Relative Intensity Noise Performance of Wavelength Converters Based on Four-Wave Mixing in Semiconductor Optical Amplifiers

We present a numerical and experimental study of the relative intensity noise (RIN) induced by the four-wave mixing (FWM) based wavelength conversion process in a semiconductor optical amplifier (SOA). The study is based on the RIN evaluation of the converted signal, under various operating conditions (input power levels and wavelength detuning) and input noise characteristics. A detailed numerical model is employed to simulate the FWM process, taking into account the amplified spontaneous emission (ASE) noise, the wavelength dependent gain, and the interaction of four waves in the SOA (two input waves and two product waves). It is shown that for low noise input signals, the output RIN is determined by the power levels of the pump and signal and the wavelength detuning. Operation under saturation allows reduction of the output RIN levels with respect to the input     

Small signal analysis with higher-order dispersion for dispersive optical communication systems

In this paper an improved small-signal analysis has been presented for analyzing the influence of the higher-order dispersion terms on dispersive optical communication systems operating near zero-dispersion wavelength for single mode fiber. A generalized conversion matrix has been reported that gives the transfer function of intensity and phase from the fiber input to fiber output for laser source including the influence of any higher-order dispersion term. In addition, analysis is applicable to evaluate the impact of combined or independent dispersion terms on small-signal frequency response and relative intensity noise (RIN) response of an ultrafast laser diode including noises due to laser spontaneous emission rate and average photon density. These responses are plotted for second-, third-, and fourth-order dispersion terms and their combinations. It is observed that the higher-order dispersion terms have significant impact on frequency and RIN responses at large modulating frequencies and large propagation distances.     

Noise Transfer Characteristics in a Semiconductor Optical Amplifier With Application to Wavelength Conversion Based on a Delay Interferometer

A theoretical model is presented to analyze the noise transfer characteristics in a semiconductor optical amplifier (SOA) under the excitation of a noisy pump signal and a noise-free probe signal. An analytical expression is derived for the optical signal-to-noise ratio (OSNR) of the output probe signal from the SOA. The influence of the gain saturation of the SOA, and the pump and probe signal powers on the noise transfer characteristics is investigated. The noise transfer model is used to determine the output noise power of a delay interference wavelength converter. An analytical expression is obtained for the nonlinear phase change in the SOA, which determines the output power of the wavelength-converted signal. These results show that the noise transfer in the wavelength conversion can be suppressed by increasing the probe signal power, but that the improvement in the output signal OSNR relative to the input signal OSNR is accompanied by a reduction in the conversion efficiency. This fundamental tradeoff can be readily investigated during the design optimization process using the concise results derived in this paper.     

Noise in IMPATT diodes: Intrinsic properties

The design of low-noise IMPATT diodes has been aided by theories describing the noise generation under small-signal conditions. A major deficiency in this procedure has existed in that there is no apparent connection between the small-signal behavior and the great increases in the noise observed in large-signal operation. As a remedy a theory has been developed for the noise generation at arbitrary signal levels by using a Read diode model. The theory is based on a linearization technique for calculating the spectrum of homogeneous noise with linear damping resulting in a separation of the large-signal and noise problems. The open-circuit noise voltage increases strongly at high signal levels due to nonlinear parametric interactions and gives rise to a rapid increase in the noise measure as a function of the generated microwave power. Operating parameters are derived that optimize the power-noise ratio. A long intrinsic response time is found to be beneficial in achieving high power as well as low noise. Other factors affecting the design and choice of material for IMPATT diodes are discussed. An important feature of the presented theory is that a complete design optimization with respect to the power-noise characteristics can be carried out provided reliable information exists about the ionization rates and the drift velocities. A simpler alternative is to obtain the physical quantities governing the power-noise behavior from small-signal admittance and noise measurements. Good agreement has been obtained with experimental power-noise measurements by this method. As an application of this procedure a state of the art comparison is given for GaAs, Ge, and Si diodes at 6 GHz.     

Effects of electron-beam confinement on klystron efficiency

This paper describes experiments which were devised to examine large-signal performance in klystrons focused by magnetic fields of finite magnitude. The experimental results are related to the theoretical analyses of Brewer, who treated small-signal bunching for magnetic fields of various magnitudes, and of Webber, who treated large-signal bunching for an infinite magnetic field intensity. The experiments agree with these theories where they apply, and additionally, the Webber theory continues to describe the transition from small to large signal for finite magnitude magnetic fields, when one uses the appropriate reduced plasma frequency in the bunching parameter. The relation of output cavity RF driving current, output cavity voltage, and magnetic field to conversion efficiency is also examined. It is found that maximum output power is obtained by maximizing driving current, but that with driving currents being equal, higher efficiency may be obtained in the beams which are magnetically more tightly confined; it appears that the higher magnetic field allows the output voltage to rise higher without defocusing the beam in the output cavity.     

Small signal frequency response of a linear dispersive single-modefiber near zero first-order dispersion wavelength

A theory describing the propagation of signal and noise through a lossless linear dispersive singlemode fiber near zero first-order dispersion point is presented. Using the small signal analysis approach, this theory has been applied to modulation and noise characteristics of a laser diode, emitting near the fiber zero-dispersion wavelength, to obtain the small signal frequency response and the signal relative intensity noise (RIN) at the output of the fiber. It has been shown that, for very long broad-band transmission lines, dispersion effects may be no longer negligible also at zero-dispersion wavelength     

Theory of gyro-travelling-wave tubes at cyclotron harmonics

The generalized theory of gyro-travelling-wave tubes operating at arbitrary cyclotron harmonics is developed taking into account the transverse drift of electron guiding centres. In the framework of small-signal theory, the dispersion equation is obtained and analysed and a problem of spurious counterpropagating waves excitation is discussed. The results of the large-signal theory are presented, which predict a high efficiency of operation at the first four cyclotron harmonics.     

High output power from an efficient praseodymium-doped fluoridefiber amplifier

A praseodymium doped fluoride fiber amplifier (PDFFA) exhibiting efficient operation in both the large- and the small-signal regimes is described. The amplifier, based on a high NA fluoride fiber, exhibited a maximum small-signal gain of 29 dB and a low-pump small-signal. efficiency of 0.13 dB/mW. In the saturated regime a maximum output power of 212 mW was achieved. In this format, the signal-out versus pump-in characteristic, exhibited a slope efficiency of 30%, representing the most efficient conversion from pump to signal yet reported. Detailed spectral characterization reveals small-signal gain in excess of 20 dB over a wavelength range of almost 50 nm and in excess of 100 mW of saturated output available over a 30-nm wavelength range     

Premature collection mode in IMPATT diodes

Experimental efficiencies of up to 35.5 percent have been reported for Read-type GaAs diodes, whereas theoretical calculations have predicted an upper limit of approximately 30 percent for the conversion efficiency of IMPATT diodes. The concept of a premature collection mode is shown to resolve this discrepancy by predicting maximum efficiencies close to 40 percent. Premature collection refers to large-signal conditions where the modulation of the drift width is sufficiently large to result in collection of the avalanche current pulse at drift angles smaller than the small-signal angle. It is shown that a discontinuous transition between the IMPATT and the premature collection modes takes place when the drift angle in the small-signal limit is greater than π. Designing the diode for close to punchthrough conditions in small-signal operation extends the practical frequency range for inducing premature collection by avoiding long drift angles and corresponding rapid conductance saturation in the IMPATT mode. The onset of premature collection is accompanied by a substantial increase in power output because of a more favorable drift angle, and in high noise because of the high RF levels involved. The jump in transit angle causes a discontinuous increase in negative conductance. The hysteresis in the tuning characteristic resulting from this discontinuity has been observed experimentally. Noise measures in the range 60-70 dB have been measured and calculated for the premature collection mode compared to 40-50 dB under large-signal conditions for the IMPATT mode. Therefore, the high efficiencies available with the premature collection mode are expected to be usable only in applications where high noise levels can be tolerated.     

基于掺锗光子晶体光纤的可调谐多波长转换器

为了提高光网络中波长资源的使用效率,利用光纤中瞬态受激拉曼散射效应分析理论,设计了一种基于掺锗光子晶体光纤的可调谐四通道波长转换器。由于受激拉曼散射效应的增益随信号光与探测光波长之间的频移量变化,波长转换器各个转换信道波长可由探测光波长调谐。分析了泵浦信号光输入功率对多波长转换器性能的影响,结果表明:随着输入泵浦功率的增大,多路波长转换器的转换性能更好。用OptiSystem对四通道可调谐波长转换进行仿真,结果表明:所设计的波长转换器能够同时实现四通道波长转换,各信道波长可在1 511~1 569nm进行调谐。     

Configurable three-wavelength Raman fiber laser for Ramanamplification and dynamic gain flattening

A three-wavelength Raman fiber laser is presented that is suitable for the generation of a flattened Raman gain curve over a broad range of extended telecommunications bandwidths. This laser utilizes an 1100-nm Yb-doped cladding pumped fiber laser and a cascaded Raman resonator to generate output radiation at 1427, 1454, and 1480 nm. The slope efficiency for conversion from the 1100- to 14xx-nm radiation is 0.38, the total output power is 1.1 W, and the fiber laser "wall-plug" efficiency is 0.10. The output power at 1454 and 1480 nm is adjusted by varying the reflectivity of the fiber grating output couplers and the suppression of lower order spectral components is between 15 and 20 dB. Low frequency relative intensity noise (RIN) measurements indicate RIN values less than -90 dB/Hz below 100 kHz     

Performance analysis of electro-absorption modulator-based multi-wavelength converter

This paper reports the performance analysis of multi-wavelength converter based on electro-absorption modulator at 10 Gb/s. It has been observed that the input signal wavelength converted into fourteen different wavelength and we observed the performance of converted wavelength only at 1552.02, 1550, 1553.99, 1542.03, 1546.03, 1556.02 and 1558 nm because the rest of converted wavelength quality has poor. The influence of two parameters such as input signal power (?30 to 30 dBm) and probe signal power (0, ?2, 2, 6.99, 10 dBm) of EAM on multi-wavelength conversion performance analyzed by measuring quality factor and converted output signal power. The best converted signal quality is obtained for 10 dBm input signal power with 6.99 dBm probe signal power for this multi-wavelength converter. Further, quality factor, conversion efficiency and extinction ratio are observed as the function of input signal power.     

Noise behavior of microwave amplifiers operating under nonlinear conditions

The noise behavior of microwave amplifiers operating under a large-signal condition has been studied in this paper. A Gaussian noise is added to a microwave signal and they are applied at the input of several amplifying devices. Experimental data show a decrease of the output noise spectral density when the power of the microwave signal at the input of the devices increases due to the compression of the amplifiers. A distortion component due to the interaction of the signal and its harmonics with the noise is also demonstrated from a simplified theoretical model. The statistical properties of the signal and the noise have also been investigated in order to verify the Gaussianity of the noise at the output of the nonlinear circuits. We have also observed that the majority of the measured devices show some variations of their additive noise versus the input power level.     

Large-signal theory of the effect of dispersive propagation on theintensity modulation response of semiconductor lasers

We have derived an exact large-signal theory of propagation in a dispersive fiber of an optical wave with sinusoidal amplitude and frequency modulation. This has been applied to the study of large-signal direct-modulation of semiconductor lasers. It is shown that the large-signal response can significantly deviate from the predictions of the small-signal theory. In particular, the improvement in modulation response caused by frequency-to-intensity modulation conversion in propagation that occurs with small-signal modulation is no longer achieved with large-signal modulation, which could affect systems such as dispersion supported transmission. Experimental results confirm our theory     

Mean Relative Intensity Noise Transfer in Fiber Raman Amplifiers Using Multiple-Wavelength Pumps

We present a numerical analysis of mean relative intensity noise (RIN) transfer in a fiber Raman amplifier (FRA) using multiple-wavelength pumps. For the first time, the RIN transfer is investigated to include pump-to-pump and signal-to-signal RIN transfers, together with the ldquotraditionalrdquo pump, to signal transfer in a multiple-wavelength pumped FRA. We show that for a multiple-wavelength pumped FRA, pump-to-pump and signal-to-signal RIN transfers induce an increase of RIN transfer magnitude. For a multiple-wavelength counter pumped FRA, in additional to the major pump to signal RIN transfer, pump-to-pump RIN transfer is considerable, and signal-to-signal RIN transfer is very small. For a multiple-wavelength co-pumped FRA, both pump-to-pump and signal-to-signal RIN transfers increase RIN transfer magnitude significantly. In addition, for both a co- and counter pumped FRAs, particularly for the co-pumped FRA, the longer wavelength signals experience more RIN transfer than the shorter wavelength signals. Furthermore, the bandwidth of RIN transfer is significantly increased in a multiple-wavelength co-pumped FRA with multiple input signals compared to a single-wavelength co-pumped FRA with an input signal.     

Gain enhancement in gain-shifted erbium-doped fiber amplifiers forWDM applications

Utilization of amplified spontaneous emission (ASE) for gain enhancement in gain-shifted erbium-doped fiber amplifier (GS-EDFA) is analyzed. It is shown that in a multiwavelength regime, gain enhancement as high as 5 dB can be achieved when the counterpropagating ASE is used to invert an unpumped first stage of a two-stage GS-EDFA. We present design criteria for flat gain operation and noise penalty minimization of two-stage GS-EDFAs determined for multiwavelength operation in the 1580-nm wavelength band. The analysis is based on an application of a comprehensive large-signal numerical model which takes into consideration propagation of signal, pump and both the downstream and upstream ASE powers     

Theory of distributed mixing and amplification in a superconductingquasi-particle nonlinear transmission line

An analysis is presented for the behavior of the superconducting quasi-particle nonlinear transmission line driven by a high-frequency local oscillator (LO). The theory developed includes a large-signal nonlinear analysis, a small-signal analysis, and a noise analysis. This model is used to simulate the conversion loss and noise temperature of distributed quasi-particle mixers based on the nonlinear transmission line. The numerical results are compared to an experimental mixer at 460 GHz. The theory also predicts that the nonlinear transmission line will provide parametric amplification when the idler frequency is inductively tuned. This new phenomenon may present new opportunities to low-noise receiver systems at submillimeter wavelengths     

Influence of Pump-to-Signal RIN Transfer on Noise Figure in Silicon Raman Amplifiers

The effect of the relative intensity noise (RIN), transferred from the pump to the signal, in 1-cm-long chip scale silicon Raman amplifiers is investigated in the presence of nonlinear losses. We show that due to the short waveguide length, the reduction in fluctuations that normally occurs due to “walk-off” between pump and signal waves in fiber amplifiers is inefficient in chip scale Raman amplifiers. In the counterpropagating pump configuration, which leads to minimum frequency RIN transfer, fluctuations up to 1.5 GHz are transferred from the pump to the signal. As a case study, the noise figure degrades by as much as 11 dB in the silicon waveguide with the free carrier life time of 0.1 ns, when it is pumped with a laser with a RIN value of $-$125 dB/Hz.      

Power penalty due to optical reflections in erbium-doped fiberpreamplifier

The effect of optical reflections in erbium-doped fiber amplifiers (EDFAs) is investigated primarily as they are used for preamplifying the received signal in a digital transmission system. Both an increase in amplified spontaneous emission (ASE) due to its multiple reflection and an increase in relative noise intensity (RIN) due to interferometric phase-to-intensity-noise conversion are considered as sources of degradation of the receiver sensitivity. The power penalty is calculated and confirmed experimentally     

High-gain two-wave mixing in a Ce-SBN photorefractive crystal:large modulation effect and dynamic properties

We have studied two-wave mixing in a Ce-SBN crystal at different pump-to-signal ratios. A small-signal gain coefficient as high as 15.9 cm^-1 at the 633 nm wavelength was obtained. We also observed the pseudo-periodical variation of the two-wave mixing signal with time at different pump-to-signal ratios. The variation is less significant and aperiodic at small pump-to-signal ratios. This time-variation property is a result of the competition between two-wave mixing, beam fanning, as well as modulation due to the Fabry-Perot cavity formed by the crystal surfaces. We have compared the experimental measurements with the theoretical calculation. There is a large discrepancy between the small-signal theory and experimental observations. An improved theory has been developed by taking into account the large-modulation effect. Although the large-signal theory agrees with the experimental results better than the small-signal theory, it still does not precisely describe the two-wave mixing process with the coexistence of various effects