John C. Cartledge

The effect of laser chirping on lightwave system performance

Directly modulated semiconductor lasers exhibit a dynamic wavelength shift (chirping) arising from gain-induced variations of the laser refractive index. The effect of laser chirping on the performance of multi-Gb/s lightwave systems operating at a wavelength of 1550 nm is investigated. Models suitable for computer-aided analysis are used to describe the dynamic response of the laser and the propagation of chirped optical pulses through a step-index single-mode optical fibre. A truncated pulse train, Gauss quadrature rule method is used to evaluate the average bit error rate of the receiver. This permits pattern effects in the transmitted optical waveform due to the laser dynamics and nonlinear optical power transmission properties of optical fibers to be included in the system model. The influence that modulation and device parameters have on the receiver sensitivity is assessed. >

Linewidth-Tolerant and Low-Complexity Two-Stage Carrier Phase Estimation for Dual-Polarization 16-QAM Coherent Optical Fiber Communications

Two novel linewidth-tolerant, low-complexity feedforward carrier phase estimation algorithms are described for dual-polarization 16-ary quadrature-amplitude-modulation with coherent detection. For both algorithms, the carrier phase is estimated in two stages. The first stage employs either a simplified quadrature-phase-shift-keying (QPSK) partitioning algorithm or the blind phase search (BPS) algorithm. The second stage employs a novel QPSK constellation transformation algorithm. The performance and linewidth tolerance of both algorithms are evaluated using experimental and simulation data, and the hardware complexity is assessed. For both proposed two-stage algorithms, the linewidth symbol duration product is 1.3 × 10-4 for a 1 dB penalty in optical signal-to-noise ratio at a bit error ratio of 10-3. This performance is comparable to a single-stage BPS algorithm with a large number of test phases, but with a reduction of the hardware complexity by factors of about 2.5-11.

Performance of smart lightwave receivers with linear equalization

Signal processing techniques can be used to reduce linear and nonlinear distortion in high-speed lightwave systems caused by fiber dispersion and nonideal responses of optoelectronic and electronic components. The improvement in the performance of 2.5 and 10 Gb/s intensity modulation, direct detection systems is assessed for receivers which utilize an analog taped delay line equalizer to compensate for signal distortion. Synchronous and fractionally spaced equalizers are evaluated. Smart receivers that jointly optimize the decision time, decision threshold, and equalizer tap weights under a minimum bit error ration criterion are considered. This yields the optimum system performance and allows consideration of both reduced distortion and enhanced noise arising from the signal processing. The effectiveness of the equalization is determined as a function of several important system parameters. Three-tap and five-tap synchronous equalizers yield virtually the same improvement in receiver sensitivity. Depending on the system, a five-tap fractionally spaced equalizer with half-bit-period tap spacing may or may not be significantly more effective than a three-tap synchronous equalizer. >

Theoretical performance of 10 Gb/s lightwave systems using a III-V semiconductor Mach-Zehnder modulator

The system performance is assessed for 10 Gb/s lightwave systems that use a III-V semiconductor Mach-Zehnder intensity modulator and operate at a wavelength of 1.55 /spl mu/m with standard single-mode optical fiber. The modulator is modeled by an equivalent circuit and an experimentally derived functional dependence of the attenuation and phase constants on applied voltage. The influence of the format of the modulating signal and the splitting ratio of the input Y-branch waveguide on the chirp characteristics of the modulator is examined based upon calculation of the receiver sensitivity. Optimum system performance is obtained for an asymmetric splitting ratio.<<ETX>>

Reducing the complexity of perturbation based nonlinearity pre-compensation using symmetric EDC and pulse shaping

Perturbation based nonlinearity pre-compensation has been performed for a 128 Gbit/s single-carrier dual-polarization 16-ary quadrature-amplitude-modulation (DP 16-QAM) signal. Without any performance degradation, a complexity reduction factor of 6.8 has been demonstrated for a transmission distance of 3600 km by combining symmetric electronic dispersion compensation and root-raised-cosine pulse shaping with a roll-off factor of 0.1. Transmission over 4200 km of standard single-mode fiber with EDFA amplification was achieved for the 128 Gbit/s DP 16-QAM signals with a forward error correction (FEC) threshold of 2 × 10(-2).

Linewidth-Tolerant and Low-Complexity Two-Stage Carrier Phase Estimation Based on Modified QPSK Partitioning for Dual-Polarization 16-QAM Systems

Three novel linewidth-tolerant, low-complexity, two-stage feed-forward carrier phase estimation algorithms are introduced for dual-polarization 16-ary quadrature amplitude modulation (DP 16-QAM) with coherent detection. The first stage employs either the quadrature phase-shift keying (QPSK) partitioning algorithm, simplified QPSK partitioning algorithm, or blind phase search (BPS) algorithm. The second stage employs a novel modified QPSK partitioning algorithm. Based on experimental data, all three algorithms achieve comparable performance for DP 16-QAM back to back and transmission systems. The linewidth tolerance for the three algorithms is numerically studied. A linewidth symbol duration product of 1.3×10-4 is demonstrated for a 1 dB optical signal-to-noise-ratio penalty at a bit error ratio 10-3 of for all the proposed algorithms, which is comparable to the single-stage BPS algorithm with a large number of test phases. Reductions in the hardware complexity by factors of about 1.7-5.3 are achieved in comparison to the single-stage BPS algorithm.

Optimum operating points for electroabsorption modulators in 10 Gb/s transmission systems using nondispersion shifted fiber

The chirp, optical extinction ratio and insertion loss of an electroabsorption modulator (EAM) depend on the properties of the bulk or multiple-quantum-well absorption layer of the device and on the bias voltage and modulating voltage waveform. For 10 Gb/s transmission over nondispersion shifted fiber, joint optimization of the bias and modulation voltages is considered for five different EAMs. To comprehensively explore this issue, the measured dependence of the absorption and /spl alpha/-parameter on applied voltage is used to accurately model an EAM in a system simulator. The effects of group velocity dispersion and self-phase modulation arising from the Kerr nonlinearity are included to permit assessment of the dependence of the optimum bias and modulation voltages on the average transmitted optical power for a given fiber length. The improvement in receiver sensitivity, relative to that obtained with maximum optical extinction ratio, depends quite significantly on the transmitted optical power and the specific properties of the modulator. This makes it difficult to determine optimum operating conditions which apply generally.

Modulation characteristics of semiconductor Mach-Zehnder optical modulators

The transmission and chirp characteristics are described for two types of semiconductor Mach-Zehnder modulators, distinguished by the differential phase shift between the two arms of the interferometer in the unbiased state. The conventional modulator has a differential phase shift of 0 radians, while the /spl pi/-shift modulator has a differential phase shift of /spl pi/ radians. The nonlinear dependence on the applied voltage of the attenuation and phase constants of the optical signal propagating in the p-i-n waveguide leads to different characteristics for the two modulators. The influence of the splitting ratio of the Y-junctions is considered for single-arm and dual-arm (push-pull) modulation formats. The /spl pi/-shift modulator is shown to yield better transmission performance for 10 Gb/s systems compared to the conventional modulator.

On using fiber transfer functions to characterize laser chirp and fiber dispersion

Two measurement techniques for estimating the /spl alpha/ parameter of a laser, the frequency f/sub c/ for which the adiabatic and transient chirp of a laser have the same magnitude, and the dispersion coefficient D of an optical fiber are compared. The techniques rely on either a single measurement of the transfer function of a dispersive optical fiber using direct modulation of the laser, or two measurements of the fiber transfer function using external and direct modulation of the laser. Compared to the two measurement technique, the single measurement technique yields a smaller estimate of /spl alpha/ and a nearly identical estimate of f/sub c/, and can produce an unreliable estimate of D unless the frequency span in the parameter extraction procedure is sufficiently large.<<ETX>>

Effect of chirping-induced waveform distortion on the performance of direct detection receivers using traveling-wave semiconductor optical preamplifiers

The influence of chirping-induced waveform distortion on the performance of multigigabit-per-second traveling-wave semiconductor optical amplifier (TWSOA)/p-i-n direct detection receivers is evaluated. The results are based on a novel method of evaluating the probability of error in the presence of the signal-spontaneous and spontaneous-spontaneous beat noise components. Laser chirping causes the dependence of the receiver sensitivity on the fiber dispersion coefficient*length product DL to be different for TWSOA/p-i-n and avalanche photodiode (APD) receivers. Compared to the APD receiver, the sensitivity of the TWSOA/p-i-n receiver degrades less quickly. So for cases of practical interest, the TWSOA/p-i-n receiver is more tolerant of chirping-induced waveform distortion. >