Search

Orthogonal frequency division multiplexing (OFDM) can achieve high utilization efficiency. In this paper, the authors combine OFDM with wavelength division multiplexing (WDM) to investigate the optical fiber transmission characteristics of a WDM/OFDM hybrid system that can easily achieve high utilization efficiency even when the number of channels is high. They used intensity modulation (IM), differential binary phase-shift keying (DBPSK), and differential quadrature phase-shift keying (DQPSK) as modulation methods to show transmission characteristics related to multi-repeater and nonrepeater systems. © 2007 Wiley Periodicals, Inc. Electron Comm Jpn Pt 1, 90(10): 14– 24, 2007; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecja.20387

This paper proposes a method which provides fast and precise evaluation of waveform degradation due to cross-phase modulation (XPM) in intensity modulation-direct detection systems with wavelength-division-multiplexing. The method consists of two corrections applied to the conventional method. The first is to take account of second-order self-phase modulation (SPM), which is induced by the waveform affected by XPM. The second is to take account of the higher-order Kerr effect. The proposed method requires only a single evaluation of an infinite integral, which can be performed in a shorter time than the SSF method. It is shown by comparison to the conventional method that very precise evaluation can be realized despite the simplicity. © 2006 Wiley Periodicals, Inc. Electron Comm Jpn Pt 1, 90(1): 58–67, 2007; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecja.20306

In wavelength division multiplex (WDM) systems for long distance and large capacity, it is desirable that the modulation format be tolerant to degradation by nonlinear effects and achieve high frequency utilization efficiency. This paper compares modulation formats by numerical calculation from the viewpoints of frequency utilization efficiency and tolerance for fiber nonlinearities. It is shown that duobinary, RZ-DQPSK (0.5) and CSRZ-DQPSK achieve high transmission performance in DWDM (dense WDM). It is also shown that NRZ-DBPSK, RZ-DBPSK (0.5), and CSRZ-DQPSK achieve high transmission performance in CWDM (coarse WDM). Before concluding, various methods of bit-error rate calculation are compared and discussed, aiming at the correct evaluation of the bit-error rate. The application of the Q method to the differential quadrature phase-shift keying format is shown to be feasible. © 2006 Wiley Periodicals, Inc. Electron Comm Jpn Pt 1, 89(8): 50–64, 2006; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecja.20236

The Q method and the modified Q method considering the waveform distortion have been proposed as methods of estimating the transmission performance based on the measured bit error rate (BER) in an optical fiber communication system. In these estimation methods, it is assumed that the noise after detection follows a Gaussian distribution. Since square-law detection in the optical receiver is not considered, the estimation error is increased. Consequently, this paper proposes a method for estimation of the system parameters, considering the noise after optical square-law detection and waveform distortion in the received signal. The system parameters for determining the BER are estimated in two steps. The transmission performance is estimated from the system parameters determined by the proposed estimation method, and its accuracy is examined. It is shown that the proposed method can estimate the transmission performance more accurately than the conventional estimation methods. © 2005 Wiley Periodicals, Inc. Electron Comm Jpn Pt 1, 89(1): 54–66, 2006; Published online in Wiley InterScience (www.interscience. wiley.com). DOI 10.1002/ecja.20230

The influence of the word length (WL) of a pseudo-random bit sequence (PRBS) on nonlinear crosstalks is examined. It is shown that the crosstalk due to stimulated Raman scattering (SRS) greatly depends on the WL, while that due to a self-phase modulation and cross-phase modulation-group velocity dispersion (SPM/XPM-GVD) effect has little dependency on the WL. As for the evaluation of SRS crosstalk, we also derive the WL required to obtain a result that is independent of the WL. In the evaluation of nonlinear crosstalks for systems that are mainly degraded by SRS, a sufficiently long WL must be adopted.

Power impairments due to stimulated Raman scattering (SRS) in dispersion-managed (DM) fiber links are evaluated theoretically. We extend previous work on the statistical analysis of SRS crosstalk to the case of multiple fiber segments. Closed-form formulas are derived, and the applicable range is presented by comparison with simulation results. The SRS crosstalk in DM fiber links is evaluated using derived formulas, and the preferable configuration of DM fiber links for suppressing SRS crosstalk is discussed. We also evaluate the exact power penalty induced by the SRS crosstalk in consideration of the log-normal waveform distribution due to SRS and non-Gaussian noise and clarify the system bounds in some typical DM fiber links. The developed approach provides a design rule for DM fiber links from the viewpoint of SRS crosstalk suppression.

Power impairments due to stimulated Raman scattering (SRS) are evaluated in wideband wavelength division multiplexing (WDM) systems. We extend previous work on statistical analysis of SRS crosstalk so that it takes not only pump power depletion but also Stokes power amplification into account. We derive simple closed-form formulae that can evaluate SRS crosstalk of all WDM channels accurately and can specify the worst channel from among all WDM channels. The validity of our approach is confirmed by comparison to the split-step Fourier method. The developed approach can evaluate the power impairments due to SRS quite accurately with very little calculation, and can be applied to almost all WDM systems.

We present an accurate method for the Q-factor estimation even in the presence of waveform distortions in optically amplified systems. The method stands on the assumption that the dominant noises over the bit error rate (BER) are those superimposed on the nearest rails to the decision level. It is clarified that the assumption is usually satisfied. The results also show that, in numerical system evaluation, we should evaluate not only the eye opening but also the probability of occurrence of the rail level.

The influence of stimulated Raman scattering (SRS) taking into account the random modulation and the walk-off effect on wide-band wavelength division multiplexing (WDM) transmission systems are studied theoretically. First, it is shown that power depletion due to SRS can be separated into average power loss and waveform distortion. The waveform distortion is evaluated for various types of pulse shapes and fibers, and simple equations for evaluating the waveform distortion due to SRS are derived. These equations can be easily applied for designing wide-band WDM transmission systems from the viewpoint of the SRS waveform distortion. We also compare results obtained from our method with those from the split-step Fourier method, and confirm validity of our method. Our method should be useful in the design of wide-band WDM transmission systems, in which SRS is expected to be a serious limitation.

This paper reports 1580-nm band wavelength division multiplexed (WDM) transmission employing optical duobinary coding over dispersion-shifted fibers. By using the 1580 nm band, the generation of four-wave mixing (FWM) over dispersion-shifted fibers (DSFs) can he suppressed. Optical duobinary coding is dispersion-tolerant because of its narrow bandwidth, and enables the use of the conventional binary intensity modulated direct detection (IM-DD) receiver. First, comparisons are made for WDM transmission performance in the 1580-nm band between conventional binary nonreturn-to-zero (NRZ) coding with and without postdispersion compensation, and optical duobinary coding by computer simulation is described. From the numerical simulations, it is found that the optical duobinary coding has superior transmission performance to the conventional binary coding without any dispersion compensation, and that the difference in the transmission performance between two coding methods is very small even if postdispersion compensation at the optical receiver is applied to the NRZ coding method. Second, transmission performance between the conventional binary NRZ and the optical duobinary signals without any dispersion compensation is compared with the straight-line experiment over 500-km dispersion-shifted fiber. The experimental results reveal that the transmission distance with optical duobinary coding is doubled in comparison with that of the conventional binary NRZ signals. Finally, 16-channel, 10-Gb/s optical duobinary WDM signals in the 1580-nm band are successfully transmitted over 640 km (80 km/spl times/8) of DSF without any dispersion compensation or management.

This paper investigates the effect of channel-to-channel interference in optical frequency division multiplexing (FDM)-homodyne detection systems that use optical filters for densely-spaced optical FDM systems. Power penalties at BER = 10−9 for double-sideband (DSB), single-sideband (SSB), and vestigial-sideband (VSB) signals were evaluated by approximating the channel-to-channel interference as Gaussian noise. the authors investigated four cases: where no optical filter was used; where an optical channel selection filter was used at the receiver, where an optical filter suppressing spectrum broadening was used for each channel before multiplexing at the transmitter; and where both of these filters were used. It has been shown that using an optical filter before multiplexing is effective for reducing channel spacing. the optical filterings are effective especially for SSB and VSB signals. Extremely densely-spaced multiplexing with a channel spacing of less than twice the bit rate can be achieved with a power penalty of less than 0.1 dB.

An optical 90/spl deg/-hybrid coherent receiver module is constructed that uses the planar lightwave circuit technique. The module consists of an optical 90/spl deg/-hybrid silica planar waveguide and two 14 GHz balanced O/E converters; the solder-bump technique is employed. For both sets of paths between the optical coupler and the twin-p-i-n photodetector pairs, the effective propagation time differences were reduced to within 4 ps. A 10 Gb/s BPSK homodyne detection experiment confirms the feasibility of the receiver. >

Even if the amplitude of a phase-modulated optical signal is constant before transmission, amplitude modulation is caused by fiber chromatic dispersion. As a result, self-phase modulation (SPM) is induced. In optical heterodyne detection, SPM cannot be compensated for by the delay equalizer (electrical domain) used to compensate fiber chromatic dispersion. However, the transmission distance limitation of multi-repeatered coherent transmission systems has not been investigated in the presence of SPM. This paper theoretically and experimentally investigates the transmission distance achievable with a phase-shift-keying (PSK) heterodyne detection system employing in-line optical amplifiers and delay equalization. The calculated results show that equalization is effective when /spl gamma/P/sub 0//2B/sup 2//spl verbar//spl beta//sub 2//spl verbar/ >

Spectral linewidth requirements for optical phase-shift-keying (PSK) coherent detection systems are found to depend on the phase-locked loop (PLL) parameters. Until now, the damping factor of the PLL has been assumed to be 1/ square root 2 when deriving the required spectral linewidth of a light source, because it is at this value that an electrical PLL offers near optimum performance in many cases. By increasing the PLL damping factor above 1/ square root 2, it is shown that there exists a maximum value of the required linewidth that achieves a received optical power penalty of 1 dB at a bit error rate of 10/sup -10/. The required beat linewidths so obtained are 50% larger than previously reported results (which assume a damping factor of 1/ square root 2). As for PLL frequency acquisition performance, it is shown that raising the camping factor above 1/ square root 2 does not seriously affect the hold-in limit or the pull-in limit. It is also shown that the normalized loop gain that optimizes PLL performance is roughly one half the normalized loop gain at which the PLL oscillation commences. >

A combination of optical frequency division multiplexing (FDM) and phase-shift-keying (PSK) homodyne detection can increase transmission capacity. However, phase sensitive transmission systems, especially repeatered ones, suffer from data-dependent optical amplitude fluctuation that is converted to phase fluctuation by fiber nonlinearity. The authors discuss how this data-dependent amplitude fluctuation affects the error rate performance of optical FDM PSK homodyne detection systems. If only the optical amplitude fluctuation induced by phase modulators is taken into account, the allowable power fluctuation to keep the power penalty at 0.5 dB at a bit error rate (BER) of 10/sup -10/ is below 0.17 mW for BPSK homodyne detection and 0.09 mW for QPSK homodyne detection. However, if only the amplitude fluctuation induced by the fiber chromatic dispersion is taken into account, the allowable number of repeaters to keep a 0.5-dB power penalty due to XPM at a BER of 10/sup -10/ is 1 for BPSK homodyne detection and below 5 for QPSK homodyne detection. >

The spectral spreads of quadrature-phase-shift-keying (QPSK) modulation and minimum shift keying (MSK) modulation are narrower than those of other kinds of modulation with the same bit rate. Therefore, they are attractive for systems limited by electrical bandwidth and available optical frequency. The synchronous demodulation offers better sensitivity than differential demodulation. However, the propagation delay time of the phase locked loop affects the performance of synchronous detection systems. Therefore, linewidth requirements while considering the loop delay time must be considered in designing these systems. The linewidth requirements for BPSK, QPSK, and MSK homodyne/heterodyne detection systems are obtained by applying the Pade approximation in evaluating the receivers' performance. A change in the power penalty with respect to the change in these values is also obtained. >

Phase-locked loop (PLL) propagation delay-time influence on optical homodyne detection was investigated both theoretically and experimentally. Applying the Pade approximation, which is often used in the control system, to the calculation of the phase-error variance with the nonzero loop delay time, a high-accuracy analytic expression phase-error variance is obtained. The linewidth requirement with the nonnegligible loop delay time for phase-shift-keying (PSK) homodyne detection is obtained as delta nu =2.04*10/sup -3// tau where delta nu (hertz) is beat linewidth and tau (seconds) is the loop delay time. The linewidth requirement with small delay time approaches delta nu =6.2*10/sup -4/ R/sub b/ where R/sub b/ (bits-per-second) is the system bit rate. Results were confirmed by a 10-GB/s optical PSK homodyne detection experiment using external cavity laser diodes. Receiver sensitivity degradations due to loop delay time and beat linewidth are in good agreement with theoretical results. >

An experimental verification of a 3-D eddy current analysis code based on the T-method was performed by applying it to an eddy-current testing (ECT) problem. Specifically, signal trajectories for ECT of a block with a flaw were measured. The experiment was performed for the ECT benchmark problem defined by the TEAM Workshops for two frequencies. Numerical results for 500 Hz and 1 kHz agreed well with measured values, demonstrating the validity of the method and the code for 3-D eddy-current problems. >

This letter demonstrates fiber chromatic dispersion compensation for optical homodyne detection systems, using a VSB filtered optical PSK signal. An optical PSK signal suitable for SSB and VSB filtering is newly employed. A 10-Gb/s optical PSK-VSB signal is transmitted with an optical carrier through 126 km conventional single-mode fiber. Penalty-free transmission is achieved by fiber chromatic dispersion compensation in the baseband with a 10-cm microstrip line. >

This Letter proposes a novel optical duobinary transmission system with no receiver sensitivity degradation. The transmitter yields a narrowband optical signal, and the receiver configuration is as simple as a binary IM-DD receiver. The feasibility of the proposed system is experimentally confirmed at 2.5, 5 and 10 Gbit/s.

Solitary DFB laser diodes with 174 kHz beat linewidth are applied to a 10 Gbit/s optical BPSK homodyne detection system for the first time: a sensitivity of -38.9 dBm was achieved. A three-electrode laser forms the local oscillator and is modulated with a partial push-pull scheme to improve its frequency response. >

Previous work on the statistical analysis of power impairments due to stimulated Raman scattering (SRS) was extended to fiber Raman amplifier systems and examples of SRS crosstalk in some typical dispersion-managed fiber links are shown.

Summary form only given. This paper demonstrates 8/spl times/10 Gbit/s WDM transmission over 500 km (100 km/spl times/5) of DSF in 1580-nm band without any dispersion compensation or management by employing optical duobinary coding, which reduces the signal bandwidth, and increases resistance to the fiber chromatic dispersion effect.

This case study illustrates the application of a software development process that is intended to produce low-defect code for reliable systems. Recognizing that traditional testing is conducted too late in development to ensure the quality of mission critical software, the process emphasizes verification through applying formal methods, inspections in all phases, and a unique testing philosophy. This process is demonstrated in a well instrumented candidate project that is the subject of the case study. The product is control software for a robotic data tape system used in the B-2 test program. Only three defects were found in the unit and integration test phases of the project, despite being over 7,000 lines of code in size. Since deployment no major defects have been found in more than eight months of operational use.

A new technique has been developed for packaging an optical 90/spl deg/-hybrid balanced receiver using a planar lightwave circuit (PLC) for the application to an optical homodyne detection system. This receiver consists of a silica-based PLC as an optical coupler, two polarization beam splitters (PBSs), two GRIN rod lenses to achieve perfect optical interconnections, two photoreceivers that each have a preamplifier, and twin p-i-n photodiodes (PIN-PDs). The photoreceivers are fabricated using microsolder bump bonding to eliminate parasitic elements in the interconnections between the preamplifiers and the PIN-PDs. An impedance-matched film carrier technique is used to achieve impedance-matched interconnections between the preamplifiers and the ceramic packages. Experiments showed that a receiver fabricated using this packaging technique has excellent performance: a broad-band frequency response of 14 GHz and, consequently, 10-Gb/s binary phase shift keying (BPSK) homodyne detection.

A quadrature-phase-shift-keying optical homodyne detection experiment was conducted at a bit rate of 8 Gb/s. External-cavity laser diodes and a single-body QPSK phase modulator were used. The optical carrier was recovered by a decision-directed phase-locked loop. The first reported measurement of bit-error-rate performance is given. A receiver sensitivity of -27.8 dBm was achieved. >

With advance in silicon technology and system integrating technology, utmost care for RC-delay must be taken in high performance LSI design. This paper describes the new design method applied to the Vector Unit (VU) implemented in the recently announced high-performance 3D-graphics engine, 'Emotion Engine'. The key features of the design method are the following: careful functional design of the VU architecture; design hierarchy consistency between RTL descriptions and floor plan; accurate estimation of wire load in pre-layout static timing analysis. With these design features, the VU is has achieved the operating frequency of 300 MHz.

The residual amplitude modulation of adjacent channels causes cross-phase modulation which is responsible for degradation in the error rate performance of a phase detection system. This influence on a two-channel 10 Gb/s optical phase-shift-keying (PSK) homodyne transmission system was experimentally evaluated. A penalty was observed for more than 6 dBm fiber input power with 28% residual amplitude modulation passing through a 100 km dispersion shifted-fiber. Since cross-phase modulation does not require phase matching, which is required for four-wave mixing, the optical frequency range in which the influence occurs is broadband. These results show that residual amplitude modulation in synchronous optical frequency division multiplexing transmission systems severely limits the fiber input power. >

5-Gb/s optical PSK (phase-shift keying) homodyne detection experiments are discussed. In these experiments, the optical carrier is recovered by a Costas optical phase-locked loop using a multielectrode local oscillator (DFB) laser diode at 1.55 mu m with a flat FM response. Although the beat linewidth of 80 kHz is broad compared to the loops in other phase-locked loop (PLL) experiments, phase locking with Costas loop is confirmed at 5 Gb/s by increasing the loop natural frequency. The receiver sensitivity is -42.2 dBm or 93 photon/bit for a 2/sup 7/-1 pseudorandom bit sequence (PRBS) in front of a 90 degrees hydride. >

This paper describes the evolution of future network which uses multi-wavelength as several network functions together with implication of technology enhancement, key issues to increase the number of wavelengths, and new 1580 nm band technology to extend transmission bandwidth.

Quandrature-phase-shift keying (QPSK) homodyne detection system can double the system bit rate without increasing the electric circuits bandwidth. Moreover, we can get a more compact spectrum at the same system bit rate than is possible with binary modulation. However, QPSK optical homodyne receiver has a 3-dB worse sensitivity than a binary-phase-shift keying receiver,1 and requires very narrow linewidth lasers or short loop delay times.2 The only reported QPSK optical homodyne detection experiment used two gas lasers,3 whose linewidths were very narrow. However, the use of solitary laser diodes as light sources is expected in future systems because of their reliability and compactness. This paper reports what we believe to be the first demonstration of QPSK homodyne detection with external cavity laser diodes and measured bit error rate (BER) performance.

Optical phase-shift-keying (PSK) homodyne detection offers the best sensitivity of any binary signaling technique and requires only the same electrical bandwidth as the bit rate. Therefore, it is suitable for multigigabit coherent transmission systems.1,2 However, it requires lasers with extremely narrow linewidths. Laser linewidth requirements depend on the configuration of the phase-locked loop (PLL). A decision- driven PLL can tolerate a larger laser linewidth (e.g., 3.1 × 10−4 times the bit rate 3) than can a pilot-carrier optical PLL, which requires a linewidth of 6 × 10−6 times the bit rate.4 Moreover, the requirement is restricted by the PLL propagation time.5 PLL propagation time is important for high-speed homodyne detection systems using broad-linewidth laser diodes. This paper reports the linewidth requirements for 10-Gb/s PSK homodyne detection in a decision-driven PLL, taking into consideration the loop propagation delay.

10 Gbit/s optical PSK homodyne transmission experiments using a modified decision-driven optical phase locked loop with 1.55 μm external cavity laser diodes were conducted. A receiver sensitivity of −34.2 dBm or 297 photon/bit was experimentally achieved. There was no degradation in receiver sensitivity after signal transmission through 151 km of dispersion shifted fibre.

16-channel, 10 Gbit/s optical duobinary WDM signals in the 1580 nm band are successfully transmitted over 640 km (80 km /spl times/8) of dispersion-shifted fibre with no dispersion compensation or management.

There is some confusion about what the effective potential is. To remove this confusion, boundary conditions in the path-integral formulation are discussed. For a system with several phases, there are several kinds of physically acceptable boundary conditions. The choice of condition determines the phase. Therefore without specifying the phase we cannot discuss the generating functionals. By taking this fact into account, the meanings of the effective potential and its convex property are clarified. The confusion is removed and the consistent interpretation is obtained.

A numerical analysis technique for 3-D eddy-current analysis using the T-method is described. The technique is based on an iterative solution method using two matrices, which are generated by splitting a complex dense matrix that appears in the AC problem. Numerical results for a simple problem show the validity of an over-relaxation method in the iteration procedure. An application to eddy-current testing verified the technique's applicability to practical problems. >