Crosstalk Power Research Articles

Reflection-induced penalty in Raman amplified systems

In this letter, we describe the measurement of reflection-induced penalties in Raman amplified systems. We find that the reflection-induced inband crosstalk power ratio (R/sub c/) provides a convenient means for predicting system penalties from crosstalk due to either a pair of discrete reflections or a discrete reflection combined with Rayleigh backscattering, occurring under a variety of system parameters, such as fiber attenuation, Raman gain efficiency, Raman pump power, mode field size, connector reflectance, and fiber length. A 1-dB penalty in our optically preamplified receiver is observed when the reflection-induced inband crosstalk level is -25.2 dB for nonreturn-to-zero (NRZ) signals and -22.8 dB for 50% return-to-zero signals. We also estimate the penalties of multispan systems using R/sub c/ per span. We find, for instance, that a Raman amplified system consisting of 30 spans with an R/sub c/ of -40 dB per span would sustain a 1-dB penalty for NRZ signals.

Study of crosstalk property in an optical packet switching node

The effect of queuing delay of output buffer on the crosstalk property in optical packet switching nodes is investigated. The relationship between crosstalk and buffer length is obtained. From the calculation and simulation results, it is concluded that the crosstalk power penalty predominantly depends on the buffer length, the longer the buffer, the greater the penalty, as well as the random range of the penalty. While comparing with the effect of queuing delay, the effect of random routing path delay takes very little proportion in the total power penalty.

Noise generation, propagation and effects on RF components in system-on-a-chip packaging

Substrate coupling noise effects in wireless receiver systems in terms of the crosstalk power spectral density induced from the fast switching digital circuits is the center of study in this paper. Deterioration in performance of a low noise amplifier is plotted against various values of die-attach inductance, inductance on digital ground pins, physical separation between the analog and digital circuits on-chip, number of simultaneous switching output buffers, etc. Different Ball Grid Array Packages, both the wire bonded and flip-chip attached versions have been studied. The return loss and insertion loss for paths from the on-chip wire bond pad to connect pads on the printed circuit boards have been plotted. Results show that noise reduces by a greater amount for reduction in die-attach inductance as compared to a reduction in inductance on the digital ground pin.

Physics-based CAD models for the analysis of vias in parallel-plate environments

In this paper, physics-based computer-aided-design (CAD) models for through and buried vias in parallel-plate environments are presented based on radial transmission-line theory. The crosstalk power transferred by the TEM parallel-plate mode between vias is characterized, and extended to the treatment of vias in finite substrates by means of image theory. The presented CAD models can be combined with lumped and distributed circuit elements, as well as linear and nonlinear devices, providing an accurate and fast procedure for the global modeling of high-speed electronic circuits. The corresponding simulation time for representative single or multiple via configurations has been drastically reduced compared to full-wave simulations while maintaining comparable accuracy.

Substrate crosstalk suppression capability of silicon-on-insulator substrates with buried ground planes (GPSOI)

Experimental s/sub 21/ transmission crosstalk studies have been conducted on silicon-on-insulator substrates with buried ground planes (GPSOI's) where a 2 /spl Omega/ per square metal-silicide buried ground plane existed between a 15 /spl Omega/-cm p-type silicon substrate and a 1 μm thick buried CVD oxide layer. Locally grounded transmission test structures fabricated on GPSOI were found to exhibit 20 dB increased crosstalk suppression compared to published data for high resistivity (200 /spl Omega/-cm) SOI substrates incorporating capacitive guard rings over a frequency range from 500 MHz to 50 GHz. This represents an order of magnitude improvement in crosstalk power suppression capability compared to existing state-of-the-art suppression techniques in silicon substrates.

Reduction of coherent crosstalk in WDM add/drop multiplexing nodes by bit pattern misalignment

Coherent crosstalk in wavelength-division-multiplexed (WDM) add/drop multiplexers and switches places severe limitations on the acceptable crosstalk suppression level. We propose and analyze a method of reducing the effects of coherent crosstalk in add-drop multiplexers by purposely introducing fiber delays on the order of bit times between different crosstalk paths. These delays decorrelate the bit pattern of the main signal with the interfering signals and place some of the crosstalk power in the 0 bits of the main signal instead of the 1 bits. For a 16-channel WDM system, this method reduces the device requirement for crosstalk suppression by /spl sim/10 dB for return-to-zero and /spl sim/5 dB for nonreturn-to-zero transmission.

Experimental comparison of the effect of discrete and distributed path inband crosstalk on system performance: application to predicting system performance penalties

The system performance degradations of inband crosstalk produced by distributed Rayleigh scattering and a single discrete time delayed path have been measured in the same system. Rayleigh scatter crosstalk degrades performance more than equal amounts of discrete crosstalk. By parameterizing the system bit error rate (BER) and received power, we have separated the optical and electrical noise contributions to the BER based on their intensity and crosstalk scaling. We observed behavior consistent with earlier models, but found an unexpected increase in the total optical noise. This excess noise had a very regular linear scaling with crosstalk power. Because of its well defined crosstalk and intensity scaling, this easily measured noise term can be used as a correction to established models to more accurately estimate system performance at high optical powers and low crosstalk levels.

Expected performance of waveguide grating routers in the presence of random phase and amplitude errors

The expected waveguide grating router response is calculated analytically in the presence of random, uncorrelated phase and amplitude errors. It is shown that the expected crosstalk, as well as the standard deviation of the crosstalk power, is indirectly proportional to the number of grating arms.

Highly stable optical add/drop multiplexer using polarization beam splitters and fiber Bragg gratings

We demonstrate a novel wavelength-division add/drop multiplexer employing fiber Bragg gratings and polarization beam splitters. The multiplexer is easy to fabricate without any special technique such as UV trimming, and yet shows very stable performance with less than 0.3-dB crosstalk power penalty in a 0.8-nm-spaced, 2.5-Gb/s-per-channel wavelength-division multiplexing (WDM) transmission system.

System design and performance of reconfigurable and simultaneous 2D multiple-plane WDM optical interconnects

A system design for reconfigurable and simultaneous 2D multiple-plane WDM optical interconnects has been investigated. Device characteristics such as detector responsivity shape, efficiency, and complexity affect the design requirements for wavelength separation and channel power. With decreasing wavelength separation, increased optical crosstalk and signal power loss result in performance penalties most notably for the middle wavelength channels. When signal power loss and optical crosstalk from the detection of undesired channels are evaluated, it is found that wavelength channel spacing down to 10 nm can be tolerated for the specific detector responsivity shape and efficiency attainable from currently available devices.

Crosstalk caused by fabrication errors in a generalised Mach-Zehnder interferometer

The relationship between crosstalk and fabrication errors in a waveguide grating router is derived. The average crosstalk power from each unwanted channel is simply the ratio of the errors variance to the effective number of arms Neff, hence it can be reduced by increasing Neff. Very low levels of crosstalk are reported.

BER impairment due to laser linewidth in OFDM-OOK systems using double-cavity Fabry-Perot demultiplexers

We report on the impact of finite laser linewidth on the bit error rate (BER) performance in directly detected optical frequency-division multiple-access (OFDMA) employing double-cavity Fabry-Perot (FP) filters. Three different double-cavity FP demultiplexers have been considered in this study. Also the laser linewidth impairment on the crosstalk probability mass functions, the crosstalk power penalties and the OFDMA network capacity are calculated. The results demonstrate that, even though these demultiplexing schemes are much superior than single-cavity FP filters for highly coherent sources (laser linewidth lower than a few megahertz), when laser linewidth increases this feature appears to be much less significant due to the severe impairment of the laser linewidth on the BER performance, specially for structures with higher optical filter finesse.

Analysis of linear crosstalk in photonic crossbar switches based on on/off gates

This paper presents expressions for the linear crosstalk and the resulting error rate degradation in a photonic switch based on on/off gates, considering cyclostationary interference among digital signals. The periodically time-varying interference accounts for the dependencies of error probability both on the spectrum of transmitted signals and on random bit-skews between channels. This paper quantitatively verifies the intuitions that bit-aligned channels increase the crosstalk power penalty, and that a sharp modulating pulse improves error probability. These dependencies on spectrum and bit-skews become noticeable at high crosstalk levels caused by either a low contrast ratio or a large switch size.

Performance assessment of two-channel dispersion-supported transmission systems using single- and double-cavity Fabry-Perot filters as demultiplexers

This letter presents a theoretical performance assessment assuming that single- and double-cavity Fabry-Perot filters are used as demultiplexers in two-channel dispersion supported transmission systems operated at 10 Gb/s. A suitable simulation methodology has been used that combines signal simulation with noise analysis in optically amplified multichannel direct detection systems. The results obtained show an improved system performance if a double-cavity Fabry-Perot filter is used as a demultiplexer, resulting in a WDM transmission with a crosstalk power penalty within 1 dB in the region of low dispersion penalty of the DST method, for a channel spacing of 1 nm. The robustness against dependencies on data patterns is verified.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Performance implications of component crosstalk in transparent lightwave networks

We find both theoretically and experimentally that crosstalk due to small component imperfections in transparent lightwave networks will tend to generate large performance degradations. The most severe degradations, which arise from interference between signal and crosstalk electric fields, are far larger than those predicted from simple power-addition without field interference. Systems corrupted by crosstalk from a single source of optical-field leakage appear to require component crosstalk levels /spl epsiv//spl les/-20 dB, if conventional receivers are used. Since the degradations are believed to depend on N/spl radic//spl epsiv/, where N is the number of crosstalk generators, each producing a crosstalk power level of /spl epsiv/, the problem is expected to grow rapidly severe with network size. >

波長多重によるインサービス光線路監視試験技術

This paper describes the in-service optical fiber monitoring for gigabits/s optical telecommunications where signal and monitor light waves are on a single-mode fiber using wavelength division multiplexing technique. In general, this monitoring is subject to crosstalk, which causes an increase in bit error rate in the signal channel and decreases in dynamic range in loss and reflection distribution measurement using optical time domain reflectometry in monitor channel. First, permissible conditions for the crosstalk power levels of signal and monitor channels were studied theoretically and experimentally. According to these conditions, the performance of optical monitor equipment and components have been clarified. It was found that optical time domain reflectometry using heterodyne detection was efficient for its large dynamic range and maximum crosstalk power level. As a result, a practical in-service monitoring system has been demonstrated.

Crosstalk penalties in a bidirectional fiber-optic WDM system

Crosstalk power penalties were calculated for a bidirectional fiber-optic wavelength division multiplexed (WDM) system (622 Mb/s at 1300 nm in one direction and 155 Mb/s at 1550 nm in the other) using the system simulation program Simulation of Fiber-Optic Communication Systems (SIFOCS). Nonreturn-to-zero (NRZ) and return-to-zero (RZ) modulation formats were investigated with the result that, in most cases, the power penalty was shown to depend only weakly on individual system properties and is close to the values found in the literature, but there are exceptions where it may be greater than usual (e.g. at 622 Mb/s RZ). These can be detected and investigated with the methods reported. The crosstalk power penalty becomes excessive only for signal-to-crosstalk power ratios well below 10 dB and is negligible for values greater than 15 dB. The calculation methods used can be easily adapted to other systems, e.g. to systems with more closely spaced wavelengths. >

Crosstalk analysis and filter optimization of single- and double-cavity Fabry-Perot filters

Since Fabry-Perot (FP) filters are major candidates for use as demultiplexers in wavelength division multiple access (WDMA) networks, the authors have analyzed, in an exact and unified manner, the crosstalk degradation for several different variations of the basic FP filter, compared their performances, and optimized their design parameters. A description is given of the system model (a wavelength division multiple access system consisting of M connected transmitters and receivers with each transmitter consisting of an on-off modulated fixed-frequency laser, with the 'off' power level equal to zero) and the parameters used throughout, along with a brief discussion of relevant Fabry-Perot equations and terms and the four different filter structures analyzed. The crosstalk is examined for the cases of a single-cavity FP filter, a double-pass FP filter, a two-stage double-cavity FP filter, and a vernier double-cavity FP filter. The criteria for the performance comparison were the worst-case crosstalk, BER, and crosstalk power penalty.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Crosstalk and mode coupling between cores of image fibres

The phenomenon of crosstalk between the cores of image fibres is studied experimentally. Illuminating a core of the image fibre, the crosstalk power appearing on the unilluminated cores is measured. A two-mode coupling theory is presented to describe the crosstalk.

Statistical Behavior of Crosstalk Power Sum With Dominant Components

The literature of digital transmission on wire-pair cables generally considers the probability distributions of both pair-to-pair crosstalk loss and its power sum to be normal on a dB scale. This paper presents extensive measured data of crosstalk among connectors and wire pairs on the backplane and associated stub cable of 466-type apparatus cases of the existing T1 system. The measured probability distribution of crosstalk power sum “bends” toward more severe crosstalk levels in the lower tail region, which is important for T1 system engineering. This bend is because of the effects of a few dominant components (i.e., within-slot or within-harness crosstalk) in the power sum. The simple normal model is too optimistic by 4 dB in estimating apparatus-case-crosstalk power sum at 0.1 percentile level This paper shows that both the Monte Carlo and the lower bound methods for power sum calculations predict this bend in close agreement with the measured data. Although apparatus-case-crosstalk power sum is worse than previously assumed, the performance of T1 system has been adequately protected by the extra margin in the previous engineering rules to cover unknowns.