A quantum amplifier model for predicting temporal variations in theoutput power from a semiconductor laser on a picosecond time scale

A new model for simulating temporal fluctuations in the power emitted by a semiconductor laser is described. Light in the cavity is assumed to circulate in the form of traveling photon packets, in which the photon number fluctuates due to the processes of spontaneous emission, stimulated emission, absorption, scattering, and reflection. The dipole dephasing time T plays a critical role in modeling the interaction of the photon packets and gain medium. The Monte Carlo method is used to simulate the temporal behavior of a continuously pumped Fabry-Perot laser. The laser output power is found to exhibit periodic fluctuations at the cavity transit time frequency (longitudinal mode beat frequency). The amplitude of these fluctuations, as well as the relaxation oscillation, which occurs at a much lower frequency, is strongly influenced by the magnitude of T. The results of these simulations are related to the temporal behavior expected from a conventional FP laser     

A 533-MHz BiCMOS superscalar RISC microprocessor

This 533-MHz BiCMOS very large scale integration (VLSI) implementation of the PowerPC architecture contains three pipelines and a large on-chip secondary cache to achieve a peak performance of 1600 MIPS. The 15 mm×10 mm die contains 2.7 M transistors (2M CMOS and 0.7 M bipolar) and dissipates less than 85 W. The die is fabricated in a six-level metal, 0.5-μm BiCMOS process and requires 3.6 and 2.1 V power supplies     

AlGaAs vertical-cavity surface-emitting laser responses to 4.5-MeV proton irradiation

We have irradiated single- and multimode AlGaAs vertical-cavity surface-emitting laser (VCSEL) arrays operating at a nominal wavelength of 780 nm with 4.5-MeV protons and doses ranging from 10 to 30 Mrad in the active region. We observed a peak power reduction of about 2% per Mrad in the 14-/spl mu/m aperture, multimode VCSELs. Single-mode VCSELs having an aperture of 6 /spl mu/m exhibited a smaller peak power reduction of 0.4%-1% per Mrad. A slight shift in the current threshold was observed only for the multimode VCSELs at dose levels above 10 Mrad. First results indicate a reduced VCSEL peak laser power output that is dominated by a temperature shift caused by the radiation induced increase in resistive heating. In contrast, the power reduction in edge-emitting lasers is dominated by the enhanced radiation induced nonradiative recombination rate. The VCSEL irradiation was performed with a focused ion micro beam that was rastered over the device surface, ensuring a very uniform exposure of a single device in the array.     

The use of atmospheric pressure MOVPE for the growth of high performance uncooled 1300 nm DFB lasers

Manufacture of high performance uncooled 1300 nm distributed feed-back (DFB) lasers operating single mode over the −40 to +85°C range requires control of the wavelength variation across a 2″ wafer to less than 10nm and preservation of grating definition during processing and regrowth. We have used atmospheric pressure metalorganic vapor phase epitaxy, without substrate rotation to achieve the necessary uniformity. Material was assessed using photoluminescence, x-ray diffraction, transmission electron microscopy, electrochemical current/voltage profiling, and secondary ion mass spectroscopy. The devices are based on a strained quantum well structure with an n-type grating layer to provide gain coupling. The best result gave a wavelength spread across 32×32 mm center square of a 2″ InP wafer of 3 nm. Buried heterostructure DFBs manufactured with high yield in this way operate from −40 to +85°C, with thresholds at 85°C as low as 18 mA.     

Multilevel differential polarization shift keying

We analyze the performance of coherent optical communication systems which employ polarization shift keying (PolSK) modulation, introducing for the first time the requirements for multilevel, differential PolSK. For M-ary signal constellations with M>2, the three-dimensional (3-D) nature of the signal constellations within Stokes space requires the use of double-differential modulation and detection (M-DDPolSK). Coding constraints on the signals place an additional restriction on such systems. We present detailed analysis and simulations for a 6-DDPolSK modulation scheme in which the signal points are located at the vertices of an octahedron in Stokes space     

Quantum-efficiency measurements in oxygen-containing gallium lanthanum sulphide glasses and fibers doped with Pr/sup 3+/

The production of competitive fiber amplifiers in the 1.3-/spl mu/m region requires both good quantum efficiency in the lasing ion and the capability to produce low-loss fibers. Oxygen-doped gallium lanthanum sulphide (GLS) doped with Pr/sup 3+/ may provide a route to both, We describe measurements of the quantum efficiency of Pr/sup 3+/ emission at 1.3 /spl mu/m from the /sup 1/G/sub 4/-/sup 3/H/sub 5/ transition in GLS glass and fiber containing varying quantities of lanthanum oxide. We show that oxide-containing GLS glasses, which are known to have considerably better thermal and glass-forming properties than pure GLS, can show quantum efficiencies of up to 84% of that of pure GLS, No degradation of quantum efficiency is seen when oxide-containing GLS glass is pulled into fiber form.     

Series-system reliability-estimation using very small binomialsamples

This investigation explored the effect of incorporating prior information into series-system reliability estimates, where the inferences are made using very small sets (less than 10 observations) of binomial test-data. To capture this effect, the performance of a set of Bayes interval estimators was compared to that of a set of classical estimators over a wide range of subsystem beta prior-distribution parameters. During a Monte Carlo simulation, the Bayes estimators tended to provide shorter interval estimators when the mean of the prior system-reliability differed from the true reliability by 20 percent of less, but the classical estimators dominated when the difference was greater. Based on these results, the authors conclude that there is no clear advantage to using Bayes interval estimation for sample sizes less than 10 unless the poor mean system reliability is believed to be within 20 percent of the true system reliability. Otherwise, the Lindstrom-Madden estimator, a useful classical alternative for very small samples, should be used     

Gain enhancement of multistage parametric intracavity frequencyconversion

Multistage intracavity frequency conversion in optical parametric oscillators (OPOs) can generate new frequencies efficiently over a large dynamic range of pump intensity. These frequencies can be either higher or lower than the pump frequency. High efficiency is correlated with steady states exhibiting nearly complete pump depletion (CPD). Placing a second OPO crystal with high small-signal gain in the cavity does not affect the steady states with CPD, if the pump is depleted and the idler out-coupled before these fields reach the high-gain crystal. However, the high-gain crystal promotes rapid saturation, and hence higher time-averaged efficiency, in Q-switched devices. Numerical modeling of injection-seeded Q-switched devices in three spatial dimensions suggests that excellent beam quality at high peak power is obtainable. However, instabilities can occur in this modeling, and it is not known if these numerical instabilities entail instabilities in the physical devices     

Generating, capturing and processing supply current signatures from analogue macros in mixed-signal ASICs

Transient Response Testing is a powerful test technique for analogue macros in mixed-signal electronic systems which with some enhancement can be particularly useful for testing deeply buried circuit structures. Supply current testing is finding widespread application in the digital domain and its use in the analogue domain may lead to integrated test methodologies for mixed-signal systems. This paper shows that by utilizing both these techniques, and a low-cost test shell, deeply buried analogue macros can be partitioned, tested using Transient Response Testing and the resulting response accurately captured from the total device supply current. It also contains an analysis of the noise on the supply current, due to digital circuit activity during testing, and demonstrates a test response analysis technique which is insensitive to it.