NARROW optical filtering, novel modulation formats,

Transcription

1 JOURNAL OF LIGHTWAVE TECHNOLOGY, VOL. 26, NO. 1, JANUARY 1, Tb/s WDM Transmission of Polarization- Multiplexed RZ-DQPSK Signals A. H. Gnauck, Senior Member, IEEE, G. Charlet, P. Tran, P. J. Winzer, Senior Member, IEEE, C. R. Doerr, Fellow, IEEE, J. C. Centanni, E. C. Burrows, T. Kawanishi, T. Sakamoto, and K. Higuma Abstract We demonstrate record 25.6-Tb/s transmission over 240 km using 160 WDM channels on a 50-GHz grid in the C+L bands. Each channel contains two polarization-multiplexed 85.4-Gb/s RZ-DQPSK signals, yielding a spectral efficiency of 3.2 b/s/hz in each band. Index Terms Differential quadrature phase shift keying, high spectral efficiency, modulation formats, optical fiber communications, optical modulation. I. INTRODUCTION NARROW optical filtering, novel modulation formats, multilevel signaling, forward error correction (FEC), and polarization interleaving or multiplexing have been used in the quest to increase the spectral efficiency (SE) of wavelength-division multiplexed (WDM) optical transmission systems. A number of high-se experiments have used a return-to-zero (RZ) format with a duty cycle of 50% to 67% (duty cycles below 50% typically just increase the filtering losses) because RZ formats are generally more robust than nonreturn-to-zero (NRZ) formats to the intersymbol interference produced by narrow optical filtering. Fifty copolarized 85.4-Gb/s RZ differential quadrature phase-shift keyed (DQPSK) signals have been transmitted through four 75-km fiber spans at an SE of 1.14 b/s/hz [1]. Sixty-four 85.4-Gb/s RZ-DQPSK signals have been transmitted through four 80-km spans at an SE of 1.6 b/s/hz using orthogonal-polarization launch of adjacent channels [2]. Polarization multiplexing and demultiplexing can increase the SE in systems where the effects of polarization-mode dispersion (PMD) and nonlinear depolarization are small. A capacity of 14 Tb/s was achieved using polarization-multiplexed 111-Gb/s 67%-RZ-DQPSK signals in a single 59-nm-wide extended L band ( nm) [3]. That experiment demonstrated transmission at an SE of 2.0 b/s/hz through two 80-km spans of fiber. The single band was later expanded to 84 nm ( nm), and 20.4-Tb/s transmission through three 80-km spans was achieved with the same spectral efficiency [4]. Manuscript received August 2, 2007; revised October 30, This work was supported in part by DARPA under Contract HR C-0153 supervised by Dr. J. Lowell. A. H. Gnauck, P. J. Winzer, C. R. Doerr, J. C. Centanni, and E. C. Burrows are with Alcatel-Lucent, Bell Labs, Holmdel, NJ USA ( gnauck@alcatel-lucent.com). G. Charlet and P. Tran are with Research and Innovation, Centre de Villarceaux, Alcatel-Lucent, Nozay, France. T. Kawanishi and T. Sakamoto are with the National Institute of Information and Communications Technologies (NICT), Tokyo , Japan. K. Higuma is with Sumitomo Osaka Cement, Chiba , Japan. Digital Object Identifier /JLT Recently, we reported a record SE of 3.2 b/s/hz using seventy-seven polarization-multiplexed 85.4-Gb/s 67%-RZ- DQPSK WDM channels on a 50-GHz grid, demonstrating 12.3-Tb/s capacity in the C band and transmission through three 80-km spans using erbium-doped fiber amplifiers (EDFAs) [5]. We subsequently reported a record 25.6-Tb/s transmission through three 80-km spans by utilizing both the C and L bands [6]. Distributed Raman amplification was added to allow the use of a single dispersion-compensating fiber (DCF) for both bands immediately following each span. This architecture also allowed the use of simple, single-stage EDFA optical repeaters. In addition, we employed an optical equalizer (OEQ) at the receiver to reduce distortions caused by narrow optical filtering in the system [7]. In this paper, we review this transmission experiment, describe additional measurements, and elaborate on the key enabling technologies. II. EXPERIMENTAL SETUP The DQPSK system is shown in Fig. 1. Eighty DFB lasers were operated on a 50-GHz ITU grid in both the C band ( to nm, to THz) and the L band ( to nm, to THz), for a total of 160 WDM channels. Two transmitters were used, corresponding to odd and even channels on a 100-GHz grid. The lasers in each band were combined in an arrayed waveguide grating router (AWG), and the two bands were combined in a C-band/L-band combiner. Each transmitter employed two external LiNbO3 modulators. The first modulator was a dual-drive Mach-Zehnder (MZ) type, driven sinusoidally in a push-pull configuration, and generating a 42.7-GHz, 67%-duty-cycle RZ optical pulse train. A second, nested MZ modulator [8] produced DQPSK modulation using two parallel dual-drive MZ modulators with a relative optical phase of. The parallel modulators were driven by identical 42.7-Gb/s NRZ electrical signals that were delayed relative to each other before being applied to the modulators in order to decorrelate the two quadratures. The delay was 23 bit slots in the even-channel transmitter, and 44 bit slots in the odd-channel transmitter. The 42.7-Gb/s electrical signals were true pseudorandom bit sequences (PRBS) of length, created by multiplexing four quarter-pattern-delayed copies of a 10.7-Gb/s PRBS. This was the longest practical pattern length that could be used, because the bit-error-rate (BER) test set receiver had to be programmed with the expected received pattern for each quadrature. After modulation, the C- and L-band channels from each transmitter were separated, amplified, and recombined. The odd and even channels were each prefiltered using one port of a 50 GHz/100 GHz interleaver (44-GHz passbands) /$ IEEE

2 80 JOURNAL OF LIGHTWAVE TECHNOLOGY, VOL. 26, NO. 1, JANUARY 1, 2008 Fig. 1. Experimental setup. C/L: C-band/L-band splitter or combiner; INT: 50-GHz/100-GHz interleaver or deinterleaver. and combined in a third interleaver. Finally, the 160 combined WDM channels were split into two paths (relative delay of approximately 20 ns) using a 3-dB coupler, and recombined in a polarization beamsplitter (PBS), using manual polarization controllers (PCs). Transmission was performed through three 80-km spans of SSMF. The power launched into each span was approximately dbm in the C band, and dbm in the L band. The average span loss was 17 db. Each span was backward Raman pumped at four wavelengths (1427, 1439, 1450, and 1485 nm) with powers ranging from 80 to 150 mw, yielding an on/off gain of about 15 db. Each span was immediately followed by a DCF (10-dB loss). After the DCF, the channels were split into separate C-band and L-band EDFAs, recombined, and sent to the next span. The residual dispersion in the 240-km link ranged from ps/nm to ps/nm across the signal channels. After transmission, a 0.6-nm-bandwidth (75-GHz) tunable grating filter and two optical deinterleavers were used to select a WDM channel. The appropriate deinterleaver ports were manually connected to pass either odd or even channels. Perchannel dispersion compensation was applied. After re-amplification and filtering by an additional 0.6-nm-bandwidth filter, a manual PC and a PBS were used to separate the two polarizations, and the polarization to be measured was selected using an optical switch. The selected polarization was passed through the optical equalizer, which will be described in detail in Section III. A 42.7-GHz clock was recovered by tapping off part of the DQPSK signal prior to demodulation and using a high- Q clock-recovery circuit. The signal was demodulated in a Mach-Zehnder delay interferometer (relative delay of 23.4 ps), which was phase-tuned to demodulate either the in-phase (I) or quadrature-phase (Q) component of the DQPSK signal. The 42.7-Gb/s demodulated signal was detected using a 40-GHz-bandwidth balanced photodiode circuit connected to an electronic decision circuit and 1:4 demultiplexer. The BER test set receiver was programmed for the expected 10.7-Gb/s output sequence. BERs on the four 10.7-Gb/s tributaries were nearly identical. Fig. 2. Examples of the OEQ response: The dashed curve is when tuned for a nearly flat response. The solid curve is tuned to create a 10-dB dip in the response at THz. III. THE OPTICAL EQUALIZER The OEQ, with thermooptically adjustable amplitude and phase, contained two taps integrated on a single silica-on-silicon chip with differential time delays of 10 ps and minimum loss of 2.2 db [9]. It was manually tuned. The passbands were repetitive, with a free spectral range of 100 GHz, thus enabling equalization of multiple channels on a 100-GHz grid. The OEQ could be tuned over about 200 GHz (twice the free spectral range), and amplitude dips could be adjusted from 0 db to greater than 20 db. Examples of the OEQ response are shown in Fig. 2. The dashed curve is when tuned for a nearly flat response. The solid curve is when tuned to create a 10-dB dip centered at THz. The benefit of the OEQ was demonstrated in back-to-back single-channel and three-channel WDM experiments without polarization multiplexing and demultiplexing [10]. We note that these experiments were done with a 50-GHz-bandwith AWG in place of the first 0.6-nm tunable filter in the receiver chain. However, it was later confirmed that results were nearly identical when a 0.6-nm tunable filter was used, because the limiting bandwidth was set by the interleavers and deinterleavers. In Fig. 3 the BER for the channel under test (odd

3 GNAUCK et al.: 25.6-TB/S WDM TRANSMISSION 81 Fig. 3. BER as a function of OSNR for several scenarios using the odd-channel transmitter for the channel under test. Open symbols are single-channel results, and solid symbols are three-channel WDM results. Fig. 5. BER as a function of OSNR using the even-channel transmitter for the channel under test. Open triangles are single-channel results, and solid triangles are three-channel WDM results. The five-channel result (solid squares) shows the additional penalty from adjacent even channels. Fig. 4. Received single-channel eye patterns after demodulation in a back-toback experiment. channel at THz) is plotted as a function of optical signal-to-noise ratio (OSNR, referenced to a 0.1-nm noise bandwidth) for several scenarios. In Fig. 3 the open circles are baseline single-channel single-polarization results without the OEQ, transmitter-side interleavers, or receiver-side AWG and deinterleavers. The combination of RZ-DQPSK format and a single 0.6-nm optical filter in the receiver yielded a required OSNR of db for a BER of. When the interleavers, 50-GHz AWG, and deinterleavers were added, the performance degraded, as shown by the open squares. The penalty was just over 2 db at a BER of, and increased quickly at lower error rates, reaching 5 db at a BER of. Adding the two copolarized adjacent WDM channels (the even channels at and THz) induced an additional penalty of 0.5 db at a BER of, and the additional penalty increased at lower error rates (solid squares). As can be seen in Fig. 3, the combined penalty from filtering and copolarized WDM crosstalk was about 8.7 db at a BER of. The OEQ was installed and the measurements were repeated. In the single-channel case with only 0.6-nm filtering, the OEQ could not improve performance, indicating that the intersymbol interference was negligible. In fact, the best performance was obtained when the OEQ was tuned for no equalization (flat response). However, with the interleavers, 50-GHz AWG, and deinterleavers present, the optical equalizer could improve the performance dramatically when tuned to create a dip of several db at the channel center. The open triangles in Fig. 3 show the single-channel performance with the OEQ. The penalty at a BER of was reduced from 5 db without the OEQ to less than 1 db with the OEQ. In the three-channel WDM case (solid triangles), the penalty at a BER of was reduced from 8.7 db without the OEQ to 2 db with the OEQ. A demodulated eye pattern is shown in Fig. 4(a) for the single-channel RZ-DQPSK signal with 0.6-nm optical filtering. Fig. 4(b) and (c) shows the eye patterns with tight optical filtering (interleavers, 50-GHz AWG, and deinterleavers added) without and with the OEQ, respectively. The eye pattern is clearly more open when the OEQ is used. Fig. 5 shows back-to-back measurements without polarization multiplexing using the even-channel transmitter to provide the test channel. With all the filtering in place, and using the OEQ, the result for a single channel at THz (open triangles) is very close to the result obtained for the odd channel in Fig. 3. This indicates that the two transmitters are well matched. Similarly, when the adjacent channels at and THz are added, the performance of the test channel (solid triangles) closely matches that shown in Fig. 2. Fig. 5 also shows a fivechannel result, obtained by the further addition of the even channels at and THz. These adjacent even channels, generated in the same transmitter as the test channel, create a significant additional crosstalk penalty (approximately 4 db at a BER of ). This is because the RZ-DQPSK spectrum of each channel is wide enough to overlap with its neighbors 100 GHz away. We stress that this penalty is due to the experimental constraint of having only two transmitters. The transmission results reported below could easily be improved if four transmitters (with appropriate interleaving filters) were to be used, such that the channels from each transmitter were spaced

4 82 JOURNAL OF LIGHTWAVE TECHNOLOGY, VOL. 26, NO. 1, JANUARY 1, 2008 Fig. 8. OSNR (squares) and BERs for the I and Q tributaries (circles) for every fifth channel in a 120-channel single-polarization transmission experiment. Fig. 6. (a) Transmitted (bottom, offset 10 db for clarity) and received (top) spectra. Transmitted L-band channel powers are preemphasized for line-amplifier gain peaking. (b) Close-up of a portion of the transmitted spectrum in the C band. Fig. 7. BERs for the I and Q tributaries for both polarizations in each WDM channel (triangles: odd channels, circles: even channels). There are 640 data points. by 200 GHz. This improvement would also be obtained in an actual system with separate transmitters for each WDM channel. IV TB/S TRANSMISSION USING POLARIZATION MULTIPLEXING Fig. 6(a) shows the transmitted (bottom trace) and received (top trace) optical spectra for the full polarization-multiplexed WDM system. In the L band, the transmitted channel powers were reduced over the range of nm in order to compensate for gain peaking in the in-line amplifiers. The received optical spectrum was taken after the last DCF, and before the final C- and L-band EDFAs. Fig. 6(b) shows a portion of the transmitted spectrum ( channels) in the C band. After the EDFAs, the OSNR ranged from 21 to 25 db in the C band, and from 23 to 27 db in the L band (reported in a 0.1-nm noise bandwidth). Fig. 7 shows the BERs after transmission for the I and Q components in each polarization for all the channels (a total of 640 data points). Because manual polarization tracking was employed, the BERs were obtained by counting errors for a period of 15 to 30 s after the polarization controller was optimized. This time period was short enough to avoid degradations from the slow changes occurring in the output polarization state, while still providing good error statistics at the observed error rates. The highest BERs occur at the edges of the C and L bands, where the OSNR was the lowest. However, all BERs are, the threshold of enhanced FEC with 7% overhead that enables a post-fec BER of less than. V. 9.6-TB/S TRANSMISSION IN A SINGLE POLARIZATION Single-polarization transmission was studied by disconnecting one of the paths in the polarization-multiplexing apparatus at the transmitter and eliminating the PBS in the receiver. In this experiment, only 120 channels (60 in C band and 60 in L band) were used, in order to avoid the edges of the bands where the received OSNR was typically lower due to the roll-off in the response of our EDFAs. The C-band channels ranged from to THz ( to nm), and the L band channels ranged from to THz ( to nm). The power launched into each span was kept at dbm in the C band and dbm in the L band, so that the average power per WDM channel increased by 1.25 db compared to the 160-channel case. Additionally, the OSNR required to obtain a given BER was (ideally) reduced by 3 db compared to the polarization-multiplexed transmission. Therefore, all channels were received with BERs substantially below the FEC threshold of. Fig. 8 shows the OSNR (squares) and BER for the I and Q components (circles) for every fifth channel across both bands, starting with the third channel in each band. VI. CONCLUSION We have achieved a record 25.6-Tb/s transmission over three 80-km spans of SSMF using 160 WDM channels on a 50-GHz grid in the C and L bands. We employed 85.4-Gb/s 67%-RZ-DQPSK modulation and polarization multiplexing to attain 160 Gb/s in each WDM channel, resulting in a spectral efficiency of 3.2 b/s/hz in each band. We used backward Raman pumping to increase the received OSNR and simplify the optical repeaters, while an optical equalizer was used in the receiver to minimize channel distortion caused by narrow optical filtering. We also demonstrated 9.6-Tb/s single-polarization WDM transmission on the same system. In this case, we used 120 WDM channels on a 50-GHz grid in the C and L bands, attaining a spectral efficiency of 1.6 b/s/hz in each band.

5 GNAUCK et al.: 25.6-TB/S WDM TRANSMISSION 83 ACKNOWLEDGMENT The authors would like to thank R. M. Jopson and D. Kilper for the loan of equipment. They also thank S. Cabot, M. A. Cappuzzo, E. Y. Chen, A. Wong-Foy, L. T. Gomez, and M. T. Santo for fabrication of the OEQ. REFERENCES [1] N. Yoshikane and I. Morita, 1.14 b/s/hz spectrally efficient Gb/s transmission over 300 km using copolarized RZ-DQPSK signals, J. Lightw. Technol., vol. 23, pp , Jan [2] N. Yoshikane and I. Morita, 160% spectrally efficient 5.12 Tb/s ( Gb/s RZ DQPSK) transmission without polarization demultiplexing, in Proc. ECOC 2004, Stockholm, Sweden, Postdeadline paper Th [3] A. Sano, H. Masuda, Y. Kisaka, S. Aisawa, E. Yoshida, Y. Miyamoto, M. Koga, K. Hagimoto, T. Yamada, T. Furuta, and H. Fukuyama, 14-Tb/s ( Gb/s PDM/WDM) CSRZ-DQPSK transmission over 160 km using 7-THz bandwidth extended L-band EDFAs, in Proc. ECOC 2006, Cannes, France, Postdeadline paper Th [4] H. Masuda, A. Sano, T. Kobayashi, E. Yoshida, Y. Miyamoto, Y. Hibino, K. Hagimoto, T. Yamada, T. Furata, and H. Fukuyama, 20.4-Tb/s ( Gb/s) transmission over 240 km using bandwidth-maximized hybrid Raman/EDFAs, in Proc. OFC 2007, Anaheim, CA, Postdeadline paper PDP20. [5] A. H. Gnauck, P. J. Winzer, L. L. Buhl, T. Kawanishi, T. Sakamoto, M. Izutsu, and K. Higuma, 12.3-Tb/s C-band DQPSK transmission at 3.2 b/s/hz spectral efficiency, in Proc. ECOC 2006, Cannes, France, Postdeadline paper Th [6] A. H. Gnauck, G. Charlet, P. Tran, P. J. Winzer, C. R. Doerr, J. C. Centanni, E. C. Burrows, T. Kawanishi, T. Sakamoto, and K. Higuma, 25.6-Tb/s C+L-band transmission of polarization-multiplexed RZ-DQPSK signals, in Proc. OFC 2007, Anaheim, CA, Postdeadline paper PDP19. [7] A. H. Gnauck, C. R. Doerr, P. J. Winzer, S. Cabot, M. A. Capuzzo, E. Y. Chen, A. Wong-Foy, L. T. Gomez, M. T. Santo, T. Kawanishi, and T. Sakamoto, Optical equalization of 42.7-Gbaud band-limited NRZ-DQPSK signals for high-spectral-efficiency transmission, in Proc. OFC 2007, Anaheim, CA, Paper OThN4. [8] T. Kawanishi, T. Sakamoto, and M. Izutsu, High-speed control of lightwave amplitude, phase, and frequency by use of electrooptic effect, IEEE J. Sel. Topics Quantum Electron., vol. 13, pp , Jan [9] C. R. Doerr, S. Chandrasekhar, P. J. Winzer, A. R. Chraplyvy, A. H. Gnauck, L. W. Stulz, R. Pafchek, and E. Burrows, Simple multichannel optical equalizer mitigating intersymbol interference for 40-Gb/s nonreturn-to-zero signals, J. Lightw. Technol., vol. 22, pp , Jan [10] A. H. Gnauck, C. R. Doerr, P. J. Winzer, and T. Kawanishi, Optical equalization of 42.7-Gbaud band-limited RZ-DQPSK signals, IEEE Photon. Technol. Lett., to be published. Gabriel Charlet was born in Rueil Malmaison, France, in He received an engineering degree from the École Supérieure d Optique, Orsay, France, in He joined Alcatel Research and Innovation, Nozay, France, in Since then, he has been working on WDM transmission systems and realized several multiterabit/s transmission records. He also addressed the topic of advanced modulation formats (PSBT, DPSK, APol RZ-DPSK, DQPSK, coherent detection of PDM QPSK, etc.). He is the author of 10 postdeadline papers in major conferences (OFC, ECOC, OAA, OECC) and more than 25 patents. Mr. Charlet received the Fabry de Gramont award for his work on fiber optics communication in Patrice Tran was born in Hatien, Vietnam, in He received the degree of high level technician in physics and chemistry in 1989 from Association Nationale Pour la Formation Professionnelle des Adultes (AFPA), Paris, France. He joined Alcatel Research and Innovation in 1990, where he worked for several years on high voltage cables. He then joined the Photonics Networks Unit, Alcatel Research and Innovation, Marcoussis, France. He is currently working on automation and realization of optical test-beds in the WDM Transmission Systems Group. Peter J. Winzer (SM 05) received the M.S. and Ph.D. degrees in electrical/communications engineering in 1996 and 1998, respectively, from the Vienna University of Technology, Vienna, Austria. His academic work, largely supported by the European Space Agency (ESA), was related to the analysis and modeling of space-borne Doppler wind lidar and highly sensitive free-space optical communication systems. In this context, he specialized on advanced digital optical modulation formats and highsensitivity optical receivers using coherent and direct detection. In November 2000, he joined Bell Labs, Holmdel, NJ, where he focused on various aspects of high-bandwidth fiber-optic communication systems, including Raman amplification, optical modulation formats, advanced optical receiver concepts, and digital signal processing techniques for 10, 40, and 100 Gb/s, as well as on dynamic data network architectures. He has widely published in international journals and at international conferences, has authored several book chapters, and holds patents in the fields of optical communications, lidar, and data networking. Dr. Winzer is a member of the OSA and is actively involved in various activities within the IEEE and the OSA. Alan H. Gnauck (SM 00) received the Masters degree in electrical engineering from Rutgers University, Piscataway, NJ. He joined Bell Laboratories, Holmdel, NJ, in 1982, where he is currently a Distinguished Member of Technical Staff. He has performed record-breaking optical transmission experiments at single-channel rates of 2 to 160 Gb/s. He has investigated coherent detection, chromatic-dispersion compensation techniques, CATV hybrid fiber-coax architectures, wavelength-division-multiplexing (WDM) systems, and system impacts of fiber nonlinearities. His WDM transmission experiments include the first demonstration of terabit transmission. He is presently involved in the study of WDM systems with single-channel rates of 40 Gb/s and higher, using various modulation formats. He has authored or coauthored more than 150 journal and conference papers, and holds 23 patents in optical communications. Mr. Gnauck is a Fellow of the Optical Society of America (OSA) and an Associate Editor for the IEEE PHOTONICS TECHNOLOGY LETTERS. Christopher R. Doerr (F 06) received the B.S. degree in aeronautical engineering and the B.S., M.S., and Ph.D. degrees in electrical engineering from the Massachusetts Institute of Technology (MIT), Cambridge. He attended MIT on an Air Force scholarship and earned pilot wings in Since coming to Bell Labs, Holmdel, NJ, in 1995, his research has focused on integrated devices for optical communication. He was promoted to Distinguished Member of Technical Staff in Dr. Doerr received the OSA Engineering Excellence Award in He is Editor-in-Chief of the IEEE PHOTONICS TECHNOLOGY LETTERS. J. C. Centanni, photograph and biography not available at the time of publication.

6 84 JOURNAL OF LIGHTWAVE TECHNOLOGY, VOL. 26, NO. 1, JANUARY 1, 2008 E. C. Burrows, photograph and biography not available at the time of publication. Tetsuya Kawanishi received the B.E., M.E., and Ph.D. degrees in electronics from Kyoto University, Kyoto, Japan, in 1992, 1994, and 1997, respectively. From 1994 to 1995, he worked for the Production Engineering Laboratory of Matsushita Electric Industrial (Panasonic) Co., Ltd. In 1997, he was with Venture Business Laboratory of Kyoto University, where he was engaged in research on electromagnetic scattering and on near-field optics. He joined the Communications Research Laboratory, Ministry of Posts and Telecommunications (from April 1, 2004, National Institute of Information and Communications Technology), Koganei, Tokyo, in In 2004, he was a Visiting Scholar with the Department of Electrical and Computer Engineering, University of California at San Diego. He is now a Research Manager with the National Institute of Information and Communications Technology, and is currently working on high-speed optical modulators and on RF photonics. Dr. Kawanishi received the URSI Young Scientists Award in He is a Senior Member of the IEEE Lasers and Electro-Optics Society (LEOS). Takahide Sakamoto was born in Hyogo Prefecture, Japan, on May 23, He received the B.S., M.S., and Ph.D. degrees in electronic engineering from the University of Tokyo, Tokyo, Japan, in 1998, 2000, and 2003, respectively. In 2003, he joined the Communications Research Laboratory (now National Institute of Information and Communications Technology), Tokyo. He has been engaged in all-optical signal processing based on nonlinear optics. His current interest is in electrooptic devices, such as LiNbO3 modulators, and their applications to photonic communication systems. Dr. Sakamoto is a member of the IEEE Lasers and Electro-Optics Society (LEOS) and the Institute of Electronics, Information and Communication Engineering (IEICE) of Japan. Kaoru Higuma received the B.E. and M.E. degrees in physics from Waseda University, Tokyo, Japan, in 1994 and 1996, respectively. In 1996, he joined Optoelectronics Research Division, New Technology Research Laboratories, Sumitomo Osaka Cement Co., Ltd., Chiba, Japan. He has been engaged in the research and development of LN optical modulators.