Abstract
This paper presents a method to improve the dynamic range of white light interferometer (WLI) based polarization mode coupling (PMC) measurement system beyond 100 dB. The limitation of interference beat noise is overcame by analyzing in detail the inherent noises that have impacts on the detection sensitivity. An improved PMC measurement system and method are proposed for testing ultra-high polarization extinction ratio (PER) of polarization-related devices. The method can improve dynamic range dramatically through eliminating interference beat noise and enhancing the tested interference intensity simultaneously, which are verified theoretically and experimentally. In addition, a Y-junction with ~80 dB PER of LiNbO3 chip corresponding to a weak signal is tested as an application example. The results demonstrate that the high PER interferogram can be identified clearly and steadily with standard deviation 0.9 dB (3σ) @ ~80 dB. This proposed method is highly beneficial in fabrication and evaluation for polarization devices with ultra-high PER.
© 2016 Optical Society of America
1. Introduction
Distributed optical fiber sensing techniques have been investigated for structural imperfection evaluation, temperature detection and strain monitoring. There exist many approaches for different physical quantity detections, such as optical time domain reflectometer (OTDR) [1], optical frequency domain reflectometry (OFDR) [2], and Brillouin optical time domain analysis (B-OTDA) [3]. Different from detecting the optical characteristics of various backscattered light, distributed polarization mode coupling (PMC) measurement system based on white light interferometer (WLI) is developed by monitoring transmission light [4]. In addition to serving as sensors, the PMC system can also be applied to distributed measurement for polarization-related characteristics with high-precision [5,6]. We could obtain the distributed features of devices depending on the locations and amplitudes of interferograms, such as the angular alignment of polarization-maintaining fiber (PMF) [7], the crosstalks along interferometric fiber optic gyro (IFOG) coil [8], the polarization extinction ratio (PER) [9] and distributed birefringence dispersion [10] of integrated-optic devices.
It has been reported that the dynamic range of distributed PMC measurement system based on WLI can achieve ~90 dB [9]. Whereas, this system requires each component and device to be kept in an optimum state. The weak PMC or high PER (>80 dB) measured with PMC measurement system is vulnerable to the environment noises [11], which will be submerged in noise floor, leading to a confusing identification. Therefore, the dynamic range and detection sensitivity of PMC measurement system should be further pronouncedly improved. Several basic studies have been reported to discuss the improvement of signal to noise ratio (SNR) or dynamic range. K. Takada et al. suggested that amplifying the output power of light source [12] and reducing the detectable bandwidth could improve the detection sensitivity of optical low coherence reflectometer (OLCR) [13]. These methods increase requirements for light source and mechanical scanning stage. W. Sorin et al. declared that the effects of intensity noise could be decreased by selectively attenuating the reference power [14], which will reduce the utilization of the light source power. W. Jing et al. described the rotation angle optimization of the polarization eigenmodes in WLI to increase the SNR of the output interferogram [15]. This measurement system however has a compromise between the measurement sensitivity and the measurement error. H. Zhang et al. proposed a signal processing method based on empirical mode decomposition (EMD) to enhance the detection sensitivity by suppressing the noise from −50 dB to −60 dB [16].
This paper analyzes three typical noises—light shot noise, interference beat noise, and circuit thermal noise—of distributed PMC measurement system. We propose a PMC measurement system to improve the system dynamic range by suppressing the interference beat noise. It is verified theoretically and experimentally that the dynamic range utilizing this proposed method can be expanded beyond 100 dB. Besides, a Y-junction with chip PER of ~80 dB is tested and its corresponding interferogram can be identified clearly and steadily. This proposed method is highly beneficial in evaluation of polarization devices with ultra-high PER and improvement of dynamic range for PMC measurement system.
2. Theory analysis
2.1 Traditional PMC measurement
The schematic diagram of traditional PMC measurement based on WLI is shown in Fig. 1 [4]. The white light from a superluminescent light-emitting diode (SLD) is transmitted into a device under test (DUT). Generally, there exist couplings or polarization extinctions at the perturbation points due to the inner structural imperfectness of DUT, external perturbations or discontinuous points (such as spliced points, connections by optical adhesive). In this case, the coupling mode with tiny energy is generated at a perturbation point and the excited mode with most energy will continue to propagate along DUT. It will generate two optical paths (OPs) with orthogonal polarization eigenmodes at the perturbation point due to the birefringence of DUT. The two OPs induce optical path difference (OPD) at the output-end of PMF, where is the PMF length between the perturbation point and the output-end of PMF. Afterwards, utilizing a scanning Mach-Zehnder interferometer (MZI), the main interferogram is detected at OPD of and the coupling interferogram is generated by a compensation OPD of , where . Finally, all the profiles of the interferograms are sampled and extracted by signal processor unit (SPU).
As the typical noise originating from distributed PMC measurement system, light shot noise , interference beat noise , and circuit thermal noise can be represented as follows [14]
where is the elementary charge, is the responsivity of photodiode (PD), is the DC light intensity, is the detection bandwidth, is the polarization degree of light source (here = 1), is the bandwidth of light source, and represent the light intensity of scanning arm and reference arm, respectively, , and are the Boltzmann parameter, thermodynamic temperature and effective noise resistance, respectively.Generally, a 50/50 coupler is adopted to divide the light into the two arms of MZI equally. The light intensities of scanning arm and reference arm can be written as
where and are the intensities of excited mode and coupling mode, respectively, in the corresponding arm of MZI. There exists a relationship of , where is the coupling coefficient introduced by perturbation point which is usually much less than 1. Therefore, Eq. (2) can be expressed as . Then we can obtain the final signalwhere is the main interferogram pattern filtering the DC component, and is the intensity of coupling interferogram.For the case of differential detections, the noise variance is given by [17]
The total signal photocurrent can be written as where and are the corresponding mean-square signal photocurrents of main interferogram and coupling interferogram, respectively. Therefore, the dynamic range of traditional PMC measurement with differential detections can be expressed asThe denominator in Eq. (6) shows that the dynamic range will be sequentially determined by the third term (-related) and the first term (-related) when the reference powers are less than a few microwatts (~2 μw). Beyond this point, increasing the light power does not improve the dynamic range due to the limitation of interference beat noise (the second term—).
2.2 PMC measurement with noise limitation
It has been recognized that polarization beam splitter (PBS) can separate or combine the orthogonal polarizations, which plays an important role in polarization-multiplexed transmissions [18]. Based on schematic shown in Fig. 1, the first coupler denoted by C1 in MZI is replaced by a PBS (see Fig. 2). Here, the two orthogonal polarization eigenmodes —excited mode and coupling mode introduced by DUT—are separated into the scanning arm and reference arm, respectively. Based on Eq. (2), the light intensities of scanning arm and reference arm with PBS are rewritten as
Here, the PBS with high extinction ratio () could provide an ideal separation for the two orthogonally polarized lights.The main interferogram will lose its original meaning due to the PER of PBS. However, the coupling interferogram still exists and can be expressed as .
Therefore, the final photocurrent of AC signal with differential detections can be expressed as
Besides, the noise variance can be rewritten asThe dynamic range of PBS-calibrated PMC measurement cannot be directly calculated according to Eq. (6) due to the meaningless main interferogram. In this case, should be calibrated by a remarkable interferogram (i.e. ), which is also measured in traditional PMC measurement. The relative dynamic range of PMC measurement with PBS can be rewritten asIn comparison to Eq. (6), Eq. (10) indicates that the signal photocurrent is amplified four times, and the detection sensitivity is enhanced because the term of interference beat noise is ignored due to such a small coupling coefficient . Accordingly, the system dynamic range of PBS-calibrated system will be improved significantly through decreasing detection sensitivity noise and increasing tested interferogram intensity simultaneously.
3. Experiments and results
3.1 PBS-calibrated PMC measurement system and method
In this section, a measurement system for verifying the dynamic range improvement is proposed. As shown in Fig. 2, the white light from a superluminescent light-emitting diode (SLD) at 1550 nm with a short coherence length (~53 μm, corresponding to full width at the half maximum of 45 nm) is divided into two beams through a 98/2 fiber coupler. 2% of the light is for monitoring the output power of light source, and the remaining light is launched into a wavelength division multiplex (WDM). Moreover, a distributed feedback (DFB) laser of 1310 nm is utilized and its light is launched into the same WDM. Then, the light transmitted through an isolator is linearly polarized by a polarizer. For simplify, the DUT is a section of PMF whose output-end is spliced with the pigtail of PBS (with operation wavelength of 1550 ± 40 nm and extinction ratio of ) utilizing an alignment device—optical fiber fusion splicer (i.e. Fujikura FSM-45PM). The output light from PBS is injected into a MZI that can compensate the OPD. Afterwards, the coupling interferograms are divided by two WDMs (with operation wavelength of 1550 nm/1310 nm, insertion loss of 0.6 dB, and isolation of ) [19] based on the wavelengths and finally detected with differential PDs.
To ensure the measurement accuracy, the system has been improved in many aspects based on our previous works: Firstly, a distributed feedback (DFB) laser of 1310 nm is adopted to eliminate the mechanical vibration influence of scanning motor [10] that is a conventional way for position calibration of mechanical scanning [19]. Secondly, a differential detection is completed by adopting two PDs [9]. Thirdly, a differential scanning MZI with two lenses is employed to suppress the optical power fluctuation [20].
Generally, the alignment device between the pigtail of PBS and DUT should work in 0°–0°, in which the excited mode and coupling mode introduced from DUT can be divided completely by PBS. However, when the alignment device keeps in 0°–45°, the light of excited mode and coupling mode are injected equally into the fast-axis and slow-axis of PBS’s pigtail, respectively. In this case, the PBS is similar to a 50/50 coupler. The calibrating method utilizing this proposed system as shown in Fig. 2 can be described as follows: (1) Varying optical fiber fusion splicer 0°–45° and 0°–0°, and acquiring the two temporal results, respectively. (2) Calibrating a remarkable interferogram in 0°–0° measurement to that in 0°–45°. (3) The relative dynamic range is obtained and the other interferograms corresponding weak-coupling points or ultra-high PER can be identified utilizing the results of 0°–0° measurement.
3.2 Dynamic range improvement results
A PMF with a length of 16 m is tested utilizing the PBS-calibrated PMC measurement system. The envelopes of interferograms versus scanning OPD with PBS angle 0°–45° and 0°–0° are plotted in Fig. 3(a) and 3(b), respectively. Based on scanning OPD relationship of , interferograms and (or in Fig. 3(c)) represent the same coupling point—the PER of system polarizer. It is well known that the real coupling intensities and dynamic range can be acquired by normalizing the main interferogram of Fig. 3(a) [9]. Besides, the dynamic range can be calculated by the difference between and the noise floor. In comparison with Fig. 3(a), the main interferogram in Fig. 3(b) will be meaningless due to the function of PBS. However, Fig. 3(b) indicates that the coupling value introduced by system polarizer is amplified from −18.6 dB to −12.0 dB, whereas the noise floor is decreased from −77.0 dB to −84.0 dB. When we calibrate the value of interferogram by interferogram , the dynamic range of PMC measurement system can be enhanced to 103.3 dB. An extra measurement with angle combination 0°–90° is shown Fig. 3(c), in which the corresponding interferograms are inverse of Fig. 3(b).
Similar to the above calibrated method, the measured results of dynamic range versus detector light intensity are shown in Fig. 4, where the dynamic range is expressed as .
The theoretical calculations indicate that dominant noise sources depend on the value of the reference arm power. When the reference light power is smaller, the dynamic range is determined by the circuit thermal noise . With the increasing of detector power, the dynamic range cannot overcome the limitation of shot noise . At last, the dynamic range is limited by the interference beat noise utilizing the traditional method. However, in the smaller segment (0.1 μw– ~2 μw) of light intensity, the dynamic range by PBS-calibrated method is amplified ~6 dB because the amplitude of AC signal is doubled and the detection sensitivity will be same to traditional method due to the limitation of circuit thermal noise . Afterwards, in the segment of ~2 μw–33 μw, the dynamic range will be enhanced dramatically because the detection sensitivity and AC signal will be improved simultaneously. As a result, the proposed method utilizing the proposed PBS-calibrated method can overcome the limitation of interference beat noise. Additionally, it can be estimated that the relative dynamic range will be continually improved with the increasing of detector light intensity.
4. Application
4.1 Evaluating Y-junction with ultra-high chip PER
LiNbO3 integrated Y-junctions are the basic components in IFOG which contains functions of beam splitter, optical polarizer and electro-optical modulator [21]. A packaged Y-junction is usually composed of LiNbO3 chip, electrodes, one input pigtail and two output pigtails. Utilizing the system as shown in Fig. 2, the connection configuration for testing packaged Y-junction in DUT is illustrated in Fig. 5. Different from testing PMF, the linearly polarization light is equally launched into the orthogonal axes of the input pigtail of Y-junction with a 45°-rotated polarizer. The light from the output pigtail of Y-junction is spliced with the PBS’s pigtail utilizing fiber fusion splicer. The methods for full evaluation of polarization characteristics of Y-junction have been reported in Ref [9]. in detail that are beyond this paper.
The results for testing Y-junction with different angle combinations are illustrated in Fig. 6(a) and 6(b), respectively. The PER of LiNbO3 chip can be distinguished by the scanning OPD
where (section = IA, AB, BC and chip) represents the length of corresponding PMF and LiNbO3 chip, and represent the birefringences of PMF and LiNbO3 chip, respectively. Based on the relationship of scanning OPD , interferograms A, B, (or ) and (or ) are induced by the coupling points A, B, C and O as shown in Fig. 5, respectively. Besides, the other interferograms , and in the inset of Fig. 5(a) are the 2nd-order couplings without the realistic perturbation points of the Y-junction [22]. Here, the analysis in detail of coupling points of Y-junction are beyond this paper and thus we only focus on the chip PER of Y-junction.The characteristics of input pigtail of Y-junction will be suppressed due to the PER of PBS. Similar to the calibration method of PMF measurement in Section 3, the results of PBS-calibrated method is corrected by the interferograms (and ) as a remarkable interferogram, which represents the coupling point C between the output-pigtail and LiNbO3 chip of Y-junction. As shown in Fig. 6, the traditional method with a PBS spliced at 45° can achieve a similar effect with Ref [9], in which the noise floor utilizing traditional method is ~90 dB (blue line). However, the noise floor by PBS-calibrated method can be improved to ~105 dB (red line). The results demonstrate that the values of PER with PBS-calibrated method can be distinguished clearly from the lower noise floor and higher SNR.
4.2 Evaluating Y-junction with ultra-high chip PER
To verify the repeatability of two measurement methods, the Y-junction is repeatedly measured 20 times by realigning the axes of PBS and the output pigtails PM fiber at 0°–45°, 0°–0° and 0°–90°, respectively. According to the relationship between PER and PMC of , the results utilizing different angle combinations are shown in Fig. 7—the black line represents the PERs of Y-junction’s chip tested with angle 0°–45° of PBS, and the blue and red lines give the PER distributions by PBS-calibrated method. It shows that the test fluctuation of high PER utilizing the traditional method will be affected more easily.
The standard deviation of PERs utilizing traditional method (0°–45°) and PBS-calibrated method (0°–0°) are 6.5 dB and 0.8 dB , respectively. Therefore, the average chip PER of Y-junction with 0°–0° and 0°–90° PBS-calibrated methods are calculated as follows
As a result, the PBS-calibrated method can also obtain more stable interferograms. Additionally, the imperfect optical devices could introduce some slight extra errors. When we employ PBS with high extinction ratio and WDMs with high isolation, the dynamic range of PMC measurement system will be extended further. However, the high dynamic range is at the expense of losing the characteristics of input pigtail of Y-junction. In spite of missing some information of Y-junction’s pigtail, pursuing the ultra-high PER measurement is more significant.
5. Conclusions
We describe the noise originating from PMC measurement system and prove that the system dynamic range can be improved by suppressing the interference beat noise. The experimental results obtained from this proposed PBS-calibrated method indicate that the dynamic range can break through 100 dB. Additionally, a Y-junction with ~80 dB PER is tested as an application example and the interferogram representing the chip PER can be identified clearly. Measurement repeatability of PER is limited to 0.9 dB @ ~80 dB. This proposed PMC measurement method is highly beneficial in evaluation of polarization devices with ultra-high PER or ultra-weak PMC.
Funding
National Natural Science Foundation of China (61422505, 61227013, 61307104); Program for New Century Excellent Talents in University (NCET-12-0623); National Key Scientific Instrument and Equipment Development Project (2013YQ040815); Specialized Research Fund for the Doctoral Program of Higher Education (20122304110022); Heilongjiang Provincial Natural Science Foundation (ZD201205); International Science & Technology Cooperation Program of China (2013DFR10340).
Acknowledgments
The Y-junction used in this work was supplied by Yong Hua and Ping Shu of Chongqing Optoelectronics Research Institute (Chongqing 400060, China). We would like to acknowledge them for the support.
References and links
1. Z. Zhang and X. Bao, “Distributed optical fiber vibration sensor based on spectrum analysis of Polarization-OTDR system,” Opt. Express 16(14), 10240–10247 (2008). [CrossRef] [PubMed]
2. S. Wang, X. Fan, Q. Liu, and Z. He, “Distributed fiber-optic vibration sensing based on phase extraction from time-gated digital OFDR,” Opt. Express 23(26), 33301–33309 (2015). [CrossRef] [PubMed]
3. Y. Dong, P. Xu, H. Zhang, Z. Lu, L. Chen, and X. Bao, “Characterization of evolution of mode coupling in a graded-index polymer optical fiber by using Brillouin optical time-domain analysis,” Opt. Express 22(22), 26510–26516 (2014). [CrossRef] [PubMed]
4. K. Takada, J. Noda, and K. Okamoto, “Measurement of spatial distribution of mode coupling in birefringent polarization-maintaining fiber with new detection scheme,” Opt. Lett. 11(10), 680–682 (1986). [CrossRef] [PubMed]
5. Y. J. Rao, Y. N. Ning, and D. A. Jackson, “Synthesized source for white-light sensing systems,” Opt. Lett. 18(6), 462–464 (1993). [CrossRef] [PubMed]
6. L. Yuan, L. Zhou, and W. Jin, “Quasi-distributed strain sensing with white-light interferometry: a novel approach,” Opt. Lett. 25(15), 1074–1076 (2000). [CrossRef] [PubMed]
7. K. Takada, K. Chida, and J. Noda, “Precise method for angular alignment of birefringent fibers based on an interferometric technique with a broadband source,” Appl. Opt. 26(15), 2979–2987 (1987). [CrossRef] [PubMed]
8. Z. Li, X. Yao, X. Chen, H. Chen, Z. Meng, and T. Liu, “Complete characterization of polarization-maintaining fibers using distributed polarization analysis,” J. Lightwave Technol. 33(2), 372–380 (2015). [CrossRef]
9. J. Yang, Y. Yuan, A. Zhou, J. Cai, C. Li, D. Yan, S. Huang, F. Peng, B. Wu, Y. Zhang, Z. Liu, and L. Yuan, “Full evaluation of polarization characteristics of multifunctional integrated optic chip with high accuracy,” J. Lightwave Technol. 32(22), 4243–4252 (2014). [CrossRef]
10. Z. Yu, J. Yang, Y. Yuan, C. Li, S. Liang, L. Hou, F. Peng, B. Wu, J. Zhang, Z. Liu, and L. Yuan, “Quasi-distributed birefringence dispersion measurement for polarization maintain device with high accuracy based on white light interferometry,” Opt. Express 24(2), 1587–1597 (2016). [CrossRef] [PubMed]
11. Z. Guo, H. Zhang, X. Chen, D. Jia, and T. Liu, “Influence of vibration disturbance during polarization coupling measurement of polarization-maintaining fiber,” Appl. Opt. 50(20), 3553–3558 (2011). [CrossRef] [PubMed]
12. K. Takada, T. Kitagawa, M. Shimizu, and M. Horiguchi, “High-sensitivity low coherence reflectometer using erbium-doped superfluorescent fibre source and erbium-doped power amplifier,” Electron. Lett. 29(4), 365–367 (1993). [CrossRef]
13. K. Takada, A. Himeno, and K. Yukimatsu, “Phase-noise and shot-noise limited operations of low coherence optical time domain reflectometry,” Appl. Phys. Lett. 59(20), 2483–2485 (1991). [CrossRef]
14. W. V. Sorin and D. M. Baney, “A simple intensity noise reduction technique for optical low-coherence reflectometry,” IEEE Photonics Technol. Lett. 4(12), 1404–1406 (1992). [CrossRef]
15. W. Jing, Y. Zhang, G. Zhou, H. Zhang, Z. Li, and X. Man, “Rotation angle optimization of the polarization eigenmodes for detection of weak mode coupling in birefringent waveguides,” Opt. Express 10(18), 972–977 (2002). [CrossRef] [PubMed]
16. H. Zhang, W. Ye, D. Jia, and Y. Zhang, “Sensitivity enhancement of distributed polarization coupling detection in Hi-Bi fibers,” Chin. Opt. Lett. 10(4), 040603 (2012). [CrossRef]
17. A. M. Rollins and J. A. Izatt, “Optimal interferometer designs for optical coherence tomography,” Opt. Lett. 24(21), 1484–1486 (1999). [CrossRef] [PubMed]
18. I. Yokohama, K. Okamoto, and J. Noda, “Analysis of fiber-optic polarizing beam splitters consisting of fused-taper couplers,” J. Lightwave Technol. 4(9), 1352–1359 (1986). [CrossRef]
19. S. H. Hsu, C. Y. Tsou, M. S. Hsieh, and C. Y. Lin, “Low-coherence interferometric fiber sensor with improved resolution using stepper motor assisted optical ruler,” Opt. Fiber Technol. 19(3), 223–226 (2013). [CrossRef]
20. C. Li, J. Yang, Y. Yuan, A. Zhou, D. Yan, J. Chai, S. Liang, L. Hou, B. Wu, F. Peng, Y. Zhang, Z. Liu, and L. Yuan, “A differential delay line for optical coherence domain polarimetry,” Meas. Sci. Technol. 26(4), 045102 (2015). [CrossRef]
21. D. Song, Z. Wang, X. Chen, H. Zhang, and T. Liu, “Influence of ghost coupling points on distributed polarization crosstalk measurements in high birefringence fiber and its solution,” Appl. Opt. 54(8), 1918–1925 (2015). [CrossRef] [PubMed]
22. W. M. J. Green, M. J. Rooks, L. Sekaric, and Y. A. Vlasov, “Ultra-compact, low RF power, 10 Gb/s silicon Mach-Zehnder modulator,” Opt. Express 15(25), 17106–17113 (2007). [CrossRef] [PubMed]