Long-haul Communication Systems Research Articles

Noise Analysis of InP/InGaAs Superlattice Avalanche Photodiode

A theoretical study of the noise characteristics of an InP/In0.53 Ga0.47 As Superlattice Avalanche photodiode (SL-APD) has been presented. It has been found that an improvement in the value of the effective hole-to-electron ionisation rate ratio (βeff/αeff) for the SL-APD structure results in a low-noise behaviour of the device. The device has a very high gain-bandwidth product (≈450 GHz) and a low excess noise factor at moderate gain. It is expected to find application as a highly sensitive low-noise photodetector in optical receiver unit for long-haul fiber optic communication system in 1.3 to 1.6 μm.

Technology of InP-based 1.55-μm ultrafast OEMMICs: 40-Gbit/s broad-band and 38/60-GHz narrow-band photoreceivers

For future long-haul communication systems operating at bitrates of 40 Gbit/s and for broad-band mobile access systems using 38- or 60-GHz carrier frequencies, ultrafast photoreceivers have to be provided. Therefore, an integration concept for InP-based optoelectronic microwave monolithic integrated circuits for the 1.55-/spl mu/m wavelength regime is demonstrated, which allows independent optimization of the constituting devices. Two different types of photodetectors (PDs), a waveguide-integrated PIN photodiode (PD) and a top-illuminated metal-semiconductor-metal PD, both having bandwidths of up to 70 GHz, have been developed. These are fabricated together with different amplifier designs employing high electron mobility transistors which exhibit transit frequencies of up to 90 GHz. The application to a 40-Gbit/s broadband photoreceiver for high-bit-rate time-division multiplexing systems is reported, as well as the application to 38- and 60-GHz narrow-band photoreceivers for use as optic/millimeterwave converters in mobile communication systems.

Noise performance of an InP/InGaAs superlattice avalanche photodiode

The noise performance of an InP/InGaAs Superlattice Avalanche Photodiode (SL-APD) has been studied theoretically to examine its suitability as a detector in optical receiver units. The results indicate that the device which has a very large gain-bandwidth product (≈450 GHz) provides a high receiver sensitivity compared to conventional APD. The device also exhibits a very low value of excess noise factor which results in a high value of signal-to-noise ratio. The SL-APD is expected to find useful applications as a high sensitivity photodetector in long-haul fibre optic communication systems. For a digital system, the device exhibits a very low value of bit-error rate (BER) even at moderate gain.

Nonlinear pulse propagation in dispersion managed system

We study nonlinear pulse propagation in an optical transmission system with dispersion managements. This is particularly important for designing an ultra-fast and long-haul communication system in the next generation. There exists a quasi-stationary pulse solution in such a system whose width and chirp are rapidly oscillating with the period of dispersion management. This pulse also has several new features such as enhanced power when compared with the soliton case with a uniform dispersion and reduction of timing jitter in a wavelength-division-multiplexed system. The method used here is so-called the guiding center (or normal form) theory based on the Lie transform and the averaging method. For the case of a small dispersion management, the core equation obtained as the normal form is given by the nonlinear Schrödinger (NLS) equation with a small correction, while for the case with a rather large management, the core equation is shown to be the NLS equation having a nontrapping quadratic potential. Because of this potential, a steady pulse propagating in such a system requires enhanced power.

Volterra series approach for optimizing fiber-optic communications system designs

A new methodology for designing long-haul fiber-optic communication systems is presented. We derive the overall Volterra series transfer function of the system including linear dispersion, fiber nonlinearities, amplified spontaneous emission (ASE) noise from the fiber amplifiers, and the square-law nature of the direct detection (DD) system. Since analytical expressions for the probability of error are difficult to derive for the complex systems being used, we derive analytical expressions for an upper bound on probability of error for integrate-and-threshold detection at the receiver. Using this bound as a performance criterion, we determine the optimal dispersion parameters of each fiber segment required to minimize the effects of linear dispersion, fiber nonlinearities and ASE noise from the amplifiers. We study the dependence of optimal dispersion parameters on the average power levels in the fiber by varying the peak input power levels and the amplifier gains. Analytical expressions give us the freedom to choose system parameters in a practical manner, while providing optimum system performance. Using a simple system as an example, we demonstrate the power of the Volterra series approach to design optimal optical communication systems. The analysis and the design procedure presented in this work can be extended to the design of more complex wavelength division multiplexed (WDM) systems.

Stabilized 20 GHz LiNbO_3 Polarization Scrambler

We developed a polarization scrambler with an embedded polarizer, angled at approximately 45° on the input side of the waveguide. The degree of polarization (DOP) of this scrambler is highly stable with the DOP less than 5% at 1558 ± 2 nm and less than 6% at approximately 10-40 °C. These scramblers are useful for long-haul communication systems and WDM systems.

Directly modulated sampled grating DBR lasers for long-haul WDM communication systems

The limitations on high-speed data transmission, using widely tunable sampled-grating distributed Bragg reflector (SGDBR) lasers, are investigated. We demonstrate 1.244-Gb/s data transmission over a wavelength range of 45 nm using a single directly modulated tunable SGDBR laser diode. Data transmission was evaluated on four separate wavelength channels each spaced 15 nm apart. Less than 0.6 dB of dispersion penalty was measured on all four channels for transmission over 50 km of standard single-mode optical fiber.

Fiber Bragg gratings for dispersion compensation in transmission: theoretical model and design criteria for nearly ideal pulse recompression

We propose a transmission-based dispersion compensator employing an apodized, unchirped fiber Bragg grating (FBG). A theoretical model for dispersion compensation in transmission based on the dispersive properties of the periodic structure is developed. A figure of merit is defined for optimization of the grating parameters for maximum recompression of dispersion-broadened optical pulses in long-haul communication systems. Numerical examples confirm that nearly perfect compensation with very low insertion losses can be achieved for many practical cases of interest.

Design of multiwavelength optical communication systems based on an integrated waveguide amplifier compensator

We theoretically analyze the performance of longhaul multichannel optical communication systems based on an integrated WDM power level compensator. By placing an array of Er/sup 3+/-Yb/sup 3+/ co-doped silica waveguides in parallel between a demultiplexer/multiplexer pair and preceeding it by an EDFA, we demonstrate the possibility of transmitting eight unequally spaced WDM channels over long distances, without electronic regenerators and using span lengths as high as 50 km. This technique allows substantial reduction of both four-wave mixing crosstalk and gain peaking.

Effect of optical amplifier noise on laser linewidth requirements in long haul optical fiber communication systems with Costas PLL receivers

The impact of optical amplifier noise is analyzed in investigating the performance of optical long-haul PSK homodyne communication systems with Costas phase locked loop (PLL) receivers. The laser linewidth requirement for an optically amplified system becomes relaxed in comparison with the system with no optical amplifier, owing to the fact that the effect of incomplete phase tracking becomes less important as a larger signal power is demanded to maintain a fixed bit-error rate. Also, it is found that the power splitting ratio regarding the power distributions for the I-arm and the Q-arm of a Costas loop can vary in a wide range without having much influence on the performance of an optically amplified system. As a matter of fact, the power penalty induced by incomplete phase tracking for a system with a large number of cascaded optical amplifiers is mainly due to the finite phase error and not due to the power splitting ratio, and this may fail a previously-reported method for finding the required laser linewidth by assigning a certain amount of power penalty that is due to the power splitting ratio.

Optimum phase deviations between mark-and space-state bits in long-haul coherent communication systems with balanced PLL receivers

In this paper, we report the importance of the optimum phase deviation between mark-state and space-state bits in a long-haul PSK homodyne system with a balanced phase-locked-loop (PLL) receiver. The laser phase noise, data-to-phase-lock crosstalk, receiver shot noise and amplifier noise are considered in analyzing the systems with cascaded optical amplifier chains. The signal power penalty incurred by improper use of phase deviations in a long-haul system is evaluated. It is found that such power penalty can amount to larger than 1.5 dB in a 10-Gbt/sec system.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Wigner distribution and spectral dynamics of laser diodes

The Wigner distribution is a simultaneous representation of a signal in both time and frequency domains. We discuss the Wigner distribution as an instrument-independent method of displaying the time–frequency response of a laser transmitter. This is important in long-haul communication systems that use semiconductor lasers, which exhibit chirp. If the type of spectrometer for measuring the spectrum as a function of time is known, so that the instrument's impulse response is known, then the Wigner distribution can be recovered. This distribution is then identical to that obtained with a different instrument with a different resolution. Examples of the calculated time-resolved spectra, obtained with two different instruments, are shown for a gain-switched laser. They are compared with the Wigner distribution obtained by deconvolution.

Improved robustness in long-haul lightwave all-amplifier systems

The issue of system robustness of a long-haul lightwave communication system that uses erbium-doped fiber amplifiers to periodically boost the signal power level is addressed. The signal power level at the output of each amplifier is the focus. It is found that this power level is the same throughout the system, and that it depends only on the amplifiers' gain characteristic and the span loss and is independent of the launched power at the input to the chain, provided it is close to the design value. As such, the systemwide signal power level is sensitive to systematic design and/or manufacturing errors. The sensitivity of the signal power level to errors in amplifier design and span loss is studied, and a simple technique is proposed that reduces this sensitivity, thus making the system more robust.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Multigigabit networks: the challenge

Candidate transmitters, optical amplifiers, and detectors that can meet the very demanding constraints for digital multigigabit long-haul terrestrial communication systems through conventional fiber are reviewed. The focus is on optical devices capable of approximately 10-GHz response. Electronic issues are not addressed. The performance of existing and future external modulators, optical fiber amplifiers, and detectors is summarized. Some dispersion compensation schemes are examined, and the use of dispersion-shifted fiber for extending the reach of terrestrial networks is examined.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Channel capacity optimization of chirp-limited dense WDM/WDMA systems using OOK/FSK modulation and optical filters

Frequency chirp in semiconductor lasers that are directly modulated through carrier injection not only induces dispersion in long-haul single-mode fiber communication systems, but also causes energy loss and signal distortion when frequency selective elements such as Fabry-Perot filters are present. This effect is significant for high-bit-rate channels in a dense wavelength-division-multiplexed system in a LAN/MAN environment which requires narrow-band filters to suppress crosstalk from adjacent channels. A simulation approach is used to determine the channel capacity in the presence of chirp and crosstalk. The results show that a minimal channel spacing of less than 37 and 10 GHz with each channel operating at 2 Gb/s using OOK and FSK, respectively, can be achieved by a dense wavelength division multiple access system. It is also shown that this system performance can be optimized through adjusting various parameters of the laser and the filter.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Power limitations due to four-wave mixing effects in frequency division multiplexed coherent systems using cascaded optical amplifiers

The impact of four-wave-mixing (FWM) effects on phase modulated coherent multichannel systems is investigated numerically for long-haul communications systems using optical amplifiers. Assuming ideal amplifiers spaced at 100 km and compensation of fiber dispersion at the receiver end, the performances of a 2.5-Gb/s 15-channel system are compared for conventional and dispersion-shifted fibers. For 1 mW per channel and 12.5-GHz channel separation, FWM leads to a channel averaged power penalty of 3.3 dB (up to 7 dB in particular channels) after 800 km in a conventional fiber, and 8 dB after only 300 km in the dispersion-shifted fiber. As expected, an increase of channel separation from 12.5 to to 37.5 GHz leads to improved system performance for both types of fibers. The improvement is, however, limited by modulation instability occurring in the individual channels so that for very large channel separations a better system performance is achieved using dispersion-shifted fibers.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Long-haul undersea communications systems: hardware redundancy techniques for ultra-reliability : Robert H. Murphy. Proc. A. Reliab. Maintainab. Symp., 164 (1990)

Long-haul undersea communications systems: hardware redundancy techniques for ultra-reliability : Robert H. Murphy. Proc. A. Reliab. Maintainab. Symp., 164 (1990)

Performance of fractional tap-spacing equalizers for long-haul high-capacity digital fibre-optic communication systems

It is known that for data transmission above 100 Mbits s−1 multimode fibre systems, working at 1.3 μm wavelength, are more modal-dispersion-limited than loss-limited. Then some form of baseband equalization is highly desirable. Baseband equalizers are usually made in the form of a transverse filter with variable tap gains and tap spacing equal to the symbol spacingT. However, the performance of these equalizers depends critically on the timing phase. To overcome this difficulty, in this paper we analyse the performances of fractional tap-spacing equalizers that have a tap spacing smaller thanT. Results obtained show that this type of equalizer gives satisfactory performance for a broad continuous range of timing phase values. It allows the application of simple clock recovery schemes in the receiver.

Principles of piezoelectric ceramics for mechanical filters

Piezoelectric ceramics are prepared for fabricating the electromechanical transducers used in the mechanical frequency filters that find application in long-haul communications systems. These ceramics have to satisfy specifications that can only be met by utilizing all the possibilities offered by the physical effects of the ferroelectric material. The material properties are stabilized by modification with Mn. The required positive temperature coefficient of the frequency constant is realized with the aid of elastic anomalies in the region of ferroelectric phase transitions. Complex additives sharply increase permittivity while keeping mechanical losses low. Through the careful coordination of additives for softening and stabilization it is possible to prepare stabilized ceramics with the typically high piezoelectric coupling coefficient of soft ceramics accompanied by a high mechanical quality factor and effective resistance to aging. Moreover, any sharp increase in mechanical losses due to a rise in temperature is largely avoided.

Distortion analysis of bipolar transistor circuits

In the design of long-haul analog communication systems, it is essential to understand the nonlinear distortion behavior of the electronic circuit realization. Distortion analysis of weakly nonlinear circuits has been developed based on the well-known perturbation method. In the frequency domain, the nonlinearities of a transistor are shown to be equivalently represented by intermodulation distortion sources whose amplitudes and phases are iteratively determined by the linear circuit characteristics and the Taylor-series coefficients associated with the nonlinearities. For all practical purposes, only two iterations are sufficient to yield accurate results. An algorithm for computing the second- and third-order intermodulation distortions is described. This algorithm has been implemented in a program called NODAP (nonlinear distortion analysis program). NODAP computes the smallsignal nonlinear transistor model from the recently developed integral change-control model (ICM). It then passes this information through a linear circuit analysis program for distortion computations. Computer results for several transistor circuits using Western Electric highfrequency transistors are all in excellent agreement with the measured data over a wide range of bias conditions and frequencies. This demonstrates the validity of the approach treated in the paper.