Treatment of municipal wastewater UASB reactor effluent by unconventional flotation and UV disinfection.

Post-treatment of an UASB reactor effluent, fed with domestic sewage, was conducted using two-stage flotation and UV disinfection. Results were compared to those obtained in a parallel stabilisation pond. The first flotation stage employed 5 - 7.5 mg L(-1) cationic flocculant to separate off more than 99% of the suspended solids. Then, phosphate ions were completely recovered using carrier flotation with 5-25 mg L(-1) of Fe (FeCl3) at pH 6.3-7.0. This staged flotation led to high recoveries of water and allowed us to separate organic matter and phosphate bearing sludge. The water still contained about 1 x 10(2) NMP/100 mL total coliforms, which were removed using UV radiation to below detection levels. Final water turbidity was < 1.0 NTU, COD < 20 mg L(-1) O2 and 71 mNm(-1), the liquid/air interfacial tension. This flotation-UV flowsheet was found to be more efficient than the treatment in the stabilisation pond and appears to have some potential for water reuse. Results were discussed in terms of the biological, chemical and physicochemical mechanisms involved.     

MOS RF reliability subject to dynamic voltage stress-modeling and analysis

Dynamic stress on MOSFETs with 900-MHz inverter-like waveforms as well as static (or dc) stress were evaluated experimentally. It showed that the degradation due to dynamic stress is less than that of dc stress for our test transistors. A compact model is used to evaluate the degradation in radio frequency performances, such as transconductance, cutoff frequency, linearity, and noise figure. A class-AB power amplifier is presented as an example to demonstrate the effect of dynamic stress on RF circuit performance.     

Improved Estimates for the Accuracy of Small Disjuncts

Learning systems often describe a target class as a disjunction of conjunctions of conditions. Recent work has noted that small disjuncts, i.e., those supported by few training examples, typically have poor predictive accuracy. One model of this accuracy is provided by the Bayes-Laplace formula based on the number of training examples covered by the disjunct and the number of them belonging to the target class. However, experiments show that small disjuncts associated with target classes of different relative frequencies tend to have different error rates. This note defines the context of a disjunct as the set of training examples that fail to satisfy at most one of its conditions. An empirical adaptation of the Bayes-Laplace formula is presented that also makes use of the relative frequency of the target class in this context. Trials are reported comparing the performance of the original formula and the adaptation in six learning tasks.     

8-Gbit/s transmission experiment over 68 km of optical fiber using a Ti:LiNbO3external modulator

We describe the performance of an experimental 1.5-μm lightwave transmission system operating at 8 Gbit/s over 68.3 km of single-mode fiber. The dispersion penalty is limited to 1 dB through the use of external modulation and is attributable to the intrinsic information bandwidth.     

Hydrological simulations in the Rhine basin.

Simulations with regional climate models (RCMs), carried out for the Rhine basin, have been analyzed in the context of implications of the possible future discharge of the Rhine river. In a first analysis, the runoff generated by the RCMs is compared to observations, in order to detect the way the RCMs treat anomalies in precipitation in their land surface component. A second analysis is devoted to the frequency distribution of area averaged precipitation, and the impact of selection of various driving global climate models.     

Stabilization of hyperbolic systems using concentrated sensors and actuators

Certain hyperbolic systems of partial differential equations which are known to be uniformly asymptotically stabilizable using point sensors/actuators (S/A) are considered. The issue to be investigated is the effect on stability when point S/A's are replaced by "concentrated" S/ A's, that is, S/A's which average over small regions of the spatial domain. Although it is known that passing from point to concentrated S/ A's necessarily destroys uniform stability, a necessary and sufficient condition for strong stability is obtained in terms of the S/A weighting functions. In addition, in the special case of a cantilevered beam controlled by a single sensor/actuator pair concentrated at the free end, another, more robust type of stability is shown to hold, even when strong stability does not. The latter result shows that the system energy is bounded by a part which goes uniformly to zero at infinity and a residual which can be explicitly estimated in terms of the support of the weight functions and the initial energy. Furthermore, the residual energy converges to zero as the support reduces to the point at the free end of the beam.     

Semiconductor optical amplifier for CWDM operating over 1540-1620 nm

A semiconductor optical amplifier was developed for coarse wavelength-division-multiplexing (CWDM) operating over 1540-1620 nm (C-L band). A unique quantum-well structure was designed to meet the requirements for the CWDM operation such as wide bandwidth, low polarization-dependent gain, and high-saturation power at the short wavelength end of the band (1540 nm). Over the band, 24-dB maximum chip gain was obtained with less than 4.3-dB gain flatness and more than 14.6-dBm saturation power.