We present a monolithic polarization diversity coherent receiver by employing 120-degree optical hybrids on a silicon photonic integrated circuit (PIC). This PIC monolithically integrates silicon inverse tapers for fiber coupling, silicon polarization splitters, germanium high-speed photo detectors, and 120-degree optical hybrids based on 3x3 multimode interferometers (MMI). We demonstrate that 112-Gb/s polarization-division-multiplexed quadrature phase-shift keyed signals are detected in the wavelength range of 1530-1580 nm with comparable performance to commercial receivers.
© 2014 Optical Society of America
Advanced modulation formats are key enablers to increase the capacity of optical transport networks with a channel rate of 100 Gb/s and beyond [1–5]. Coherent optical detection with digital signal processing (DSP) not only enables high spectral efficiency systems, but provides cost-effective electronic equalization of transmission impairments and network monitoring as well. Classical coherent detection utilizes 90-degree optical hybrids to mix the incoming optical signal with a local oscillator (LO), into a pair of balanced photodiodes to detect the amplitude and phase of the signal. Such coherent receivers have been realized in discrete component formats, hybrid planar lightwave circuits (PLCs) with photo detectors [6–12], InP photonic integrated circuits (PICs) [13–21], and silicon PICs [22–24]. The PIC solution promises compactness, low cost, and low power consumption. In particular, silicon PICs offer the additional advantage of polarization diversity by using grating couplers  or on-chip polarization beam splitters/rotators (PBS/PR) [23,24].
Coherent optical detection can also be achieved by mixing a signal with an LO in other kinds of optical hybrids, such as 120-degree hybrids with three single-ended detectors . In general, balanced detection is superior in suppressing various noises, preferable for long-haul transmission driven by high performance requirements. Single-ended detection, however, may be lower in cost because of simpler transimpedance amplifiers (TIA) and less RF connections between optical and electrical interfaces. Compared with 90-degree hybrids, 120-degree hybrids have an intrinsically broader optical bandwidth , larger fabrication tolerance, and the ability to mitigate hardware-induced imperfections from analog-digital converters and digital signal processors. These advantages were numerically demonstrated by using a multi-mode interferometer (MMI) as a 120-degree hybrid [27,28]. As colorless detection over the whole telecommunication wavelength is preferable in high-capacity networks, broad-bandwidth optical hybrids for coherent detection would be demanded.
Xie et al. presented colorless detection of a large number of channels by suppressing LO relative intensity noise (RIN) and direct-detection signal terms using a 3x3 fiber coupler with discrete photo detectors (PDs) . In this paper, we present a highly compact integrated version on a silicon PIC, which monolithically integrates PBSs, MMI-based 120-degree hybrids, and germanium PDs. We demonstrate that our monolithic polarization diversity coherent receiver has a comparable performance to commercial receivers with 90-degree hybrids over a 1530-1580 nm wavelength range.
If a signal and an LO are launched into a symmetric 3x3 MMI with 1:1:1 power splitting, its three output photo currents areEq. (1) is the direct-detection term and the second term is the beat term, with a 120-degree difference among its three components. A proper rearrangement of the photo currents can yield the currents for in-phase/quadrature (I/Q) components with the suppression of the direct-detection term :
From MMI theory , the optical bandwidth of a MMI is proportional to the square of the ratio between waveguide width and MMI width. Hence, a 3x3 MMI has a larger intrinsic bandwidth than a 4x4 MMI 90-degree hybrid, resulting in larger fabrication tolerance which is needed for silicon and InP PIC fabrication.
3. Silicon PIC
We designed a silicon PIC with the circuit diagram shown in Fig. 1(a). The PIC includes two mode converters for fiber coupling based on inverse tapers, two polarization splitters based on silicon waveguide directional couplers, two silicon MMI-based 120-degree hybrids for TE and TM modes respectively, and six germanium PDs. The device information for PBSs and PDs has been reported in [30–32]. Both MMIs have a width of 12 μm and a length of 360 μm for TE and 270 μm for TM, respectively. The PIC was fabricated on an 8” silicon-on-insulator wafer with a top silicon thickness of 220 nm. The overall chip size is only 1.5 mm x 2.3 mm, with a photo in Fig. 1(b). The signal and LO are coupled into the chip from top/bottom sides [see Fig. 1(b)]. After coupling in, both signal and LO are split into TE and TM components. Same polarizations proceed to a 120-degree hybrid whose outputs are connected to three germanium PDs. After fabrication and initial characterizations, the chip was packaged with two cleaved single-mode fibers (SMFs) and two printed circuit boards with RF connectors [see Fig. 1(c)]. No TIAs were available for packaging this device.
4. Fiber-to-PD responsivities
Prior to the packaging, we measured the fiber-to-PD responsivities by launching an external-cavity tunable laser with a tuning range of 1520-1580 nm. Cleaved single-mode fibers were used. A polarization controller was used to adjust the input polarization. Figure 2 shows the responsivity for the six detectors when either TE or TM was launched into the chip. For the three PDs at the TE side, the PDs’ fiber-to-PD responsivities range between 0.13 and 0.145 A/W at the peak wavelength of ~1530 nm, and drop to about 0.1 A/W at 1580 nm. The responsivity drop may mainly be attributed to the PDs’ own spectral dependence. It is well known that for germanium on silicon, the absorption coefficients decrease beyond a wavelength of ~1565 nm. For TM-side PDs, the peak responsivities are about 0.1-0.118 A/W at a wavelength of ~1545 nm. The smaller peak responsivities for TM-side detectors are mainly induced by the insertion loss from PBSs. The PBSs based on silicon directional couplers were designed to have a coupling efficiency of >95% for TM but less than 2% for TE. Therefore, the TM light experiences more loss than TE (5% versus 2% in theory). For spectral dependence of TM, at the shorter wavelengths, the responsivity drops may also result from the PBS loss. The MMI hybrids were designed to have a 1-dB bandwidth of about 100 nm and therefore are expected to have less impact on the responsivity spectra. In summary, the spectral responses of germanium PDs, PBSs, MMI hybrids and mode converters all contribute to the responsivity spectra. Further studies of each individual component are needed to fully explain the results in Fig. 2. To break down the loss, the total fiber-to-PD insertion loss is about 4.5-5.5 dB, including a ~2.5-dB fiber coupling loss, a ~1.0-dB waveguide propagation loss, a ~0.5-dB MMI excess loss, and a ~1.0-dB detector loss. In addition, the polarization extinction ratios are more than 20 dB for the wavelength range of 1520-1580 nm, indicated by the responsivity of TE-side (or TM-side) PDs when the launched polarization is TM (or TE). These results demonstrate that the receiver can work for the wavelength range of 1520-1580 nm, covering more than the whole C-band.
We characterized the PD speed by launching an on-off-keying signal at 30 Gb/s with a wavelength of 1546 nm. The signal was modulated by a high-speed electro-absorption modulator with a CW laser input. The PD was biased at 2 V and a high-speed RF probe with a 50 ohm terminator was employed to collect the eye diagrams. The input polarization was set to either TE or TM depending on which side of the PDs locates. Figure 3 presents the eye diagrams for all six germanium PDs, with excellent eye openings. The frequency response of the PDs with the same design reported in  also verified that their 3-dB bandwidths exceed 20 GHz. These measurements indicate that the receiver can detect 30-Gbaud coherent optical signals.
5. 112-Gb/s PDM-QPSK detection
Figure 4 shows the experimental setup for testing the packaged receiver with polarization-division-multiplexed quadrature phase-shift keyed (PDM-QPSK) signals. The PDM-QPSK signal was generated by a commercial I/Q LiNbO3 modulator driven by 28-Gb/s pseudorandom bit sequences (PRBSs) of length 215–1 followed by a polarization multiplexer. We launched the signal at ~3 dBm and an LO at ~20 dBm into the chip. We used a polarization controller to adjust the polarization of LO prior to the PIC. The receiver output was connected to a real-time sampling oscilloscope equipped with six 40-Gsamples/s analog-to-digital converters with a bandwidth of 20 GHz. The received signals were analyzed with offline processing. For the offline DSP, the sampling skew was first corrected and the signal was synchronously sampled to 2 samples per symbol. The in-phase and quadrature components of the signal in each polarization were obtained with the operation described in Eq. (2). A butterfly equalizer with 19 taps adapted via the constant modulus algorithm was used for polarization demultiplexing and inter-symbol interference compensation . Carrier phase estimation was performed with the Viterbi & Viterbi algorithm . Differential decoding was used in the detection, and bit error ratios (BERs) were calculated using direct error counting.
Figures 5(a) and 5(b) show the plots of captured I1 versus I2 signals without any DSP for TE (a) and TM (b) branches. Plots of I1 versus I3 and I2 versus I3 are similar. The 2π/3 phase difference between I1 and I2 is evident from the tilt of the elliptical signal cloud. After DSP, the constellations were successfully recovered, shown in Figs. 5(c)–5(j) with an optical signal-to-noise ratio (OSNR) of 20 dB for four wavelengths from 1530.72 nm to 1580.35 nm. The performance of BER averaged over two polarizations versus OSNR is shown in Fig. 6 for four wavelengths, covering more than the whole C-band. For the wavelength of 1549.70 nm and 1565.50 nm, the device requires an OSNR of 15.4 dB at a BER of 10−3, which is 1.6 dB away from the theoretical limit. At 1530.72 nm, this OSNR requirement is 16.0 dB. The slight performance degradation at short wavelengths may be due to hybrid phase errors and also the LO high noises from optical amplifiers. For the wavelength of 1580.35 nm, an OSNR of 16.3 dB is required at a BER of 10−3, which may also be due to LO high noises. For a typical commercial coherent receiver based on discrete components with 90-degree hybrids, the OSNR requirement for a BER of 10−3 is in the range of 15.5-16 dB for differential decoding in lab experiments. Therefore, our extremely compact integrated optical device has a comparable BER performance. It is to be noted that this comparison does not include the impact of TIAs. As balanced TIAs may perform better than single-ended TIAs on suppressing noises, more rigorous comparisons are needed if both types of optical hybrids can be packaged with TIA arrays.
We have demonstrated a polarization diversity coherent receiver on a silicon PIC by using 120-degree optical hybrids instead of classical 90-degree hybrids. 112-Gb/s PDM-QPSK signals were detected over the wavelength range of 1530-1580 nm, with a BER performance comparable with commercial receivers. As coherent technology is expected to be soon employed in metro and shorter-distance networks, the demonstrated receiver may be a crucial candidate for low-cost and colorless coherent detection. Together with compact I/Q modulators demonstrated on silicon [32, 35–37], silicon photonics could be a viable and powerful platform of photonic integrated circuits in coherent optical communications.
We thank T.-Y. Liow and G.-Q. Lo of the Institute of Microelectronics, Singapore on fabrication, Y.-K. Chen, P. J. Winzer, D. Neilson, M. Zirngibl, and J. Fernandes from Bell Labs for their support.
References and links
1. T. Okoshi and K. Kikuchi, Coherent Optical Fiber Communications (Kluwer, 1998).
2. P. J. Winzer and R. Essiambre, “Advanced optical modulation formats,” Proc. IEEE 94(5), 952–985 (2006). [CrossRef]
3. A. D. Ellis, F. C. G. Gunning, B. Cuenot, T. C. Healy, and E. Pincemin, “Towards 1TbE using coherent WDM,” in Proc. OECC/ACOFT 2008, Sydney, Australia (2008), Paper WeA-1.
4. K. Kikuchi, “Digital coherent optical communication systems: fundamentals and future prospect,” IEICE Electron. Express 8(20), 1642–1662 (2011). [CrossRef]
5. M. Berger, “100G challenges and solutions,” in Optical Fiber Communication Conference/National Fiber Optic Engineers Conference (2010), Market Watch Session.
6. T. Inoue and K. Nara, “Ultrasmall PBS-integrated coherent mixer using 1.8%-delta silica-based planar lightwave circuit,” in European Conference on Optical Communication (2010), paper Mo.2.F.4.
7. J. Wang, M. Kroh, A. Theurer, C. Zawadzki, D. Schmidt, R. Ludwig, M. Lauermann, Z. Zhang, A. Beling, A. Matiss, C. Schubert, A. Steffan, N. Keil, and N. Grote, “Dual-quadrature coherent receiver for 100G Ethernet applications based on polymer planar lightwave circuit,” Opt. Express 19(26), B166–B172 (2011). [CrossRef] [PubMed]
8. Y. Nasu, T. Mizuno, R. Kasahara, and T. Saida, “Temperature insensitive and ultra wideband silica-based dual polarization optical hybrid for coherent receiver with highly symmetrical interferometer design,” Opt. Express 19(26), B112–B118 (2011). [CrossRef] [PubMed]
9. Y. Kurata, Y. Nasu, M. Tamura, H. Yokoyama, and Y. Muramoto, “Heterogeneous integration of high-speed InP PDs on silica-based planar lightwave circuit platform,” in European Conference on Optical Communication (2011), paper Th.12.5. [CrossRef]
10. S. Tsunashima, F. Nakajima, Y. Nasu, R. Kasahara, Y. Nakanishi, T. Saida, T. Yamada, K. Sano, T. Hashimoto, H. Fukuyama, H. Nosaka, and K. Murata, “Silica-based, compact and variable-optical-attenuator integrated coherent receiver with stable optoelectronic coupling system,” Opt. Express 20(24), 27174–27179 (2012). [CrossRef] [PubMed]
11. Y. Kurata, Y. Nasu, M. Tamura, R. Kasahara, S. Aozasa, T. Mizuno, H. Yokoyama, S. Tsunashima, and Y. Muramoto, “Silica-based PLC with heterogeneously-integrated PDs for one-chip DP-QPSK receiver,” Opt. Express 20(26), B264–B269 (2012). [CrossRef] [PubMed]
12. T. Ohyama, I. Ogawa, H. Tanobe, R. Kasahara, S. Tsunashima, T. Yoshimatsu, H. Fukuyama, T. Itoh, Y. Sakamaki, Y. Muramoto, H. Kawakami, M. Ishikawa, S. Mino, and K. Murata, “All-in-one 112-Gb/s DP-QPSK optical receiver front-end module using hybrid integration of silica-based planar lightwave circuit and photodiode arrays,” IEEE Photonics Technol. Lett. 24(8), 646–648 (2012). [CrossRef]
13. H. Takeuchi, K. Kasaya, Y. Kondo, H. Yasaka, K. Oe, and Y. Imamura, “Monolithic integrated coherent receiver on InP substrate,” IEEE Photonics Technol. Lett. 1(11), 398–400 (1989). [CrossRef]
14. R. Kunkel, H.-G. Bach, D. Hoffmann, C. M. Weinert, I. Molina-Fernandez, and R. Halir, “First monolithic InP-based 90°-hybrid OEIC comprising balanced detectors for 100GE coherent frontends,” in Proc. IPRM 2009 (2009), paper TuB2.2.
15. C. R. Doerr, L. Zhang, and P. J. Winzer, “Monolithic InP multi-wavelength coherent receiver,” in Optical Fiber Communication Conference/National Fiber Optic Engineers Conference (2010), paper PDPB1. [CrossRef]
16. A. Beling, N. Ebel, A. Matiss, G. Unterbörsch, M. Nölle, J. K. Fischer, J. Hilt, L. Molle, C. Schubert, F. Verluise, and L. Fulop, “Fully-integrated polarization-diversity coherent receiver module for 100G DP-QPSK,” in Optical Fiber Communication Conference/National Fiber Optic Engineers Conference (2011), paper OML5. [CrossRef]
17. R. Nagarajan, D. Lambert, M. Kato, V. Lal, G. Goldfarb, J. Rahn, M. Kuntz, J. Pleumeekers, A. Dentai, H. Tsai, R. Malendevich, M. Missey, K. Wu, H. Sun, J. McNicol, J. Tang, J. Zhang, T. Butrie, A. Nilsson, M. Reffle, F. Kish, and D. Welch, “10 channel, 100Gbit/s per channel, dual polarization, coherent QPSK, monolithic InP receiver photonic integrated circuit,” in Optical Fiber Communication Conference/National Fiber Optic Engineers Conference (2011), paper OML7. [CrossRef]
18. S. B. Estrella, L. A. Johansson, M. L. Masanovic, J. A. Thomas, and J. S. Barton, “Widely tunable compact monolithically integrated photonic coherent receiver,” IEEE Photonics Technol. Lett. 24(5), 365–367 (2012). [CrossRef]
19. H. Yagi, N. Inoue, Y. Onishi, R. Masuyama, T. Katsuyama, T. Kikuchi, Y. Yoneda, and H. Shoji, “High-efficient InP-based balanced photodiodes integrated with 90° hybrid MMI for compact 100 Gb/s coherent receiver,” in Optical Fiber Communication Conference/National Fiber Optic Engineers Conference (2013), paper OW3J.5.
20. M. Lu, H.-C. Park, A. Sivananthan, J. S. Parker, E. Bloch, L. A. Johansson, M. J. W. Rodwell, and L. A. Coldren, “Monolithic integration of a high-speed widely tunable optical coherent receiver,” IEEE Photonics Technol. Lett. 25(11), 1077–1080 (2013). [CrossRef]
21. M. Schell, H. Bach, K. Janiak, N. Keil, M. Moehrle, P. Runge, and Z. Zhang, “Coherent receiver PICs,” in Optical Fiber Communication Conference/National Fiber Optic Engineers Conference (2013), paper OW3J.6. [CrossRef]
22. C. R. Doerr, L. L. Buhl, Y. Baeyens, R. Aroca, S. Chandrasekhar, X. Liu, L. Chen, and Y.-K. Chen, “Packaged monolithic silicon 112-Gb/s coherent receiver,” IEEE PhotonicsTechnol. Lett. 23(12), 762–764 (2011). [CrossRef]
23. P. Dong, X. Liu, C. Sethumadhavan, L. L. Buhl, R. Aroca, Y. Baeyens, and Y. Chen, “224-Gb/s PDM-16-QAM modulator and receiver based on silicon photonic integrated circuits,” in Optical Fiber Communication Conference/National Fiber Optic Engineers Conference (2013), paper PDP5C.6. [CrossRef]
24. P. Dong, X. Liu, C. Sethumadhavan, L. L. Buhl, R. Aroca, Y. Baeyens, and Y. Chen, “Monolithic silicon photonic circuits enable 112-Gb/s PDM-QPSK transmission over 2560-km SSMF,” in Proc. ECOC’13 (2013), paper We.2.B.1.
25. T. Pfau, S. Hoffmann, O. Adamczyk, R. Peveling, V. Herath, M. Porrmann, and R. Noé, “Coherent optical communication: Towards realtime systems at 40 Gbit/s and beyond,” Opt. Express 16(2), 866–872 (2008). [CrossRef] [PubMed]
26. A. Besse, M. Bachmann, H. Melchior, L. B. Soldano, and M. K. Smit, “Optical bandwidth and fabrication tolerances of multimode interference couplers,” J. Lightwave Technol. 12(6), 1004–1009 (1994). [CrossRef]
27. P. J. Reyes-Iglesias, I. Molina-Fernández, A. Moscoso-Mártir, and A. Ortega-Moñux, “High-performance monolithically integrated 120° downconverter with relaxed hardware constraints,” Opt. Express 20(5), 5725–5741 (2012). [CrossRef] [PubMed]
28. P. J. Reyes-Iglesias, A. Ortega-Moñux, and I. Molina-Fernández, “Enhanced monolithically integrated coherent 120° downconverter with high fabrication yield,” Opt. Express 20(21), 23013–23018 (2012). [CrossRef] [PubMed]
29. C. Xie, P. J. Winzer, G. Raybon, A. H. Gnauck, B. Zhu, T. Geisler, and B. Edvold, “Colorless coherent receiver using 3x3 coupler hybrids and single-ended detection,” Opt. Express 20(2), 1164–1171 (2012). [CrossRef] [PubMed]
30. T.-Y. Liow, K.-W. Ang, Q. Fang, J.-F. Song, Y.-Z. Xiong, M.-B. Yu, G.-Q. Lo, and D.-L. Kwong, “Silicon modulators and germanium photodetectors on SOI: monolithic integration, compatibility, and performance optimization,” IEEE J. Sel. Top. Quantum Electron. 16(1), 307–315 (2010). [CrossRef]
33. S. J. Savory, “Digital coherent optical receivers: algorithms and subsystems,” IEEE J. Sel. Top. Quantum Electron. 16(5), 1164–1179 (2010). [CrossRef]
34. A. J. Viterbi and A. M. Viterbi, “Nonlinear estimation of PSK-modulated carrier phase with application to burst digital transmission,” IEEE Trans. Inf. Theory 29(4), 543–551 (1983). [CrossRef]
35. P. Dong, C. Xie, L. Chen, N. K. Fontaine, and Y.-K. Chen, “Experimental demonstration of microring quadrature phase-shift keying modulators,” Opt. Lett. 37(7), 1178–1180 (2012). [CrossRef] [PubMed]
36. P. Dong, C. Xie, L. L. Buhl, and Y. Chen, “Silicon microring modulators for advanced modulation formats,” in Optical Fiber Communication Conference/National Fiber Optic Engineers Conference (2013), paper OW4J.2. [CrossRef]