of a FUT (17 m in this case) or the junction of fibers with considerably different νB‘s. Another remarkable point is that the returned signal amplitude of BDG-OTDR could be considerably decreased if the fluctuation of νB within the width of the pump pulse is larger than Brillouin gain bandwidth, which is also due to reduction of ASBS for the generation of BDG. I think this feature could limit the applications of this scheme [14]. Finally, the demonstrated measurement range of 935 m is the best result ever reported in the BDG measurements, almost 9 times better than the former work [2], showing the polarization crosstalk of the PMF is not so serious up to this range in the BDG operation.
Acknowledgments
The author is grateful to FiberPro for offering PMF samples. This work was supported by the National Research foundation of Korea (NRF) grant funded by the Korean Ministry of Education, Science, and Technology (MEST) (2012-009103).
References and links
1. K. Y. Song, W. Zou, Z. He, and K. Hotate, “All-optical dynamic grating generation based on Brillouin scattering in polarization-maintaining fiber,” Opt. Lett. 33(9), 926–928 (2008). [CrossRef] [PubMed]
2. K. Y. Song, W. Zou, Z. He, and K. Hotate, “Optical time-domain measurement of Brillouin dynamic grating spectrum in a polarization-maintaining fiber,” Opt. Lett. 34(9), 1381–1383 (2009). [CrossRef] [PubMed]
3. K. Y. Song, K. Lee, and S. B. Lee, “Tunable optical delays based on Brillouin dynamic grating in optical fibers,” Opt. Express 17(12), 10344–10349 (2009). [CrossRef] [PubMed]
4. Y. Dong, L. Chen, and X. Bao, “Truly distributed birefringence measurement of polarization-maintaining fibers based on transient Brillouin grating,” Opt. Lett. 35(2), 193–195 (2010). [CrossRef] [PubMed]
5. W. Zou, Z. He, and K. Hotate, “Complete discrimination of strain and temperature using Brillouin frequency shift and birefringence in a polarization-maintaining fiber,” Opt. Express 17(3), 1248–1255 (2009). [CrossRef] [PubMed]
6. K. Y. Song and H. J. Yoon, “High-resolution Brillouin optical time domain analysis based on Brillouin dynamic grating,” Opt. Lett. 35(1), 52–54 (2010). [CrossRef] [PubMed]
7. K. Y. Song, S. Chin, N. Primerov, and L. Thévenaz, “Time-domain distributed sensor with 1 cm spatial resolution based on Brillouin dynamic grating,” J. Lightwave Technol. 28(14), 2062–2067 (2010). [CrossRef]
8. D. P. Zhou, Y. Dong, L. Chen, and X. Bao, “Four-wave mixing analysis of Brillouin dynamic grating in a polarization-maintaining fiber: theory and experiment,” Opt. Express 19(21), 20785–20798 (2011). [CrossRef] [PubMed]
9. K. Y. Song, “Operation of Brillouin dynamic grating in single-mode optical fibers,” Opt. Lett. 36(23), 4686–4688 (2011). [CrossRef] [PubMed]
10. J. Sancho, N. Primerov, S. Chin, Y. Antman, A. Zadok, S. Sales, and L. Thévenaz, “Tunable and reconfigurable multi-tap microwave photonic filter based on dynamic Brillouin gratings in fibers,” Opt. Express 20(6), 6157–6162 (2012). [CrossRef] [PubMed]
11. Y. Antman, N. Primerov, J. Sancho, L. Thevenaz, and A. Zadok, “Localized and stationary dynamic gratings via stimulated Brillouin scattering with phase modulated pumps,” Opt. Express 20(7), 7807–7821 (2012). [CrossRef] [PubMed]
12. K. Y. Song, “Effects of induced birefringence on Brillouin dynamic gratings in single-mode optical fibers,” Opt. Lett. 37(12), 2229–2231 (2012). [CrossRef] [PubMed]
13. G. P. Agrawal, Nonlinear Fiber Optics, 2nd Ed. (Academic Press, 1995).
14. R. Bernini, A. Minardo, and L. Zeni, “Dynamic strain measurement in optical fibers by stimulated Brillouin scattering,” Opt. Lett. 34(17), 2613–2615 (2009). [CrossRef] [PubMed]
