Error-free operation of wavelength conversion laser for photonic crossconnect node at 125 Mbit/s

Error-free operation of a wavelength conversion laser diode (WCLD) is demonstrated. The optical input/output interface scheme of the WCLD has been examined and wavelength switching at a data rate of 125 Mbit/s has been experimented on. Good error rates below 10/sup -10/ and a small sensitivity degradation of 2.0 to 2.9 dB were achieved.<>     

Wavelength conversion at 10 Gb/s using a semiconductor opticalamplifier

Data at 10 Gb/s has been translated from an input signal wavelength to another wavelength, either longer or shorter, using gain compression in a 1.5-μm semiconductor optical amplifier for wavelength conversion. To achieve operation at such high bit rates, the probe (shifted) input must be intense enough to compress the gain of the amplifier significantly. This reduces the gain recovery time of the amplifier because of probe stimulated emission. A consequence of the intense probe is an extinction ratio deduction. Using moderate input powers, wavelength conversion is achieved over a 17-nm (2-THz) range, with 0.7-3-dB power penalties     

All-optical wavelength conversion at 160 Gbit/s using SOA and silicon-on-insulator photonic circuit

Error-free operation of an all-optical wavelength converter at 160 Gbit/s based on a semiconductor optical amplifier and a silicon-on-insulator photonic circuit, consisting of two cascaded Mach-Zehnder delay interferometers, is demonstrated.     

连续锁模Ar~+激光器的噪声特性

本文从时域特性出发,分析了锁模Ar~+激光器中噪声的来源,假设表征光脉冲强度涨落的随机函数为A(t)=A_oexp[-i(Ωt+(?)_t)],得到描述光强涨落的噪声谱为罗仑兹线形,这与实验结果是一致的.     

光模数转换系统中幅度不均匀性的仿真

分析了高速光模数转换系统中采样时钟幅度不均匀性的产生机理及其对高速光模数转换整体系统性能的影响.结合MATLAB软件和OptiSystom软件,仿真了光模数转换系统的架构和性能,论证了实验系统中各个环节产生的采样时钟幅度不均匀性.仿真结果与实验结果相符,说明了仿真论证的可靠性.     

Measurement of the timing jitter and pulse energy fluctuation of aPLL regeneratively mode-locked fiber laser

We measured the timing jitter and pulse energy fluctuation from the noise RF power spectra of a 10-GHz regeneratively mode-locked erbium-doped fiber laser in both free-running and phase-locked-loop (PLL) operation. The measured timing jitter was 90 and 120 fs for free-running and PLL operation, respectively. These values are to our knowledge the smallest timing jitter reported in actively mode-locked fiber lasers. The pulse energy fluctuation was 0.24% for both operation     

High-quality optical pulse train generation at 80 Gb/s using a modified regenerative-type mode-locked fiber laser

In this paper, we present an experimental demonstration of a high-quality optical pulse train generation at 80 Gb/s. The successful operation was achieved via two steps. In the first step, an ultrastable 2-ps (FWHM), 40-Gb/s pulse train was generated using the technique of fourth-order rational harmonic mode locking. The mode-locked laser was stabilized using a regenerative-type base-line extraction feedback technique. In the second step, an external fiber loop mirror consisting of several meters of polarization-maintaining fiber was used to double the repetition rate from 40 to 80 Gb/s. The output of the 80-Gb/s pulse train shows very small pulse-to-pulse amplitude fluctuation with good long-term stability.     

Generation of a 4 /spl times/ 100 GHz pulse-train from a single-wavelength 10-GHz mode-locked laser using superimposed fiber Bragg gratings and nonlinear conversion

In this paper, the design of a simple and practical repetition-rate multiplier based on superimposed-chirped fiber Bragg gratings (FBGs) is presented. A tenfold increase in the repetition rate of a mode-locked fiber source, by generating a 100-GHz optical pulse train from a 10-GHz train, is demonstrated experimentally. As compared with previous demonstrations, the superimposed FBG filter was specifically designed to decrease the duty cycle of the generated pulse train or, in other words, decrease the pulsewidth. In addition, a fiber nonlinear optical loop mirror (NOLM) is used to eliminate the pulse-to-pulse phase fluctuations in the output high-repetition-rate train and to achieve a wavelength-tunable transform-limited pulse sequence. Moreover, it is shown that nonlinear conversion using the NOLM can be used to simultaneously generate multiwavelength high-repetition-rate optical pulse trains (4 /spl times/ 100 GHz in the example shown here).     

Wavelength conversion at 10 Gb/s by four-wave mixing over a 30-nminterval

We show that the use of a long semiconductor optical amplifier increases the error-free conversion interval of a four-wave mixing (FWM)-based wavelength converter. 30-nm wavelength down-conversion and 15-nm up-conversion have been obtained at 10 Gb/s. This result is a significant improvement over the previous best performance of a FWM-based wavelength converter and suggests that the full erbium-doped fiber amplifier bandwidth can be covered with FWM wavelength converters     

10-Gb/s agile wavelength conversion with nanosecond tuning timesusing a multisection widely tunable laser

Simultaneous wavelength conversion and high-speed wavelength switching has been demonstrated using a multisection grating-assisted coupler sampled reflector laser. A high-output extinction ratio (>20 dBm) over a 30-nm wavelength range was observed. The 10-Gb/s data was switched between different output wavelengths at a 2.5-Gb/s rate and transmitted over 50 km of dispersion-shifted fiber with a small measured power penalty (<2 dB)     

Retiming and reshaping function of all-optical clock extraction at 160 Gb/s in monolithic mode-locked laser diode

Retiming and reshaping behaviors in an all-optical clock extraction operation at 160 Gb/s using a monolithic 160-GHz mode-locked laser diode was investigated in detail. The investigation of the basic clock extraction performance revealed that the clock extraction was stable against the input of data signals with a long zero sequence and with a poor extinction ratio. The clock extraction performance when the input data had random intensity and timing fluctuations was also discussed. We revealed that the generated clock pulse train absorbed well the intensity and timing fluctuations of the input data. Stable clock extraction remained even when the input data had 25% of the peak-intensity fluctuations. A timing jitter reduction from 0.85 to 0.32 ps was also achieved. The detailed investigation in the frequency-domain analysis of the jitter spectra revealed that the timing jitter reduction originated from the short-term stability of the slave 160-GHz mode-locked laser diode.     

Jitter analysis and modeling of a 10 Gbit/s SerDes CDR and jitter attenuation PLL

Jitter analysis and a linear model is proposed in this paper which predicts the characteristics of serial-deserial(SerDes) clock and data recovery circuit,and the characteristics include jitter transfer,jitter tolerance and jitter generation are particularly analyzed.The simulation results of the clock data recovery(CDR) model show that the jitter specifications exceed the mask of ITU-T optical transport network(OTN) G.8251 recommendations.The whole systems are validated by 9.95-11.5 Gbit/s CDR and the jitter attenuation phase locked loops(PLL) circuits using TSMC 65 nm CMOS technology.     

锁模Cr4+:YAG固体激光器脉冲的时间抖动理论性分析

固体激光器的脉冲稳定性限制了激光器在很多方面的应用.从被动锁模Cr4 :YAG激光器腔内脉冲的传播方程出发,得到了带噪声的非线性Schr(o)dinger方程,进一步分析了脉冲在激光器腔内的时间传播形式,建立了锁模Cr4 :YAG固体激光器振荡脉冲参数的线性扰动方程,推导出了脉冲参数波动的动力学方程以及时间和相位的均方根波动关系式.提出了一个减少锁模Cr4 :YAG固体激光器脉冲时间抖动的实验方案,旨为锁模Cr4 :YAG固体激光器能够用作光纤通信和量子通信中的光源提供理论上的帮助.     

Hybridization platform assembly and demonstration of all-optical wavelength conversion at 10 Gb/s

The development of a high-performance hybrid integration platform is demonstrated using an all-optical wavelength converter based on an integrated semiconductor optical amplifier (SOA) Mach-Zehnder interferometer (MZI). The device structure, transfer functions, power penalties and regenerative properties are presented for data rates up to 10 Gb/s.     

Cascaded wavelength conversion by four-wave mixing in a strained semiconductor optical amplifier at 10 Gb/s

We demonstrate for the first time cascaded wavelength conversion by four-wave mixing in a semiconductor optical amplifier. Bit-error-rate performance of <10/sup -9/ at 10 Gb/s is achieved for two conversions of up to 9 nm down and up in wavelength. For two wavelength conversions of 5 nm down and up, a power penalty of 1.3 dB is measured. A system of two wavelength converters spanning 40 km of single-mode fiber is also demonstrated.     

Investigation of a tuneable mode-locked fiber laser for application to multipoint gas spectroscopy

This paper reports on an initial investigation into the operation of a mode-locked fiber laser system for application in gas spectroscopy as a multipoint multigas sensor. Wavelength selection is performed by use of multiple chirped gratings and fine tuning by using the dispersion properties of the chirped gratings. A tuning rate of /spl sim/0.014 nm per kHz change in mode-lock frequency (at the third harmonic) has been demonstrated, which is suitable for scanning across gas absorption lines. Key issues that have an important bearing on the tuning are discussed, including gain flattening and polarization drift. This paper investigates a method of multiplexing the sensor cells with the mode-locked system. Preliminary results for a two cavity system are presented to verify the principles of the technique.     

Multi-wavelength conversion at 10 Gb/s and 40 GHz based on nonlinear effects in HNLF

We present experimental results and numerical simulations of multiple-wavelength conversion of 10 Gb/s non-return-to-zero (NRZ) signals based on 2-pumps four-wave mixing and of 40Gb/s return-to-zero (RZ) signals through supercontinuum (SC) generation in a highly nonlinear fiber (HNLF). Two pumps with a 100 GHz separation located in the region of anomalous dispersion of the HNLF were combined to form a beat signal The 10 Gb/s NRZ signal anchored on ITU grid was converted with gain into 6 new signals and 6 idlers with 100 GHz separation. High quality sc with bandwidth exceeding 50 nm was generated when the HNLF was pumped with 25 dBm average power. Sliced sc can be used as a source for all-optical multicasting.     

Demonstration and analysis of a 40-gigasample/s interferometric analog-to-digital converter

A novel interferometric scheme for photonic analog-to-digital conversion is for the first time experimentally demonstrated at a real-time sample rate of 40 gigasamples/s. The scheme includes sampling as well as binary encoding, and the input signal in the experiment was a 1.25-GHz sinusoidal tone that was successfully digitized with a nominal resolution of 21 digital levels. Single-sample measurements yielded an effective number of bits (ENOB) of 2.6, which was limited by thermal detection noise while multisample averaged measurements resulted in an ENOB of 3.6 bits, mainly limited by phase drift. Apart from the experimental results, this paper covers an extensive theoretical analysis of the system, including calculations on the fundamental maximum bandwidth, the required optical power, the generated binary code, and its error robustness, as well as the impact of detection noise on the signal-to-noise ratio of the digitized signal. The major benefits of this interferometric scheme are that only one standard phase modulator is required and that the phase swing does not have to be larger than /spl plusmn//spl pi/ to reach the full digital value space.     

On quantization,truncation and jitter errors in the sampling theorem and its generalizations

Any band-limited signal f(t) can, according to the sampling theorem, be exactly reconstructed from its sampled values. If the signal is not necessarily band-limited, an alternative model states that it can be approximately reconstructed from its samples. Such signals can also be approximated by generalized sampling sums which can be interpreted as discretized convolution integrals of Fejér's type. In all three cases the physically realized signal is often only roughly equal to f(t) due to errors caused e.g. by the sampling mechanism.In this paper the following types of errors are treated: (1) round-off error arising when quantized sampled values are used, (2) truncation error, arising when a truncated sum is used for representation, (3) time jitter error, a result of sampling at instants slightly different from the sample values. All proofs employ deterministic methods.     

Wavelength conversion up to 18 nm at 10 Gb/s by four-wave mixing in a semiconductor optical amplifier

We characterize the conversion bandwidth of a four-wave mixing semiconductor optical amplifier wavelength converter. Conversion of 10-Gb/s signals with bit-error-rate (BER) performance of <10/sup -9/ is demonstrated for wavelength down-shifts of up to 18 nm and upshifts of up to 10 nm.