Large electroabsorption effect in GaInAs/InP multiple quantum well(MQW) optical modulator grown by OMVPE

To make short, high-speed electroabsorption modulators, it is necessary to use a material and device structure that displays a large change in absorption coefficient, Δα. The authors report a device with Δα=7800 cm^-1, which provides an on/off ratio of 44:1 at λ=1.6 μm with a drive voltage of 10 V     

Electroabsorption properties of InGaAs/InAlAs multiple quantum wellstructures at 1.5 μm

The electroabsorption properties of InGaAs/InAlAs MQW structures are characterised in terms of Δα, Δα/F and Δα/α₀, where Δα is the electroabsorption, α₀ is the residual absorption coefficient under zero bias, and F is the applied electric field. The limitations of these structures for 1.5 μm modulators are primarily due to the relatively small Δα/F values as a result of the small well width. The results are compared with the literature     

On the figures of merit for electroabsorption waveguide modulators

ΓΔα/F and Δα/α₀ (where Δα is the absorption change, α₀ is the residual absorption, F is the applied electric field, and Γ is the optical confinement factor in the waveguide) have been separately proposed as the relevant figure of merit for electroabsorption waveguide modulators. Using a quantitative and systematic argument, the authors show that they are both necessary and important to the total performance of the modulator     

Independence of absorption coefficient-linewidth product tomaterial system for multiple quantum wells with excitons from 850 nm to1064 nm

The authors have measured the absorption coefficient (α) and linewidth (Δ) of the excitons of GaAs/AlGaAs and strain-balanced InGaAs/GaAsP multiple-quantum-well (MQW) modulators with wavelengths from 850 to 1064 nm. They find that α decreases and Δ increases as wavelength increases, but their product, and thus the integrated absorption coefficient, remains roughly constant. Thus, the reduced performance observed for longer wavelength modulators is due to exciton broadening     

Theoretical design optimization of multiple-quantum-wellelectroabsorption waveguide modulators

Optical on-off modulators require low insertion loss, high contrast ratio (CR), small drive power and large bandwidth or bit-rate. A systematic approach to optimize the total performance of these modulators based on the quantum-confined Stark effect is presented here. The approach consists of minimizing the power/bandwidth ratio while satisfying a given CR and insertion loss. Our design consists of a large-core multimode passive waveguide with a thin buried active layer. The passive waveguide is designed to yield a high coupling efficiency to conventional single-mode fibers. The quantum well material structure is designed to maximize Δα/ΔF², while maintaining a sufficiently large Δα/α₀, where Δα is the absorption change, α₀ is the residual absorption at zero bias, and ΔF is the swing of the applied electric field. Our theoretical model shows that i) wider quantum wells give larger Δα/ΔF², and ii) the bandwidth/power ratio as high as 4 GHz/mW can be achieved simultaneously with small insertion loss, For example, with a drive voltage of 3 V, an RC limited bandwidth as high as 60 GHz is predicted, while a contrast ratio of 20 dB and a total insertion loss of 4.5 dB may also be obtained     

Modeling of excitonic electrorefraction in InGaAsP multiple quantumwells

The absorption spectra in an InGaAsP/InP multiple quantum well for different values of electric field, applied perpendicularly to the quantum well layer planes, are delineated by modifying the model given by P.J. Stevens et al. (IEEE J. Quantum Electron., vol.24, pp.2007-2015, 1988). The essential deviations in the present model lie in the inclusion of the additional broadening of excitons due to composition fluctuation in the quaternary, the forbidden transition between the second heavy hole subband and the first conduction subband, an improved excitonic envelope function dependent on both the in-plane and transverse separation of electrons and holes, and a modified calculation of oscillator strength. The modeled curve's are then used to calculate the values of Δn, the change in refractive index due to field, and the ratio Δn/Δk, where Δk is the extinction coefficient, using Kramers-Kronig relations. The calculated values are found to agree with the experimental data for 1.537 μm     

Design and demonstration of polarization-insensitive Mach-Zehnderswitch using a lattice-matched InGaAlAs/InAlAs MQW and deep-etchedhigh-mesa waveguide structure

The authors have designed and demonstrated a 2×2 Mach-Zehnder switch in view of polarization independence as well as low propagation loss (α) and absorption change (Δα). To obtain polarization-insensitive refractive index change (Δn), a lattice-matched InGaAlAs-InAlAs multiple quantum-well (MQW) with a large detuning wavelength was used. Moreover, to reduce the insertion loss difference between polarizations, we applied a multimode-interferometer 3-dB coupler and a deep-etched high-mesa waveguide structure. This switch, therefore, can provide polarization-independent operation about both driving voltage and insertion loss, which is indispensable to practical optical switching applications. We also paid attention to Δα suppression when we decided the value of wavelength detuning and the length of the phase shift region. We also investigated the wavelength dependence of the switch. Within 1530-1560 nm, which is the erbium-doped fiber amplifier (EDFA) gain band, polarization independence in the driving voltage and the crosstalk was maintained. This result shows that the switch is also applicable in wavelength division multiplexing (WDM) applications     

Dispersive subband-resonant nonlinearity in amorphous Si/SiO2 quantum well structures

Dispersive nonlinearity in amorphous Si/SiO₂ quantum well structures (QW's) has been investigated. The refractive index changes obtained from the intensity-dependent reflection spectra are nonlinearly dependent on the excitation intensity and can be described by the model of the saturating nonlinearity at low pump intensities. The nonlinear refractive index reveals resonant behavior associated with the subband structure of the QW's. The saturated nonlinear index and the saturation intensity have been obtained as Δn_s=-0.11 and I_s=1.9 MW/cm² at the transitions between the lowest subbands, and Δn_s~0.3 and I_s~0.5 MW/cm² at the transitions between the second subbands of the valence and conduction bands. The nonlinearity for the second subband transitions has been found high enough to provide potentially bistable operation, but the bistability is not expected at the transitions between the ground subbands. Carrier lifetime less than 1 ps restricting the switching time of the nonlinearity has been estimated from the saturation intensity     

Hg0.4Cd0.6Te 1.55-μm avalanche photodiodenoise analysis in the vicinity of resonant impact ionization connectedwith the spin-orbit split-off band

The authors describe the electrical and optical characterization of three Hg_1-xCd_xTe avalanche photodiodes manufactured using planar technology with composition parameter x near 0.6. This alloy composition leads to devices that are well suited for 1.55-μm detection. From the noise analysis under multiplication, the authors show the tight dependence of the ratio β/α (of the hole; and electron ionization coefficient, respectively) upon x and the ratio Δ/E_g where Δ is the spin-orbit splitting energy and E _g is the bandgap energy. It turns out that in these alloys around x=0.6, Δ is very close to the bandgap energy so β/α reaches its maximum value. Owing to this property, which is characteristic of II-VI compounds, Hg_1-xCd_xTe is a good candidate for 1.3-μm to 1.6-μm avalanche photodiodes     

The expansion factor of error-control codes

We investigate the expansion factor Δ(C) of an (n, k, d) linear block code. Δ(C) indicates how well the code preserves and expands the distance relations of the input. We relate Δ(C) to n, k, d, the total weight, and the weight distribution, and we find bounds on Δ(C)     

On the optimization of InGaAs-InAlAs quantum-well structures forelectroabsorption modulators

Several InGaAs-InAlAs multiple quantum-well structures grown by metalorganic vapor phase epitaxy (MOVPE), with various Ga content and quantum-well width, have been investigated for electroabsorption modulators (EAM's). The light-hole heavy-hole splitting, the chirp parameter, the insertion loss and the figure of merit ΓΔα/F of the different InGaAs-InAlAs structures have been evaluated with photocurrent, photoluminescence, absorption and X-ray measurements. It was then possible to experimentally study the influence of different parameters of the multiple quantum-well structures on the device performance. The use of tensile strained barriers are believed to be responsible for the improvement in the figure of merit. Structures with unresolved light-hole and heavy-hole transitions, with negligible chirp, with adequate insertion loss and with extremely high values for ΓΔα/F have been obtained, however, not simultaneously     

Loss-imbalance equalization in arrayed waveguide-grating (AWG)multiplexer cascades

How to connect arrayed-waveguide-grating (AWG) multiplexers in cascade is discussed with the goal of equalizing the loss imbalance among frequency-division-multiplexed (FDM) channels. This paper proposes to average the FDM-channel loss over the cascaded multiplexers by shifting port connections between each adjacent multiplexer. A simulation predicts that, in a cascade of M periodic N×N multiplexers, shifting the connections by N/M reduces the loss imbalance from MΔα to Δα/M where Δα [dB] denotes the loss imbalance per multiplexer. This improvement will extend the cascadable node number in all-optical FDM networks. The prediction is confirmed experimentally in an optical add-drop filter (M=2) constructed with a silica-based AWG 16×16 multiplexer; the largest loss difference among 15 FDM channels is reduced from 5.0 dB to 1.5 dB. This paper also reports that imperfect multiplexer periodicity due to waveguide dispersion restricts the equalizable frequency bandwidth to less than several free spectral ranges (FSR's)     

Amplitude and phase modulation in a 4 μm-thick GaAs/AlGaAsmultiple quantum well modulator

The experimental results of a 4 μm-thick GaAs/AlGaAs MQW modulator show an ~10:1 on/off ratio with an applied voltage of 20 V (Δα~6×10³/cm at E~50 kV/cm) and ~0.4 πrad of phase shift with an applied voltage of 10 V (Δn~0.04 at E~25 kV/cm). Such high electro-optical modulations have previously been reported only in MQW optical waveguide modulators     

On the prediction of the mode-partitioning floor in injectionlasers with multiple side modes at 2 and 10 Gb/s

An analytical method is presented for estimating the error-rate floor to be expected in a digital transmission system using an injection laser. A computer model of the laser, including Langevin noise sources and with random sequences of signal pulses, is used to find the probability of the main-mode average power during an ON pulse being less than 0.5 times normal average power during an ON pulse event. With sufficient fiber dispersion, this constitutes a useful estimate of the mode-partitioning floor. Three methods are included for determining the floor: (1) direct calculation using a single Monte Carlo run; (2) extrapolation to floors in the 10^-10 region of a series of Monte Carlo runs for different laser modal loss differences (Δα); and (3) extrapolation of a single Monte Carlo run to low error-rate floors using a new analytical expression with parameters derived from the Monte Carlo run. In order to avoid a serious error-rate floor at 2 or 10 Gb/s, it is found that Δα must be about 10 or 30-35 cm^-1, respectively     

The influence of directly modulated DFB LD sub-mode oscillation onlong-span transmission system

Error-rate floors have been observed in several long-span transmission experiments at 500 Mb/s, using 1.5-μm distributed-feedback laser diodes (DFB LDs) and 1.3-μm zero dispersion optical fibers. It is proposed that for the threshold gain difference between main and submode (for DFB LDs), Δα is a good parameter to specify the submode oscillation characteristics. It is experimentally and theoretically confirmed that the threshold gain difference Δα must be greater than 5-6 cm^-1, to avoid the error rate floor at 500 Mb/s. It was also confirmed that λ/4 phase-shifted DFB LDs can easily satisfy this condition     

Δk-radar equivalent of interferometric SAR's: a theoreticalstudy for determination of vegetation height

Theoretical aspects of estimating vegetation parameters from SAR interferometry are presented. In conventional applications of interferometric SAR (INSAR), the phase of the interferogram is used to retrieve the location of the scattering phase center of the target. Although the location of scattering phase center for point targets can be determined very accurately, for a distributed target such as a forest canopy this is not the case. For distributed targets the phase of the interferogram is a random variable which in general is a function of the system and target attributes. To relate the statistics of the interferogram phase to the target attributes, first an equivalence relationship between the two-antenna interferometer system and an equivalent Δk radar system is established. This equivalence relationship provides a general tool to related the frequency correlation function (FCF) of distributed targets, which can conveniently be obtained experimentally, analytically, or numerically, to the phase statistics of the interferogram. An analytical form for the p.d.f. of the interferogram phase is obtained in terms of two independent parameters: 1) ζ: mean phase and 2) α: degree of correlation. ζ is proportional to the scattering phase center and n is inversely proportional to the uncertainty with which ζ can be estimated. It is shown that α is directly related to the FCF of the distributed target which in turn is a function of scattering mechanisms and system parameters. It is also shown that for a uniform closed canopy the extinction and the physical height of the canopy top can be estimated very accurately. Some analytical and numerical simulations are demonstrated     

Analysis of transient interband light modulation by ultrashortintersubband resonant light pulses in semiconductor quantum wells

We explore the transient characteristics of an interband resonant light modulation process by ultrashort intersubband resonant light pulses in semiconductor quantum wells (QW's). The modulation characteristics in a three-level semiconductor QW system, including the effects due to in-plane momentum, have been investigated by a numerical analysis of the coupled Bloch equations using a density-matrix approach. We have studied the effect of the carrier density and the nature of the intersubband coupling light on the transient absorption features of the interband light. The modulation process has been compared in doped and undoped QW's. The switching behavior of the strong interband light field in presence of a train of intersubband light pulses has also been discussed     

一种可预测TFT-LCD垂直串扰水平的定量分析方法

电容耦合和TFT漏电引起的垂直串扰问题在高分辨率液晶显示器产品上变得较为突出,大大影响了产品良率。业内通常采用VESA 2.0标准利用窗口画面测试串扰水平,但目前还没有一种预测串扰水平的定量分析方法。本文从垂直串扰形成机理出发,提出了一种可预测垂直串扰水平的定量分析方法,可预测出不同模式产品垂直串扰最严重的画面,有利于我们更好地研究和分析产品的品质。首先,通过分析垂直串扰机理,得到了垂直串扰现象与源电压差之间的定性对应关系。然后,通过分析V-T曲线,得到窗口画面下的亮度变化与源电压差之间的定量关系|ΔL|=kα|ΔV|。通过PCB板上的输出节点可以得到各灰阶对应的正负源电压,依据灰阶画面对应的源电压找到V-T曲线上对应的点可求出对应的斜率k值,依据漏电机理可求出对应的源电压差值|ΔV|,|ΔV|变化不大时可认为漏电电压降系数α为定值,故可计算出不同灰阶背景画面在窗口画面影响下的亮度变化值|ΔL|,将|ΔL|除以V-T曲线上对应灰阶的亮度值即可得到串扰值,通过比较不同窗口画面的串扰计算值即可得出串扰最严重的画面。最后,采用VESA 2.0标准方法测试不同窗口画面下的垂直串扰水平,与此方法的计算结果进行比较,串扰变化趋势吻合较好,TN和ADS模式下的线性相关系数分别达0.98和0.93以上。结果表明,此方法可以用来定量地研究产品垂直串扰的问题。     

A new construction for n-track (d, k) codes with redundancy

A new construction for n-track (d, k) codes with redundancy r, referred to as (d, k; n, r) codes, is presented. This construction applies single-track (d, k+Δk) codes (with certain extra constraints and appropriate amounts of delay) on each of the n tracks. This construction achieves a large part of the capacity increases possible when using (d, k; n, r) codes, has simple encoders and decoders, and exhibits considerable robustness to faulty tracks. It is shown that under this construction, (d, k; n, r) codes can achieve at least (n-r-1:)/n*100% of the gap in capacity between conventional (d, k) and (d, ∞) codes. Several practical examples of (d, k; n, r) codes under this construction are presented