Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Reduction of polarization-induced performance degradation in WDM PON utilizing MQW-SLD-based broadband source

Open Access Open Access

Abstract

We report on the reduction of polarization-induced performance degradation in WDM PON utilizing MQW-SLD-based ASE source for injection locking to FPLD. The results show that, to suppress the polarization-induced Q penalty sufficiently less than 0.5 dB, the MQW-SLD output should be depolarized within the locking range of the wavelength-locked FPLD.

©2007 Optical Society of America

1. Introduction

Wavelength-division-multiplexed (WDM) passive optical networks (PONs) offer many advantages including large capacity and upgradability [1]. From the perspective of practical deployment, however, initial high installation cost and complicated inventory management have been the most critical issues in these networks. To overcome this problem, it has been proposed to utilize wavelength-locked Fabry-Perot laser diodes (FPLDs) with the external injection of a broadband incoherent optical source, rather than wavelength-selected optical sources [2-4]. Until now, the amplified spontaneous emission (ASE) of an erbium-doped fiber amplifier (EDFA) has been mainly used as an external broadband optical source. However, it has been recently reported that semiconductor-based broadband sources such as super luminescent diodes (SLDs) and semiconductor optical amplifiers (SOAs) can be used instead of costly EDFAs [5-6]. In this case, a multiple-quantum-well (MQW) semiconductor device has many advantages over a bulk-type device, including a high output power and long-term reliability [7-8]. However, because most MQW devices generate highly polarized lights (e.g., degree of polarization (DOP) of >90%), when these devices are employed, the system performance can be seriously deteriorated by polarization dependency of the wavelength-locked FPLD [5],[9]. Polarized broadband light can be depolarized either by polarization-multiplexing two orthogonally polarized lights or by using a depolarizer [10]. However, since polarization-multiplexing scheme requires two broadband light sources together with a polarization beam combiner, using a depolarizer can be preferred in cost-sensitive WDM PONs. In this paper, we investigate the requirements of depolarizer in WDM PON utilizing MQW-SLD-based ASE source for injection locking to FPLD. To this end, we utilized a bi-refringent device for depolarizing the SLD output. By changing the differential group delay (DGD) of a bi-refringent device, we can simply change the DOP of the light. In real systems, however, a simple depolarizer such as a dual fiber-ring depolarizer could be used to fully depolarize the SLD light [11-12]. We show that simply reducing the DOP of MQW-SLD output is not enough to achieve low polarization-induced Q penalty. Our findings show that to suppress the polarization-induced Q penalty sufficiently less than 0.5 dB, the broadband source should be depolarized within the locking range of the wavelength-locked FPLD.

2. Experiment and Results

Figure 1 shows the experimental setup used to evaluate the polarization-induced performance degradation of the wavelength-locked FPLD. The output power and operating center wavelength of our MQW SLD were 20 dBm and 1543 nm, respectively. Because the DOP of our SLD output was partially depolarized, it was initially set to be 100% by using a polarizer in order to simulate the extreme case. Then the SLD output was depolarized by using a polarization controller and a bi-refringent fiber (in a practical system, a long bi-refringent fiber can be replaced by the depolarizer) [10-12]. The depolarized signal was sent to an optical circulator via an optical attenuator. The output of the optical circulator was spectrum-sliced by an arrayed waveguide grating (AWG) (3-dB bandwidth: 0.6 nm) and then injected into the FPLD via a polarization controller. The inset shows the measured optical spectrum of the injected light. The FPLD was directly modulated at 622 Mbps (non-return-to-zero data, pattern length: 231-1). The modulated FPLD output was transmitted through a 20-km single-mode fiber (SMF) and sent to a receiver to measure the bit error rate (BER).

 figure: Fig. 1.

Fig. 1. Experimental setup used to evaluate the polarization-induced performance degradation of the wavelength-locked FPLD.

Download Full Size | PDF

Figure 2 shows the DOP measured at the polarization analyzer while changing the length of bi-refringent fibers. In this experiment, we adjusted the polarization controller (PC1) to minimize the DOP for a given bi-refringent fiber. As expected, the DOP was decreased with increasing the DGD. The results show that the DGD should be higher than 20 ps to fully depolarize the output of SLD (i.e., DOP: <10 %).

 figure: Fig. 2.

Fig. 2. The measured DOP while varying the DGD of bi-refringent fiber.

Download Full Size | PDF

Using these depolarized broadband sources, we performed the wavelength locking of FPLD. We first measured the side-mode suppression ratio (SMSR) of the FPLD output while varying the optical power of the injected light and DGD as shown in Fig. 3. In this experiment, by controlling PC2, we obtained SMSR variation (i.e., maximum SMSR – minimum SMSR). The measured SMSR variation was decreased as either the DGD or injection power was increased. However, unlike the case of DOP measurement, the SMSR variation was quite large (~20 dB) even when the DOP of the ASE is lower than 10% (i.e., when DGD was 20 ps. See Fig. 2 for DOP vs. DGD). The results show that, to reduce the SMSR variation within 5 dB, the injection power and DGD should be higher than -5.5 dBm and 40 ps, respectively.

 figure: Fig. 3.

Fig. 3. The measured SMSR variation as a function of injected ASE power into FPLD.

Download Full Size | PDF

In principle, this SMSR degradation due to polarization dependency deteriorates the signal’s performance since relative intensity noise (RIN) is inversely proportional to SMSR (in our experiment, the RIN was measured to be about -110 dB/Hz when the SMSR was higher than 30 dB). Thus, we measured the signal’s performance (Q) of the wavelength-locked FPLD after the transmission over 20-km SMF. Figure 4 shows the maximum polarization-induced Q penalty (i.e., maximum Q (dB) – minimum Q (dB) @10-7 BER) measured while changing the polarization of the injected light at various DGD values when the injection signal power and received power were set to be -5.5 dBm and -25 dBm, respectively. The results show that, to suppress the polarization-induced Q penalty within 0.5 dB, the required DGD should be higher than 40 ps unlike the case of DOP measurement. To analyze these results, we measured the locking range (i.e., the maximum affordable wavelength misalignment between the center wavelength of broadband source and the correspondent lasing wavelength of FPLD to maintain the SMSR more than 30 dB) by tuning the ambient temperature of FPLD [4]. As a result, the locking range was measured to be about 0.3 nm. It should be noted that, the 40-ps DGD is the value which fully depolarizes the broadband source (i.e., DOP: <10%) within 0.3-nm bandwidth. Thus, we conclude that, to suppress the polarization-induced Q penalty sufficiently, the broadband source should be depolarized within the locking range of the wavelength-locked FPLD (not within the broadband source’s linewidth). However, this locking range is basically decreased as the injection power is decreased. Thus, the DGD of depolarizer should be determined properly by considering the injection power in a practical system. To evaluate the performance of polarization-insensitive MQW-SLD-based broadband source, we measured the BER of the wavelength-locked FPLD when the DGD was 100 ps in comparison with the case using a polarization-multiplexed broadband source as shown in Fig. 5 (injection power = -5.5 dBm). For polarization multiplexing, the outputs of two SLDs were combined by using a polarization beam combiner and used as a broadband source in Fig. 1. There was no much difference between the data. Thus, our results show that, the MQW SLD can be used as a cost-effective broadband source without any significant performance degradation by using the depolarization technique.

 figure: Fig. 4.

Fig. 4. The Q penalty measured while varying the DGD (injection power = -5.5 dBm, received power = -25.2 dBm).

Download Full Size | PDF

 figure: Fig. 5.

Fig. 5. The measured BER performances measured when DGD was 100 ps in comparison with the case using a polarization-multiplexed source.

Download Full Size | PDF

3. Conclusion

We investigated the requirements of depolarizer in WDM PON utilizing MQW-SLD-based ASE source for injection locking to FPLD. We found out that simply reducing the DOP of MQW-SLD output is not enough to achieve low polarization-induced Q penalty. To verify this, we measured the signal’s performance (Q) of the wavelength-locked FPLD after the transmission over 20-km SMF. The results showed that, to reduce the polarization-induced Q penalty sufficiently (< 0.5 dB), the broadband source should be fully depolarized (DOP: <10 %) within the locking range of the wavelength-locked FPLD (not within the broadband source’s linewidth).

References and links

1. E. S. Son, K. H. Han, J. K. Kim, and Y. C. Chung, “Bidirectional WDM passive optical network for simultaneous transmission of data and digital broadcast video service,” IEEE J. Lightwave Technol. 21, 1723–1727 (2003). [CrossRef]  

2. H. D. Kim, S. G. Kang, and C. H. Lee , “A low-cost WDM source with an ASE injected Fabry-Perot semiconductor laser,” IEEE Photon. Technol. Lett. 12, 1067–1069 (2000). [CrossRef]  

3. D.J. Shin, D. K. Jung, J. K. Lee, Y. K. Oh, J. H. Lee, H. S. Kim, C. H. Lee, S. T. Hwang, J. H. Ko, Y. J. Oh, T. I. Kim, and C. S. Shim, “Transmission of HDTV and Ethernet data over a WDM-PON employing ASE-injected Fabry-Perot laser diodes,” Proc. of Optical Fiber Communication Conference (Los Angeles, CA,2004), WO3.

4. D. J. Shin, D. K. Jung, H. S. Shin, S. B. Park, H. S. Kim, S. H. Kim, S. T. Hwang, E. H. Lee, J. K. Lee, Y. K. Oh, and Y. J. Oh, “AWG misalignment tolerance of 16 × 155 Mb/s WDM-PON based on ASE-injected FP-LDs,” Proc. of Optical Fiber Communication Conference (Anaheim, CA,2005), JWA54.

5. D. J. Shin, Y. C. Keh, J. W. Kwon, E. H. Lee, J. K. Lee, M. K. Park, J. W. Park, Y. K. Oh, S. W. Kim, I. K. Yun, H. C. Shin, D. Heo, J. S. Lee, H. S. Shin, H. S. Kim, S. B. Park, D. K. Jung, S. T. Hwang, Y. J. Oh, D. H. Jang, and C. S. Shim, “Low-cost WDM PON with colorless bidirectional transceivers,” IEEE J. Lightwave Technol. 24, 158–165 (2006). [CrossRef]  

6. D. J. Shin, Y. C. Keh, J. W. Kwon, E. H. Lee, J. K. Lee, M. K. Park, J. W. Park, J. K. Kang, Y. K. Oh, S. W. Kim, I. K. Yun, H. C. Shin, D. Heo, J. S. Lee, H. S. Shin, H. S. Kim, S. B. Park, D. K. Jung, S. T. Hwang, Y. J. Oh, D. H. Jang, and C. S. Shim, “C/S-band WDM-PON employing colorless bidirectional transceivers and SOA-based broadband light sources,” Proc. of Optical Fiber Communication Conference (Anaheim, CA,2005), PDP36.

7. S. Kakimoto and H. Watanabe, “Absorption loss coefficient of the active layer for 1.48-um bulk and MQW lasers,” IEEE J. Quantum Electron. 34, 110–112 (1998). [CrossRef]  

8. T. Kallstenius, J. Backstrom, U. Smith, and B. Stoltz, “On the degradation of InGaAsP/InP-based bulk lasers,” IEEE J. Lightwave Technol. 17, 2584–2594 (1999). [CrossRef]  

9. D. Heo, J. S. Lee, I. K. Yun, H. C. Shin, S. W. Kim, D. K. Jung, D. J. Shin, H. S. Kim, H. S. Shin, S. B. Park, S. T. Hwang, Y. J. Oh, Y. K. Oh, D. H. Jang, and C. S. Shim, “Polarization-independent, high-power, and angle-flared superluminescent diode for WDM-PON applications,” Proc. of IEEE LEOS Annual Meeting (Sydney, Australia,2005), 611–612.

10. D. Derickson , Fiber Optic Test and Measurement, (Hewlett-Packard professional books, Prentice Hall PTR, Upper Saddle River, New Jersey,1998, USA), Chap. 6.

11. M. Martinelli and J. C. Palais, “Dual fiber-ring depolarizer,” IEEE J. Lightwave Technol. 19, 899–905 (2001). [CrossRef]  

12. I. Yoon, B. Lee, and S. Park, “A fiber depolarizer using polarization beam splitter loop structure for narrow linewidth light source,” IEEE Photon. Technol. Lett. 18, 776–778 (2006). [CrossRef]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (5)

Fig. 1.
Fig. 1. Experimental setup used to evaluate the polarization-induced performance degradation of the wavelength-locked FPLD.
Fig. 2.
Fig. 2. The measured DOP while varying the DGD of bi-refringent fiber.
Fig. 3.
Fig. 3. The measured SMSR variation as a function of injected ASE power into FPLD.
Fig. 4.
Fig. 4. The Q penalty measured while varying the DGD (injection power = -5.5 dBm, received power = -25.2 dBm).
Fig. 5.
Fig. 5. The measured BER performances measured when DGD was 100 ps in comparison with the case using a polarization-multiplexed source.
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.