Open Access
Add to BibSonomy
Abstract
We experimentally demonstrate a novel liquid crystal on silicon (LCoS)-based programmable spectral processor, including two cascaded photonic configurations, to realize the state of polarization (SOP) manipulation in the spatial and spectral domain. As the final SOP at each wavelength is linear polarization with a manageable polarization direction, a broadband linear polarizer is used to filter the undesired wavelengths. The polarization manipulation only needs to be implemented along the dispersion direction with a 1D-LCoS. The programmable spectral processor can experimentally reach an intensity modulation depth of 46.4 dB with less than 1 dB polarization-dependent loss (PDL). Moreover, arbitrary power spectral distribution can be obtained with around 40 dB channel isolation. In particular, our experimental results verify that the proposed setup can achieve the adjustable central wavelength and tunable filtering at a resolution of 0.08 nm.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Programmable spectral processor (PSP) is a very important device for various photonic applications, including wavelength selective switching (WSS) [1–3], adaptive spectral filter [4], optical pulse shaping [5] and secure optical communications [6–9]. In particular, PSP with tunable grid granularity is believed as a key solution for elastic optical networking (EON) [10].
Generally, a ‘disperse-and-select’ configuration is commonly used in liquid crystal on silicon (LCoS)-based photonic spectral processor [11]. Both spectrally dispersed element and spectral processor are two key components for the PSP implementation. Spectrally dispersed element usually use the bulk transmission grating, waveguide grating, or Fabry-Perot interferometer array to realize the spectral separation with fine resolution [12–14]. As for the spectral processor, spatial light modulator (SLM) is often employed to manipulate the phase of the signal for steering the angular dispersed optical beams to different directions [15]. In general, a specific phase grating is loaded into a two dimensional (2D)-LCoS device along the switching axis which is orthogonal to the dispersion axis in order to diffract selected spectral components to target output ports with the desired attenuation [16]. However, owing to the polarization-dependent characteristic of LCoS, both insertion loss (IL) and polarization-dependent loss (PDL) of PSP based on LCoS suffer great impact, which will constrain its application. Therefore, a polarization diversity configuration with two cascaded birefringent crystals has been proposed to circumvent this restriction [17]. However, it is challenging to achieve the equal optical path length for two orthogonal linear polarizations. In addition, due to imperfection of phase grating on the LCoS caused by fringing field effect, pixel discretization, phase quantization, and high order diffraction may bring additional crosstalk and the extinction ratio (ER) penalty at the output ports [18,19]. Recently, an LCoS-based PSP with fine grid granularity is reported with a power attenuation from 0 dB to 30 dB by a step of 0.1 dB and its measured channel crosstalk is less than −20 dB [20]. Besides, it was also applied for the signal monitoring over 100 spatial channels, in which its inter-spatial-channel crosstalk was typically less than −30 dB [21]. In order to improve the ER, two Mach-Zehnder-interferometer-based variable optical attenuators are integrated on the PSP to improve the ER to about 50 dB [22]. However, it is difficult to achieve such high ER over all C-band simultaneously.
In this paper, we propose a novel state of polarization (SOP) insensitive LCoS based programmable spectral processor with two cascaded photonic configurations, which can manipulate SOP of the input signals in the optical spectral domain to achieve the intensity modulation instead of steering the dispersed beams to different directions. The 1D-LCoS is divided into two sections to form two cascaded photonic configurations. One section of the LCoS is used to transform the spatially separated optical signals into linear polarization. Then, the other section of the LCoS is encoded to adjust the azimuthal angle of the linearly polarization direction. The single channel characterization result indicates that our configuration is insensitive to the SOP of input signal. The experimental result of multichannel spectral processing shows a significant increase of the channel isolation. The experimental results of variable filtering operation show that the proposed method is effective for the tunable bandwidth spectral processing with grid granularity of 0.08 nm.
2. Operation principle
The system configuration of LCoS-based PSP including four function-modules is shown in Fig. 1(a) and the principle of PSP implementation is presented in Fig. 1(b). In general, the SOPs of multichannel input signals are random. For the ease of explanation, we assume five-channel optical inputs with different SOPs. First, the five-channel input signals are launched into the beam shaping module which consists of a collimator and a beam expander. The function of beam shaping module is to enlarge the beam size for the purpose of slightly overfilling the dispersion element so as to obtain high spectral resolution. Then, the expanded beam goes into the spatial dispersion module, which consists of a dispersion element and a Fourier lens. The beam passes through the dispersion element and spread the spectrum angularly. With the use of a Fourier lens, the angular dispersion is transformed to the linear dispersion. The wavelength resolution of the proposed PSP is determined by the spectral resolution of the dispersion element. Subsequently, the dispersed signal enters the polarization manipulation module, which is composed of a 1D-LCoS, a reflective 4-f system, and a broadband quarter-wave plate. Accordingly, signals at different wavelengths are projected onto the left half section of LCoS. Under such operation, the LCoS provides the spectrally degree of freedom for achieving the polarization transformation. Generally, the LCoS can be regarded as a variable retardance waveplate with a fixed fast axis. The above polarization transformation can be written by the Jones matrix
where φ1(λ) is the phase retardation introduced by the left half section of LCoS, E (λ) and δ(λ) are the amplitude and the phase of the incident light, respectively. The subscript of p and s denote the p-polarized component and s-polarized component, respectively. If the phase pattern loaded into the LCoS can compensate the initial phase difference between the p-polarized and the s-polarized light, namely, φ1(λ) = δp(λ)-δs(λ), the field of output light will become linearly polarized. Due to the distinction of amplitude ratio of p and s for five channels, the azimuthal angles of the obtained linearly polarized lights are not uniform. In order to further manipulate the signal intensity in the spectral domain, a 4-f reflective configuration is used to project above linearly polarized signals into the right half section of LCoS and spatial polarization direction management is employed. For the purpose of rotating the azimuthal angles, the LCoS together with a quarter-wave plate (QWP) are used to form a polarization rotator. The included angel between the slow axes of QWP and the extraordinary axis of LCoS is set to 45°. Such configuration has been reported for the vector beam generation [23]. The Jones matrix description for such polarization rotator iswhere φ2(λ) is the phase retardation induced by the right half section of LCoS, R(·) represents the matrix of polarization rotation. According to Eq. (2), the phase modulation is transformed to the rotation angle of linearly polarized lights. Due to the spatial separation for the signals with different wavelength, their polarization direction can be independently modulated. Multiplying Eq. (1) with Eq. (2), and under the condition of δp(λ)- φ1(λ) = δs(λ), the Jones vector of output light from the polarization manipulation module iswhere α(λ) is the initial azimuthal angles of linearly polarized lights which is determined by Ep(λ) and Es(λ). According to Eq. (3), arbitrary angle rotation can be encoded for every channel. Finally, the beam operated by the polarization manipulation module enter the output module, which consists of a near infrared linear polarizer and a Fourier lens. The near infrared linear polarizer is used to transform the polarization modulation to the intensity modulation. The transmitted light intensity can be obtainedwhere I0(λ) is intensity of input light, β is the transmission direction of linear polarizer. By adjusting φ2(λ), the input signal can be selectively filtered or shaped in the spectral domain. In general, the polarization extinction ratio of near infrared linear polarizer is more than 50 dB over 850-1600 nm. Therefore, a high extinction ratio can be expected as long as the precision of the phase retardation introduced by LCoS is high enough.
Fig. 1 System configuration (a) and schematic drawing of operation principle (b) of the programmable spectral processor based on spatial polarization manipulation with LCoS.
3. Experimental results and discussions
The experimental setup of proposed programmable spectral processor is illustrated in Fig. 2(a). The corresponding four modules are cascaded with a denotation of the dotted box. Several tunable distributed feedback (DFB) lasers in the C-band with a channel spacing of 50 GHz are used as the light source to realize multichannel input. For the continuous spectrum measurement, DFB lasers are replaced by an amplified spontaneous emission (ASE) source together with an optical filter to generate light source with a bandwidth of 390 GHz. In order to measure the PDL and characterize the polarization sensitivity of the proposed setup, a polarization emulator including a linear polarizer (LP1) and a quarter-wave plate (QWP) is used to vary the SOP of the input light. Next, the beam is free-space coupled by a collimator (Thorlabs, F240FC-1550). Then, the beam size of the collimated beam is enlarged by a beam expander in order to slightly overfill the bulk transmission grating (Ibsen Photonics, PING-Sample-083), whose resolution is 1000 line/mm within the active area of 19 mm × 13.5 mm. Its theoretically wavelength resolution is 0.08 nm. Being diffracted by the grating, the beam is angularly dispersed. Then, the angular wavelength separation is transformed to linear wavelength separation by the Fourier lens. Subsequently, the spectrally dispersed light is divided by a beam splitter (BS) and the reflected part focuses onto the left part of LCoS, which is placed at the Fourier plane of L1. The LCoS (Meadowlark Optics Inc, HSPDM12k-1550) is 1x12288 linear array with an array size of 19.66 mm × 19.66 mm, and its pixel pitch is 1.6 μm. The measured zero-order diffraction efficiency of this LCoS is 87-91%. Subsequently, the beam is reflected to the BS again and transmitted through a 4-f system. The reflective 4-f system is composed of a mirror and a lens with a distance equal to the focal length of L2. After passing through the reflective 4-f system, the beam goes through the BS again. The reflected beam passes through another near infrared linear polarizer (LP2) for measuring the light intensity and obtaining φ1(λ) according Eq. (1). The transmitted beam is spatially separated again on the Fourier plane of L2 where LCoS is placed. It re-focuses onto the right half of LCoS again for the polarization manipulation. The broadband QWP2 (Thorlabs, AQWP10M-1600) with fast axis oriented at 45° with respect to the extraordinary axis of LCoS is inserted before LCoS to form a spatial polarization rotator. Finally, the modulated beam pass through the BS once again and the reflected part is the modulated signal for measurement. The broadband linear polarizer LP3 (Thorlabs, LPNIR050-MP2) together with lens L3 is used to convert the polarization modulation into the intensity modulation on the basis of Malus law. The typical phase pattern for programmable channel switching is shown in Fig. 2(b). The phase pattern is plane waveform for each channel. For the left half section of LCoS, the phase pattern φ1(λ) is used to compensate the initial phase difference between the p-polarized and s-polarized light, which can be adjusted to match the variation in SOP. For the right half section of LCoS, the phase modulation φ2(λ) is used to rotate the linear polarization orientation to make maximum transmittance for all channels. According to Eq. (4), the phase difference between maximum transmittance (ON) and minimum transmittance (OFF) is π. Through adjusting φ2(λ), the input signal can be selectively filtered or shaped in the spectral domain.
Fig. 2 (a) Experimental setup of proposed programmable spectral processor. CL: collimating lens, BE: beam expander, G: grating, LP: linear polarizer, QWP: quarter wave plate, L: lens, BS: beam splitter, LCoS: liquid crystal on silicon, M: mirror, OSA: optical spectrum analyzer. (b) Typical phase pattern for programmable channel switching.
In order to evaluate ER and PDL of the proposed programmable spectral processor, we firstly perform a single channel characterization. In this experiment, LP2 oriented at 45° with respect to the extraordinary axis of LCoS is used to form a polarimeter [24]. The light intensity after LP2 is measured with a power meter (Newport, 1918-R). When the phase patterns from 0 to 2π at the left half section of LCoS are loaded, the φ1(λ) used for achieving linearly polarization can be obtained once the maximum or minimum measured intensity detected. Then, we detect the optical intensity behind the LP3 by varying φ2(λ) from 0 to 2π at the right half section of LCoS and the curve of the optical intensity versus the loaded phase is illustrated in Fig. 3(a). Its theoretical function can be described as f(φ2) = I0(1-cos(φ2))/2, which can be derived from the Malus law when the initial azimuthal angle of linearly polarized light is orthogonal to the transmission direction of LP3. It is clearly observed that the measured intensity curve agrees well with the theoretical fitting curve. When the azimuthal angle of linearly polarized light is rotated from 0 to π, the transmitted power varies between −54.3 dBm and −7.84 dBm. For our configuration, the optical signal suffers 12 dB insertion loss, which is mainly caused by the 4 times transmission through the BS. To solve this problem, an oblique incident architecture can be used as an alternative for mitigating this insertion loss [25]. Theoretically, the ER can reach more than 50 dB, while the ER is 46.4 dB in our experiment. We infer that such ER gap results from the phase-modulation-dependent loss of LCoS or other phase pattern error [26]. Moreover, the PDL over 10 nm operation wavelength in the C-band is less than 1 dB, as shown in Fig. 3(b). The residual PDL is contributed by the transmittance difference between two orthogonal linear polarization components of BS, the bulk transmission grating, and LCoS. Therefore, the experimental results confirm the validity and excellent performance of our proposed method for programmable spectral processing.
Fig. 3 (a) The output signal intensity modulated with phase φ2 and its theoretical fitting curve. (b) Polarization dependent loss of the proposed programmable spectral processor.
Then, multichannel spectral processing experiment is performed to characterize the proposed programmable spectral processor. Firstly, the mapping between the optical wavelength and the pixel position on the left part and the right part of LCoS are calibrated with a near infrared charge coupled device (CCD) camera (OPHIR Photonics, SP620U-1550) and some narrow spectrum light source. The crossed-polarizer method is used to determine the position of each optical channel [27]. The optical field after the LP2 and the LP3 are detected. The pixel position on the LCoS is decided by phase scanning and simultaneously investigating whether the beam spot intensity on the CCD varied. The calibration result indicates that 203 pixels per nm is used on the LCoS. Subsequently, 10 DFB lasers, whose central wavelength are from 1552.22 nm to 1555.82 nm with a channel spacing of 50 GHz, are used as the multichannel signal source. The input power is 10 dBm for every channel. The output modulated signal after LP3 is coupled to the optical fiber for the spectrum measurement by the optical spectrum analyzer (OSA, Yokogawa, AQ6370C) with a spectral resolution of 0.02 nm. Figures 4(a) and 4(b) show the typical spectral processing results of 10-wavelength-channel optical input. Figure 4(a) is the maxim achievable signal intensity by manipulate φ2 to make the polarization direction of the output signals be parallel to the transmission axis of the LP3. The transmitted signal intensity is around −15 dBm with a 5.8 dB fluctuation. Since the 10-channel optical signals with the same power are projected to different regions on LCoS, we infer that such intensity fluctuation is caused by the non-uniform phase response of LCoS [28]. Thus, the IL of our setup can reach 13 dB regardless of the BS induced power splitting effect. Figure 4(b) shows that unwanted signal can be selectively filtered out by manipulating the SOP of the input signals independently. From the experimental results, we can find that the channel isolation is about 30-40 dB and the attenuation fluctuation in the stop band is around 7 dB. It is noticeable that there exists the consistent trends for the intensity fluctuation in Fig. 4(a) and the intensity fluctuation of the stop band in Fig. 4(b), which is depicted with the blue curve in Fig. 4. Thus, we can attribute above attenuation fluctuation to the phase response non-uniformity of LCoS. Therefore, a higher channel isolation can be expected if the non-uniform phase response of LCoS can be mitigated. Moreover, the intensity fluctuation in the stop band is larger than that in the pass band, which we can attribute to larger phase response non-uniformity [29] and stronger interference induced by the weak parasite reflections from cover glass and/or transparent electrode (ITO) for the stop band [30].
Fig. 4 The 10-wavelength-channel measurement results of the maxim achievable signal intensity (a) and the intensity with selective filtering (b).
Furthermore, we characterize the variable grid granularity operation for the proposed programmable spectral processor. Considering that the spectrum is already spatially separated by the bulk transmission grating, the spectrum can be measured directly with high spatial resolution CCD [31]. In our experiment, CCD (OPHIR Photonics, SP620U-1550) is used for spectral measurement. Before measurement, 10-channel DFB lasers from 1549.82 nm to 1553.42 nm with a channel spacing of 50 GHz are used to calibration the pixel position corresponding to the operation wavelength. The calibration result indicates that 81 pixels per nm are used on the CCD. Each spot center is calculated with gray centroid method and marked with the cross in Fig. 5. Accordingly, the operation wavelength corresponding to different positions on CCD can be obtained. Then, an ASE source together with an optical filter is employed to investigate the performance of our configuration for tunable bandwidth filtering. The dispersed continuous spectrum with a bandwidth of 390 GHz are also shown in Fig. 5(a). The input light is firstly transformed to linear polarization by loading a phase pattern φ1(λ) on the left half section of LCoS. Subsequently, the phase pattern φ2(λ) is loaded on the right half section of LCoS to selectively filter the unwanted signal. The spatial intensity pattern in the spectral domain is shown in Fig. 5(b). It is obviously that the optical signal in the short wavelength is filtered. However, there exists an intensity transition region between the pass band and the stop band, which degrades filtering performance. We think it is caused by the phase deviation arising from the non-negligible fringing field effect of LCoS [32].
Fig. 5 (a) Spatial intensity pattern for the optical signal with 390 GHz bandwidth and (b) spatial intensity pattern for the filtered signal with 200 GHz bandwidth on the CCD.
Moreover, different phase patterns φ2(λ) can be employed on the right section of LCoS to filter the input signal selectively. When φ2(λi) is set to φ2,min(λi) = π/2 + 2[β-α(λi)], wavelength λi can be filter. In the experiment, the φ2,min(λi) is employed on the LCoS pixel by pixel to investigate the filter characteristic. It is observed that the output spectrum is distinctly changed when φ2,min(λi) is loaded with 20 pixels interval. The modulated spectrum is illustrated in Fig. 6(a). The result indicates that the wavelength can be digitally tuned with 0.08 nm resolution, which agrees with the theoretical wavelength resolution of grating. If a dispersed element with a higher wavelength resolution [12–14] is used, the signal can be filtered with finer granularity. Besides, the central wavelength tuning can be realized by applying φ2,max(λi) = 2[β-α(λi)] to the selected wavelength range. The central wavelength tuned with a step of 0.16 nm is shown in Fig. 6(b). These results indicate that the central wavelength can be flexibly controlled with a variable bandwidth. Meanwhile, we noticed that the edge of spectra are flatten gradually, which is also caused by the fringing field effect of LCoS. By increasing the pixel size of LCoS [33], changing the LC modes [34], and designing new pixel electrode structure [35], such effect can be mitigated for our configuration.
Fig. 6 (a) Flexible operation of bandwidth for our proposed PSP and (b) central wavelength tuning by a wavelength step of 0.16 nm under the condition of fixed bandwidth.
4. Conclusion
In summary, we have demonstrated a novel programmable spectral processor based on spectral polarization manipulation with a spatial light modulator. This approach overcomes the polarization-dependent feature of LCoS and applies to arbitrary input polarization state. To achieve high ER and low PDL, the LCoS is suggested to manipulate the SOP of the input signals in the optical spectral domain to achieve the intensity modulation instead of steering the dispersed beams to different directions. As a result, the ER can reach 46.4 dB with a PDL of less than 1 dB. For the wavelength filtering operation, the channel isolation is around 40 dB. Meanwhile, the tunable optical filtering can be realized with a resolution of 0.08 nm. Our proposed PSP allows to modulate both the phase and polarization of input light in the spectral domain.
Funding
National Natural Science Foundation of China (NSFC) (61575071, 61377073); National Key Research and Development Program of China (2016YFE0121300).
References
1. R. Rudnick, A. Tolmachev, D. Sinefeld, O. Golani, S. Ben-Ezra, M. Nazarathy, and D. M. Marom, “Sub-GHz resolution photonic spectral processor and its system applications,” J. Lightwave Technol. 35(11), 2218–2226 (2017). [CrossRef]
2. K. Suzuki, K. Seno, and Y. Ikuma, “Application of waveguide/free-space optics hybrid to ROADM device,” J. Lightwave Technol. 35(4), 596–606 (2017). [CrossRef]
3. K. Sorimoto, K. Tanizawa, H. Uetsuka, H. Kawashima, M. Mori, T. Hasama, H. Ishikawa, and H. Tsuda, “Compact and phase-error-robust multilayered AWG-based wavelength selective switch driven by a single LCOS,” Opt. Express 21(14), 17131–17149 (2013). [CrossRef] [PubMed]
4. H. Hu, D. Kong, E. Palushani, M. Galili, H. C. H. Mulvad, and L. K. Oxenløwe, “320 Gb/s Nyquist OTDM received by polarization-insensitive time-domain OFT,” Opt. Express 22(1), 110–118 (2014). [CrossRef] [PubMed]
5. J. Pupeikis, N. Bigler, S. Hrisafov, C. R. Phillips, and U. Keller, “Programmable pulse shaping for time-gated amplifiers,” Opt. Express 27(1), 175–184 (2019). [CrossRef] [PubMed]
6. X. Wang and N. Wada, “Spectral phase encoding of ultra-short optical pulse in time domain for OCDMA application,” Opt. Express 15(12), 7319–7326 (2007). [CrossRef] [PubMed]
7. T. Yeminy, D. Sadot, and Z. Zalevsky, “Spectral and temporal stealthy fiber-optic communication using sampling and phase encoding,” Opt. Express 19(21), 20182–20198 (2011). [CrossRef] [PubMed]
8. D. Sinefeld, C. R. Doerr, and D. M. Marom, “A photonic spectral processor employing two-dimensional WDM channel separation and a phase LCoS modulator,” Opt. Express 19(15), 14532–14541 (2011). [CrossRef] [PubMed]
9. E. Wohlgemuth, Y. Yoffe, T. Yeminy, Z. Zalevsky, and D. Sadot, “Demonstration of coherent stealthy and encrypted transmission for data center interconnection,” Opt. Express 26(6), 7638–7645 (2018). [CrossRef] [PubMed]
10. D. M. Marom, P. D. Colbourne, A. D’Errico, N. K. Fontaine, Y. Ikuma, R. Proietti, L. Zong, J. M. Rivas-Moscoso, and I. Tomkos, “Survey of photonic switching architectures and technologies in support of spatially and spectrally flexible optical networking,” J. Opt. Commun. Netw. 9(1), 1–26 (2017). [CrossRef]
11. H. Yang, B. Robertson, P. Wilkinson, and D. Chu, “Small phase pattern 2D beam steering and a single LCOS design of 40 1 × 12 stacked wavelength selective switches,” Opt. Express 24(11), 12240–12253 (2016). [CrossRef] [PubMed]
12. N. Goldshtein, D. Sinefeld, O. Golani, R. Rudnick, L. Pascar, R. Zektzer, and D. M. Marom, “Fine resolution photonic spectral processor using a waveguide grating router with permanent phase trimming,” J. Lightwave Technol. 34(2), 379–385 (2016). [CrossRef]
13. S. Asraf, T. Yeminy, D. Sadot, and Z. Zalevsky, “Proof of concept for ultrahigh resolution photonic spectral processor,” Opt. Express 26(19), 25013–25019 (2018). [CrossRef] [PubMed]
14. D. Sinefeld, S. Ben-Ezra, and D. M. Marom, “Nyquist-WDM filter shaping with a high-resolution colorless photonic spectral processor,” Opt. Lett. 38(17), 3268–3271 (2013). [CrossRef] [PubMed]
15. E. Ronzitti, M. Guillon, V. de Sars, and V. Emiliani, “LCoS nematic SLM characterization and modeling for diffraction efficiency optimization, zero and ghost orders suppression,” Opt. Express 20(16), 17843–17855 (2012). [CrossRef] [PubMed]
16. S. Frisken, G. Baxter, D. Abakoumov, H. Zhou, I. Clarke, and S. Poole, “Flexible and grid-less wavelength welective switch using LCOS technology,” in Optical Fiber Communication Conference/National Fiber Optic Engineers Conference,(Optical Society of America, 2011), paper OTuM3.
17. R. Rudnick, L. Pascar, B. Frenkel, and D. M. Marom, “Polarization diverse fine resolution photonic spectral processor,” in Optical Fiber Communication Conference and Exhibition (IEEE, 2016), paper W1E.4. [CrossRef]
18. D. Sinefeld and D. M. Marom, “Insertion loss and crosstalk analysis of a fiber switch based on a pixelized phase modulator,” J. Lightwave Technol. 29(1), 69–77 (2011). [CrossRef]
19. T. Lu, M. Pivnenko, B. Robertson, and D. Chu, “Pixel-level fringing-effect model to describe the phase profile and diffraction efficiency of a liquid crystal on silicon device,” Appl. Opt. 54(19), 5903–5910 (2015). [CrossRef] [PubMed]
20. Y. Gao, G. Chen, X. Chen, Q. Zhang, Q. Chen, C. Zhang, K. Tian, Z. Tan, and C. Yu, “High-resolution tunable filter with flexible bandwidth and power attenuation based on an LCoS Processor,” IEEE Photonics J. 10(6), 1–8 (2018). [CrossRef]
21. M. Nakajima, K. Suzuki, K. Seno, T. Goh, R. Kasahara, M. Fukutoku, Y. Miyamoto, and T. Hashimoto, “Over-100-spatial-channel programmable spectral processor for SDM signal monitoring,” in Optical Fiber Communication Conference and Exhibition, (IEEE, 2018), paper W1E.2. [CrossRef]
22. M. Nakajima, K. Suzuki, K. Yamaguchi, H. Ono, T. Goh, M. I. Fukutoku, Y. Miyamoto, and T. Hashimoto, “Multilane photonic spectral processor integrated in a spatial and planar optical circuit for a space-division multiplexing network,” J. Lightwave Technol. 36(2), 309–317 (2018). [CrossRef]
23. Q. Zhan, “Cylindrical vector beams: from mathematical concepts to applications,” Adv. Opt. Photonics 1(1), 1–57 (2009). [CrossRef]
24. Z. Hong, L. Zhu, S. Fu, M. Tang, P. Shum, and D. Liu, “A robust and fast polarimeter based on spatial phase modulation of liquid crystal on silicon (LCoS),” in Asia Communications and Photonics Conference, (ACP, 2015), paper AM1A.3. [CrossRef]
25. X. Zheng, A. Lizana, A. Peinado, C. Ramírez, J. L. Martínez, A. Márquez, I. Moreno, and J. Campos, “Compact LCOS–SLM based polarization pattern beam generator,” J. Lightwave Technol. 33(10), 2047–2055 (2015). [CrossRef]
26. D. Yu, S. Fu, Z. Hong, M. Tang, P. Shum, and D. Liu, “Characterization and mitigation of phase-modulation-dependent loss of liquid crystal on silicon,” Opt. Lett. 40(7), 1484–1487 (2015). [CrossRef] [PubMed]
27. X. Xun and R. W. Cohn, “Phase calibration of spatially nonuniform spatial light modulators,” Appl. Opt. 43(35), 6400–6406 (2004). [CrossRef] [PubMed]
28. J. Xu, S. Qin, C. Liu, S. Fu, and D. Liu, “Precise calibration of spatial phase response nonuniformity arising in liquid crystal on silicon,” Opt. Lett. 43(12), 2993–2996 (2018). [CrossRef] [PubMed]
29. L. Teng, M. Pivnenko, B. Robertson, R. Zhang, and D. Chu, “A compensation method for the full phase retardance nonuniformity in phase-only liquid crystal on silicon spatial light modulators,” Opt. Express 22(21), 26392–26402 (2014). [CrossRef] [PubMed]
30. J. L. Martínez, I. Moreno, M. del Mar Sánchez-López, A. Vargas, and P. García-Martínez, “Analysis of multiple internal reflections in a parallel aligned liquid crystal on silicon SLM,” Opt. Express 22(21), 25866–25879 (2014). [CrossRef] [PubMed]
31. S. Jiang, H. Jia, Y. Lei, X. Shen, J. Cao, and N. Wang, “Novel method for determination of optical rotatory dispersion spectrum by using line scan CCD,” Opt. Express 25(7), 7445–7454 (2017). [CrossRef] [PubMed]
32. K. H. Fan-Chiang, S. T. Wu, and S. H. Chen, “Fringing-field effects on high-resolution liquid crystal microdisplays,” J. Disp. Technol. 1(2), 304–313 (2005). [CrossRef]
33. M. Wang, L. Zong, L. Mao, A. Marquez, Y. Ye, H. Zhao, and F. J. V. Caballero, “LCoS SLM study and its application in wavelength selective switch,” Photonics 4, 22 (2017). [CrossRef]
34. K. H. Fan Chiang, S. T. Wu, and S. H. Chen, “Fringing field effect of the liquid-crystal-on-silicon devices,” Jpn. J. Appl. Phys. 41, 4577–4585 (2002). [CrossRef]
35. L. Yang, J. Xia, and X. Zhang, “Optimization of fringing field effect using dielectric separation and local electric field enhancement,” J. Disp. Technol. 11(3), 242–247 (2015). [CrossRef]
