-
基于3×3耦合器的马赫-曾徳尔干涉仪传感结构如图 1所示。来自光源的激光经过一个2×2耦合器分成两束光分别进入干涉仪的信号臂和参考臂[14-16]。在实验中,利用信号臂上的压电陶瓷(piezoelectric transducers,PZT)产生振动信号模拟外界的扰动,产生相位调制。PZT工作原理是受外部信号发生器控制,由于压电效应使内部光纤的拉伸量发生改变,从而产生相位调制,具有高稳定性和高速调制等特性。外界扰动的物理量引起信号臂内光的相位变化,与参考臂中的光信号产生相位差。相干光在3×3耦合器内分成3路被三通道型光电探测器接收,最后通过数据采集卡将采集到的信号输入到电脑的解调算法中获得所需的扰动信号。
由于3×3耦合器受到产品自身工艺加工限制和易受外界的温度、湿度、压力、偏振态等因素影响,输出不能做到功率完全相等和相位差严格满足120°。在3×3耦合器非对称状态下,输出的3路信号y1, y2, y3(本文中均指代信号的电压)可以表示为:
$ {{y_1} = {D_1} + {E_1}\cos [\phi (t) + (2/3){\rm{ \mathsf{ π} }} + {\varphi _1}]} $
(1) $ {{y_2} = {D_2} + {E_2}\cos [\phi (t) + {\varphi _2}]} $
(2) $ {{y_3} = {D_3} + {E_3}\cos [\phi (t) - (2/3){\rm{ \mathsf{ π} }} - {\varphi _3}]} $
(3) 式中,D1, D2, D3为3路输出的直流分量;E1, E2, E3为3路输出的交流分量;φ1, φ2, φ3为3路输出的相位偏差,具体大小由耦合器和光电探测器的特性决定[17-18];ϕ(t)为外界的扰动信息, t是时间。根据非对称状态的输出形式,需要对原有的3×3耦合器算法进行改进,使其能够对非对称输出进行解调。新型解调算法如图 2所示。
将非对称状态下输出的其中两路信号y1和y2作相加除以2的均值变换,可以得到新的第1路信号p1:
$ {p_1} = {D_1} + {D_2}/2 + {E_4}\cos [\phi (t) + (2/3){\rm{ \mathsf{ π} }} + {\varphi _4}] $
(4) 式中,E4为新的第1路信号的交流分量,φ4为新的第1路信号的相位偏差。在3×3耦合器解调中,直流分量和相位偏差是影响解调质量的最关键参量。同样,新的第3路信号p2:
$ {p_2} = {D_3} + {D_2}/2 + {E_5}\cos [\phi (t) - (2/3){\rm{ \mathsf{ π} }} - {\varphi _5}] $
(5) 式中,E5为新的第3路信号的交流分量,φ5为新的第3路信号的相位偏差,新的第2路信号还是原来的第2路信号y2。从新的3路信号可以看出,直流分量的差值经过平均后压缩的非常小,与变换之前的3路信号相比,新的3路信号的直流分量在数值上更加接近;而相位偏差φ1和φ2本身也是数值较小,再经过一次压缩后,得到的φ4和φ5在数值上就会变得更加小,几乎可以忽略不计[19-20]。所以,光电探测器输出的3路非对称的原始信号经过上述算法后,得到的新的3路信号被矫正至几乎对称的理想状态,再利用传统的对称型解调算法对新的3路信号进行运算,在解调过程中就不会再受到功率不相等和相位偏差的影响,可以更好地得到最终的扰动信号。
-
从实验结果中可以看出,只有3路信号在在对称的条件下才能取得最好的解调效果。在非对称情况下,3路输入信号之和y为:
$ \begin{array}{*{20}{l}} {{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} y = {y_1} + {y_2} + {y_3} = }\\ {\left[ {{D_1} + {D_2} + {D_3} + W\cos \phi (t)} \right]} \end{array} $
(6) 式中,W为增益系数,大小取决于相位偏差。只有在3路输出之和为常数且直流分量相等才能满足对称性输出的条件,使传统的解调算法有效,也就是上式中W=0。如果W=0,则必须满足下列条件:
$ {\frac{{{E_2}}}{{{E_1}}} = \frac{{\sqrt 3 \cos {\varphi _2} - \sin {\varphi _2}}}{{\sqrt 3 \cos \left( {{\varphi _1} + {\varphi _2}} \right) + \sin \left( {{\varphi _1} + {\varphi _2}} \right)}} = 1} $
(7) $ {\frac{{{E_3}}}{{{E_1}}} = \frac{{\sqrt 3 \cos {\varphi _3} - \sin {\varphi _3}}}{{\sqrt 3 \cos \left( {{\varphi _1} + {\varphi _3}} \right) + \sin \left( {{\varphi _1} + {\varphi _3}} \right)}} = 1} $
(8) $ \frac{{{E_3}}}{{{E_2}}} = \frac{{\sqrt 3 \cos {\varphi _3} - \sin {\varphi _3}}}{{\sqrt 3 \cos \left( {{\varphi _2} + {\varphi _3}} \right) + \sin \left( {{\varphi _2} + {\varphi _3}} \right)}} = 1 $
(9) 从(7)式~(9)式的条件中可以看出,只有当相位偏差都尽可能等于0°时,才能满足对称性条件。假设在非对称条件下,选取下列参量:φ1=φ2=5°,φ3=6°,即3路输出信号有5°左右的相位偏差,可以计算得到E2/E1=0.870, E3/E1=0.855, E3/E2=0.855。当存在相位偏差时,该结果与对称的条件相差过大,同时也不满足传统算法的解调条件,从而无法进行准确解调。所以,在实际解调中,不能忽视该相位偏差,必须尽可能消除相位偏差。而提出的新型算法可以将偏差进行压缩处理,使3路信号接近对称性条件进行输出,从而满足传统算法解调的要求。另外,经过处理的新的3路信号具有直流分量近似相等的特点,也满足对称性的另一个条件。在未来非对称型3×3耦合器的解调研究中,将尝试利用硬件反馈电路和算法结合的方法能更加快速稳定地实现非对称状态的矫正,从而更好地投入实际工程运用。
-
在解调系统中,噪声主要来自光路和模拟电路部分。由于光路具有抗电磁干扰能力,一般不会受高频噪声干扰,主要由温度漂移等低频干扰,而且包含在相位偏差中,直流漂移噪声一般包含在直流分量中,这些噪声如果不能去除,会使解调结果严重失真。从实验的频域结果中可以发现,提出的解调方案相比传统的解调方法具有更高的信噪比。解调系统中,信噪比可以由S表示:
$ S = 10{\rm{lg}}\left( {\frac{I}{{{n_1} + {n_2}}}} \right) $
(10) 式中,I为光功率大小,n1为光噪声大小,n2为电噪声大小。3×3耦合器解调系统中的随机噪声包括随机的直流漂移噪声和相位漂移噪声。对于传统解调方案,假设选取光功率为10mW,总的噪声水平为0.1mW,信噪比约等于20dB;而对于提出的新型解调方案,由于利用均值算法压缩了直流分量和相位偏差,也意味着将包含的随机噪声进行了压缩,减小了噪声水平,有利于提升系统信噪比和稳定性,选取新解调方案中的噪声水平约为10-4mW,信噪比约为50dB。所以,提出的新型的解调方案对于抑制随机噪声和提升信噪比有更好的效果。在今后的研究中,将利用多次均值算法,尝试将噪声进一步压缩,从而得到更高的信噪比和解调准确度,为光纤传感模式识别领域的发展提供参考。
基于非对称3×3耦合器的光纤相位解调研究
Fiber optical sensor demodulation research based on asymmetric 3×3 coupler
-
摘要: 为了解决3×3耦合器相位解调中,输出的3路信号分光比不均匀和相位差不能严格满足120°的非对称问题,采用了一种新型的3×3耦合器解调方案,并进行了理论分析和实验验证。利用均值算法对输出的任意两路信号分别进行预处理,压缩原始3路输出信号之间的功率与相位的偏差,使经过矫正后新的3路信号近似为对称状态输出。根据仿真与实验的结果,分析了耦合器输出的对称性条件和新型解调方案的抗噪声能力。结果表明,该新型解调方案可以有效矫正3×3耦合器3路输出信号的非对称性,新方案的噪声水平约为10-4mW,信噪比约为50dB, 与传统的解调方案相比,可以得到准确度与信噪比更高的待测信号。这一结果对光纤相位解调领域有很好的指导作用,加速了光纤传感技术的实用化进程。Abstract: In the fiber optical sensor demodulation based on asymmetric 3×3 coupler, the fiber-optic median phase shift 3×3 adder due to the limitation of the manufacturing process and the susceptibility to external environmental interference, and then the three-way signal output has an uneven splitting ratio and an asymmetric phenomenon that the phase difference cannot meet 120°, which causes a problem that could not be accurately corrected. In order to solve these problems, a new 3×3 replacer was used, and theoretical analysis and experimental verification were performed. The mean two algorithms were used to pre-process any two signals output, and compressed the original three signals. The power and phase difference between the output signals of the two channels make the new three-channel signals after correction to be approximately symmetrical output, and then perform a symmetric algorithm operation. Simulation and experimental results show that the new scheme can effectively correct the asymmetry of the three output signals of the positive 3×3 converter, and classify it. Noise level of the new scheme is about 10-4mW and signal-to-noise ratio is about 50dB. Compared with the traditional alternative scheme, the new structure can obtain higher accuracy and signal-to-noise ratio of the signal under test. In addition, according to the simulation and experimental results, the symmetry conditions of the output of the replacer and the anti-noise capability of the new superposition scheme are analyzed. The result has a good guiding role in the field of optical fiber polarizers and accelerates the practical process of optical fiber sensing technology.
-
Key words:
- optoelectronics /
- phase demodulation /
- mean algorithm /
- 3×3 coupler
-
-
[1] LU Y, ZHU T, CHENG L, et al. Distributed vibration sensor based on coherent detection of phase-OTDR[J]. Journal of Lightwave Technology, 2010, 28(22): 3243-3249. [2] PENG F, DUAN N, RAO Y, et al. Real-time position and speed monitoring of trains using phase-sensitive OTDR[J]. IEEE Photonics Technology Letters, 2014, 26(20): 2055-2057. doi: 10.1109/LPT.2014.2346760 [3] ZHANG B, ZHANG E T, HU X Ch, et al. Amplification charactcristics of multiwavelength crbium-doped fiber laser amplifiers[J]. Laser Technology, 2018, 42(3): 325-330(in Chinese). [4] BAO X, ZHOU D P, BAKER C, et al. Recent development in the distributed fiber optic vibration and ultrasonic detection[J]. Journal of Lightwave Technology, 2016, 35(16): 3256-3267. [5] LIU Sh, HAN X Y, XIONG Y Ch. Distributed vibration detection system based on weak fiber grating array[J]. Chinese Journal of Lasers, 2017, 44(2): 0210001(in Chinese). [6] YING C. Quantitative detection of phase-demodulation techniques for phase-sensitive optical time domain reflectometry[D]. Hangzhou: Zhejiang University, 2018: 13-23(in Chinese). [7] ZHANG X, SUN Z, SHAN Y, et al. A high performance distributed optical fiber sensor based on φ-OTDR for dynamic strain measurement[J]. IEEE Photonics Journal, 2017, 9(3): 1-12. [8] XU N, DAI M. Distributed optical fiber temperature and pressure sensor design[J]. Chinese Optics, 2015, 8(4): 629-635(in Chinese). [9] LIU L. Research on light reflector based on helium pulse[D]. Shanghai: Shanghai Jiao Tong University, 2015: 17-47(in Chinese). [10] WU G X, DUAN F J. Avalanche photodiode electric heterodyne mixing technology and its parameter optimization[J]. Laser Technology, 2015, 39(6):803-804(in Chinese). [11] QIAO J P, DENG L W, HE J, et al. Optimization of fast image encryption algorithm based on chaotic mapping[J]. Laser Technology, 2017, 41(6): 897-903(in Chinese). [12] XU G, HE Ch Ch, ZHANG L N, et al. Research of position technology of Mach-Zehnder interferometer[J]. Laser Technology, 2019, 43(2): 195-200(in Chinese). [13] SU B L. Investigation on quasi-lossless transmission system based on pumping Raman amplification[J]. Laser Technology, 2017, 41(2): 265-269(in Chinese). [14] QIAN X L, KONG Y, DU T Y, et al. Study on full-sensitivity to vibration of phase sensitive optical time-domain reflectometers[J].Laser Technology, 2019, 43(5): 608-613(in Chinese). [15] CHEN K, ZHENG J Y, ZHOU J H, et al. Design of real-time fast polarization control algorithm[J]. Laser Technology, 2017, 41(5): 738-742(in Chinese). [16] KOYAMADA Y, IMAHAMA M, KUBOTA K, et al. Fiber-optic distributed strain and temperature sensing with very high measurand resolution over long range using coherent OTDR[J]. Journal of Lightwave Technology, 2009, 27(9): 1142-1146. [17] LV Y L, XING Y W. Study on rayleigh scattering waveform characteristics of phase light time domain reflectometer[J]. Acta Optica Sinica, 2011, 31(8): 0819001(in Chinese). [18] HUANG Zh H, WANG Y L, LI G F, et al. Adaptive frequency-domain equalization for few-mode fiber transmission systems[J]. Laser Technology, 2017, 41(1): 124-128(in Chinese). [19] GAO H, LIU J M, YANG Ch, et al. Compact solid-state lasers with high peak power used for remote laser rangefinders[J]. Laser Technology, 2019, 43(5): 597-600(in Chinese). [20] XU S H, XIAO Sh L. Research on reverse modulation optical communication system based on acousto-optic modulation[J]. Laser Technology, 2015, 39(5): 599-600(in Chinese).