Open Access
Add to BibSonomyAbstract
We present a new transmitter configuration using a single dual-electrode Mach-Zehnder Modulator (DE-MZM) for generating DPSK signal and subcarrier simultaneously. We then apply this new configuration to monitoring chromatic dispersion (CD) in DPSK systems by detecting the RF power at both of the clock and subcarrier frequencies. Our investigations show that the method can greatly improve CD monitoring capability. The impact of subcarrier on the DPSK performance is also investigated. Results show that by simply selecting the subcarrier close to half of the DPSK data rate frequency, we can significantly suppress the subcarrier induced power penalty to DPSK signal.
©2007 Optical Society of America
1. Introduction
Subcarrier multiplexing (SCM) has been demonstrated as a promising technique for monitoring a number of optical performance parameters such as optical signal-to-noise ratio (OSNR) and dispersion [1]. Moreover, it can be used to carry label information for switching and routing optical packets over high-bit-rate wavelength division multiplexed (WDM) networks [2]. Presently, differential phase-shift keying (DPSK) attracts significant attention in long-haul high capacity WDM systems because of its 3 dB lower requirement in OSNR and better resilience to fiber cross-phase modulation (XPM) than non-return-to-zero on-off-keying (NRZ-OOK) [3–5]. Thus, for providing efficient control and management, cost-effective SCM techniques are essential in DPSK systems. Recently, SCM techniques were demonstrated in DPSK systems for optical label switching [6, 7]. However, an additional external modulator is required in the transmitter for generating a subcarrier.
On the other hand, chromatic dispersion (CD) is an important factor limiting transmission distance of DPSK systems. It may change with the dynamic network reconfigurations and variations in environmental conditions such as temperature [8]. As a result, it is essential to monitor the residual CD for DPSK systems. Although the CD monitoring is intensively investigated in OOK systems, including using some SCM techniques [1, 9–11], little work has been done on CD monitoring for DPSK systems. Pan et al. reported a CD monitoring technique for DPSK signal, based on the regenerated clock power detection [12]. This method is quite simple. However, because the monitoring sensitivity and range depend on the frequency of clock tone, the monitoring capability is greatly limited by the data bit-rate.
In this paper, we present a new transmitter configuration using a single dual electrode Mach-Zehnder modulator (DE-MZM), which can generate a DPSK signal together with a subcarrier simultaneously (some preliminary work was reported in [13]). Then we apply this new configuration to CD monitoring in DPSK systems by detecting the power at both the clock frequency and the subcarrier frequency. By utilizing the subcarrier with different frequencies, flexible CD monitoring sensitivity and large range can be acquired. Experimental and simulation results show that this CD monitoring technique can improve the monitoring capability. The subcarrier induced power penalty to DPSK signal is also investigated. The impact of subcarrier frequency on the power penalty shows that by simply selecting the subcarrier frequency close to half of the DPSK data-rate frequency, we can significantly suppress the subcarrier induced power penalty to DPSK signal.
2. Transmitter configuration for simultaneous DPSK and subcarrier generation
DE-MZM has been reported for generating RZ/CSRZ-DPSK signal [4]. Following the similar configuration as the DE-MZM reported in [4], we here propose a new transmitter structure for modulating a DPSK data signal and a subcarrier simultaneously, as shown in Fig. 1(a). Both a data signal and a subcarrier signal (fsc) are first divided into two equal branches. The respective data and subcarrier signal of the upper branch are combined by a combiner. The subcarrier signal of the lower branch is delayed by 1/2fsc via a delay line and then combined with the data signal of the lower branch. The two combined signals are used to drive the upper and lower arms of the DE-MZM separately. Unlike the operation condition of the DE-MZM in [4], our DE-MZM is biased at the quadrature, i.e., the bias voltages of the upper and lower arms are set at Vπ/4 and -Vπ/4, respectively, where Vπ is the switching voltage of the DE-MZM for π phase shift. It is noted that with our scheme, the subcarrier frequency is not limited to half of the DPSK data rate, and no synchronization between them is required.
Theoretical analysis shows that the output of the DE-MZM E out(t) can be written as [4]:
where E in is the optical field amplitude of input signal, a is the modulator insertion loss in dB, and the phase terms ϕ upper(t) and ϕ lower(t) are given by
where Vπ is the switching voltage of the DE-MZM, D(t) is the data signal being 0 and Vπ, and VSC∙sin(2πfSCt) represents the subcarrier signal. Substituting (2) and (3) into (1), we have
Using the Jacobi-Anger identity to expand the output power of the DE-MZM into Fourier series and neglecting the high order terms, we get the approximate output power,
where J 1(x) is the first-kind Bessel function of order 1.
Eq. (4) shows that the DE-MZM encodes the data with the binary phase modulation. Eq. (5) indicates that the subcarrier signal at fsc is sinusoidal intensity modulated to the output power and m = 2J 1(2πVsc/Vπ) is the modulation index of the subcarrier. Fig. 1(b) shows the RF spectra simulated by commercial software (VPItransmissionMaker) [14] using the proposed transmitter configuration and the direct detection when the output light experiences 0 and 68 ps/nm accumulated CD. Here, the DPSK signal is 43 Gb/s and the subcarrier is 20 GHz with 30% modulation index. Without using an additional modulator, the subcarrier is modulated to the DPSK data. No DPSK signal spectrum is seen in the RF spectrum since the envelop detection removes the phase information. By varying the phase of the subcarrier signal, we have also observed that the synchronization between the subcarrier and the DPSK data does not affect the RF power of the generated subcarrier signal.
Fig. 1. (a) Transmitter configuration for simultaneous generation of DPSK signal and subcarrier using a single DE-MZM. (b) Simulated RF spectra of the 43 Gb/s DPSK with a 20 GHz subcarrier (upper: without CD; lower: with 68 ps/nm CD).
3. Chromatic dispersion monitoring for DPSK systems
When a light carrying a phase modulated DPSK signal and an intensity modulated subcarrier travels along a dispersive optical channel, CD causes phase modulation to amplitude modulation (PM-to-AM) conversion for the DPSK signal and a phase difference between the two sidebands of the subcarrier. As shown in the lower diagram of Fig. 1(b), these effects result in the regeneration of the clock tone (43 GHz) and the power fading of the subcarrier (20 GHz) after the photo detection [9, 12]. Note that another spike at 23 GHz in the lower diagram of Fig. 1(b) is the beating signal between the subcarrier and the clock tone. Consequently, CD monitoring can be realized by detecting the RF power of both the clock tone and the subcarrier.
Figure 2 illustrates the system configuration of our proposed CD monitoring scheme for DPSK systems. The SCM DPSK transmitter can be achieved by the scheme as described in Section 2. The light at 1552.5 nm propagates along a length of standard single mode fiber (SMF) followed by dispersion compensating fiber (DCF). Two erbium-doped fiber amplifiers (EDFAs) are used to compensate the attenuation induced by fiber and other components. An optical band-pass filter (OBPF) is used to remove the excess amplified spontaneous emission (ASE) noise. At the receiver side, the DPSK signal is demodulated with a Mach-Zehnder one bit delay interferometer (DI) and then detected with a balanced receiver followed by a limiting amplifier and a clock and data recovery device (CDR). By adjusting the variable optical attenuator (VOA), the transmission performance is measured by a bit-error-rate tester (BERT). For monitoring purpose, we adjust the length of fiber in the transmission link to emulate different amounts of residual CD. A small portion of light is tapped out before the DI and coupled into the CD monitoring module which consists of a photodetector and an RF spectrum analyzer (RFSA). We monitor the residual CD by measuring the RF power at the clock frequency and the subcarrier frequency. Here, the received optical power is kept constant so that the measured RF power is not affected by the optical power variation. In a practical implementation, the RFSA can be replaced by a narrow-band RF band-pass filter and an RF power detector at the corresponding frequency.
To assess the monitoring capability of our proposed scheme, we have conducted the experiment for 10.7 Gb/s DPSK system. Due to the lack of DE-MZM, we modulate a DPSK signal by a phase modulator and then add a subcarrier by a zero-chirp single electrode Mach-Zehnder modulator (MZM). We have also carried out the simulation by commercial software (VPItransmissionMaker) for both 10.7 Gb/s and 43 Gb/s DPSK systems, in which a single DE-MZM was used. Here, the 5 GHz and 20 GHz subcarriers were used in 10.7 Gb/s and 43 Gb/s DPSK systems, respectively. Fig. 3 plots the simulated and measured normalized RF power (i.e., the ratio of the RF power to the peak RF power) of the clock tone and subcarrier against the accumulated CD for 10.7 and 43 Gb/s DPSK systems. As seen in Fig. 3(a), the experiment results (symbols) match very well with the simulation results (curves), except for the places where the simulated clock tone power is very low. At these places, the clock tone may be hidden behind the PM-to-AM converted spectrum components of the DPSK signal and noise. Under such a situation, the tone power cannot be measured accurately, resulting in higher monitoring errors and narrower CD monitoring range [10]. The results have also shown that the regenerated clock tone is very sensitive to CD, but the monitoring range (the monotonically increasing region) is quite narrow, especially for the 43 Gb/s system in Fig. 3(b) (only several tens ps/nm). A subcarrier with lower frequency has a larger CD monitoring range (the monotonically decreasing region). In Fig. 3(a), the 5 GHz subcarrier power fades with CD while the 10.7 GHz clock power increases firstly and then decreases within the evaluated dispersion amount of around 1500 ps/nm. By referring to the subcarrier power, the monotonically decreasing region of the clock power can be distinguished from its monotonically increasing region. By detecting the RF power at both two frequencies, we can monitor CD with the high sensitivity of the clock tone in a more than doubled range for 10.7 Gb/s DPSK system. Similarly, for a 43 Gb/s DPSK system as shown in Fig. 3(b), by tracking the RF power at 20 GHz, we can achieve 150 ps/nm monitoring range. In such a high bit-rate system, it is imperative to make sure that the residual CD is as small as several ps/nm. Combining with the 43 GHz clock tone with 1 dB/(ps/nm) monitoring sensitivity, the CD can be monitored accurately and hence compensated fully with a tunable CD compensator. Therefore, as compared with the technique using the regenerated clock tone alone [12], by using a subcarrier together with the regenerated clock tone, we can significantly improve the monitoring range without compromising the monitoring sensitivity.
Fig. 3. Normalized RF power versus accumulated chromatic dispersion for (a) 10.7 Gb/s DPSK system; (b) 43 Gb/s DPSK system. Curves represent the simulation results and symbols represent the experimental results.
4. Subcarrier induced penalty and its suppression
As described in Section 3, the proposed method requires an additional subcarrier to the DPSK signal at the transmitter side. A subcarrier frequency less than clock frequency (in-band subcarrier) is preferred to increase the CD monitoring range. A subcarrier within the data bandwidth of the DPSK signal also decreases the cost of the subcarrier generator/detector and improves spectral efficiency [6]. However, like the case in OOK systems, the in-band subcarrier may induce impairment to the DPSK data signal and result in power penalty. To evaluate this impairment, we have experimentally measured the BER performance of a 10.7 Gb/s DPSK signal with subcarriers at 30% and 45% modulation indices after transmission of an 80-km standard SMF with full CD compensation. With the reference to the receiver sensitivity of DPSK data without subcarrier at BER of 10-9, the subcarrier induced power penalty to DPSK data is plotted against the subcarrier frequency in Fig. 4(a). We have also measured the eye diagrams before the balanced receiver. Fig. 4(b) shows the eye diagram without subcarrier and Figs. 4(c) and 4(d) show the eye diagrams with 2.5 GHz and 5.35 GHz subcarriers at 30% modulation index, measured at the destructive port of the DI. Fig. 4(a) reveals that as expected, the higher modulation index leads to higher power penalty of the DPSK data. Note that the similar conclusion has also been obtained in [5] in the context of channel identification using an SCM technique. It is also important to note that the power penalty also changes with the frequency of the subcarrier. As shown in Fig. 4(a), the lowest penalty occurs for the DPSK system with the subcarrier that has half of the DPSK data rate frequency (i.e., fSC = 5.35 GHz). The power penalty induced by the 5.35 GHz subcarrier is smaller than 0.5 dB for both modulation indices and no obvious difference can be observed between the eye diagrams in Fig. 4(b) and Fig. 4(d). Moreover, the power penalty induced by the subcarrier increases with the difference between the subcarrier frequency and the half of the DSPK data rate frequency. Fig. 4(c) shows the eye diagram which is severely distorted by 2.5 GHz subcarrier. Similar results have also been observed in [7] in the context of SCM labelled DPSK payload for IP packet routing.
Fig. 4. (a) Measured subcarrier induced power penalty versus subcarrier frequency at 30% and 45% modulation indices; (b) eye diagram without subcarrier; (c) eye diagram with 2.5 GHz subcarrier (m = 30%); (d) eye diagram with 5.35 GHz subcarrier (m = 30%).
The observations in Fig. 4 can be attributed to the DI for DPSK demodulation. As Eq. (5) shows, the subcarrier modulated optical signal has the sinusoidal power fluctuation. Therefore, due to the one bit delay between its two arms, DI not only demodulates DPSK signal back to intensity modulated data [3], but also leads to the phase difference between the sinusoidal fluctuated optical power of two arms. When the frequency of subcarrier equals half of the DSPK data rate frequency, the signal with one bit delay is out of phase with the original signal. As a result, the amplitude fluctuation induced by the subcarrier is cancelled at the DI output and the resulted power penalty is greatly suppressed. With the increase of the difference between the subcarrier frequency and half of the DSPK data rate frequency, the suppression of amplitude fluctuation becomes less effective and the power penalty induced by subcarrier is larger. Therefore, unlike the SCM technique in OOK systems, by simply selecting the subcarrier frequency close to half of the DPSK data rate frequency, we can suppress the subcarrier induced power penalty without introducing any additional components. Note that, in the CD monitoring system of Section 3, in order to keep their second harmonics away from the clock tones, the subcarriers with frequencies slightly deviated from half of the corresponding DPSK data rate frequencies were used, which were 5 GHz and 20 GHz for 10.7 Gb/s and 43 Gb/s systems, respectively.
5. Conclusion
We have presented a new transmitter configuration for generating a DPSK signal together with a subcarrier simultaneously using a single DE-MZM. This new subcarrier multiplexed DPSK transmitter can be applied to monitoring the CD in DPSK systems by detecting the RF power at both of the regenerated clock and subcarrier frequencies. Experimental and simulation results have shown that by using a subcarrier together with the regenerated clock tone, we can significantly improve the CD monitoring range without compromising the monitoring sensitivity. Furthermore, the subcarrier induced power penalty to DPSK data has been investigated experimentally. By simply selecting the subcarrier frequency close to half of the DPSK data rate frequency, we can suppress the subcarrier induced power penalty without introducing any additional components.
References and links
1. G. Rossi, T. E. Dimmick, and D. J. Blumenthal, “Optical performance monitoring in reconfigurable WDM optical networks using subcarrier multiplexing,” J. Lightwave Technol. 18,1639–1648 (2000). [CrossRef]
2. Z. Zhu, V. J. Hernandez, M. Y. Jeon, J. Cao, Z. Pan, and S. J. B. Yoo, “RF photonics signal processing in subcarrier multiplexed optical-label switching communication systems,” J. Lightwave Technol. 21,3155–3166 (2003). [CrossRef]
3. A. H. Gnauck and P. J. Winzer, “Optical phase-shift-keyed transmission,” J. Lightwave Technol. 23,115–130 (2005). [CrossRef]
4. Y. J. Wen, A. Nirmalathas, and D.-S. Lee, “RZ/CSRZ-DPSK and chirped NRZ signal generation using a single-stage dual-electrode Mach-Zehnder modulator,” IEEE Photon. Technol. Lett. 16,2466–2468 (2004). [CrossRef]
5. S. B. Jun, H. Kim, P. K. J. Park, J. H. Lee, and Y. C. Chung, “Pilot-tone-based WDM monitoring technique for DPSK systems,” IEEE Photon. Technol. Lett. 18,2171–2173 (2006). [CrossRef]
6. T. Flarup, C. Peucheret, J. J. V. Olmos, Y. Geng, J. Zhang, I. T. Monroy, and P. Jeppesen, “Labeling of 40 Gbit/s DPSK payload using in-band subcarrier multiplexing,” in Optical Fiber Communication Conference, 2005 OSA Technical Digest Series (Optical Society of America, 2005), paper OWB7.
7. I. T. Monroy, J. J. V. Olmos, M. G. Larrode, T. Koonen, and C. D. Jimenez, “In-band 16-QAM and multi-carrier SCM modulation to label DPSK payload signals for IP packet routing,” Opt. Express 14,1000–1005 (2006). [CrossRef]
8. W. Hatton and M. Nishimura, “Temperature dependence of chromatic dispersion in single mode fibers,” J. Lightwave Technol. 4,1552–1555 (1986). [CrossRef]
9. M. N. Petersen, Z. Pan, S. Lee, S. A. Havstad, and A. E. Willner, “Online chromatic dispersion monitoring and compensation using a single inband subcarrier tone,” IEEE Photon. Technol. Lett. 14,570–572 (2002). [CrossRef]
10. N. Liu, W. D. Zhong, P. Shum, C. Lu, and Y. X. Wang, “Improved chromatic dispersion monitoring technique,” Opt. Commun. 259,553–561 (2006). [CrossRef]
11. N. Liu, W.-D. Zhong, Y. J. Wen, C. Lu, L. Cheng, and Y. Wang, “PMD and chirp effects suppression in RF tone-based chromatic dispersion monitoring,” IEEE Photon. Technol. Lett. 18,673–675 (2006). [CrossRef]
12. Z. Pan, Y. Wang, Y. Song, R. Motaghian, S. Havstad, and A. E. Willner, “Monitoring chromatic dispersion and PMD impairments in optical differential phase shift-keyed (DPSK) systems,” in Optical Fiber Communication Conference, Vol.1 of 2003 OSA Technical Digest Series (Optical Society of America, 2003), paper WP1.
13. N. Liu, W. D. Zhong, and Y. J. Wen, “Simultaneous generation of DPSK and RF tone using a single DE-MZM for CD monitoring,” in Proc. 11th OptoElectronics and Communication Conference (OECC 2006), 7C2-4-1 - 7C2-4-2.
