Unraveling robustness of deep face anti-spoofing models against pixel attacks

Multimedia Tools and Applications - In the last few decades, deep-learning-based face verification and recognition systems have had enormous success in solving complex security problems. However,...     

Modeling evidential databases as possible worlds

This paper is about modeling a possible worlds' form of imperfect databases, where imperfection is handled with the evidence theory. The possible worlds' form is already developed for other types of imperfect databases such as probabilistic and possibilistic databases. Such a form is essential to prove that a compact form querying method is a strong representation system. For the same aim, we propose in this article an extended model of possible worlds's form treating the tuple‐level uncertainty in addition to the attribute‐level uncertainty.     

Maximizing efficiency and linearitry

The need for highly efficient, linear microwave power amplifiers has grown rapidly, especially as mobile and satellite communication systems have been developed and expanded [1]. These systems employ multilevel digital modulation techniques to achieve higher-data-rate and multimedia services. This results in a band-limited multicarrier signal, which is spectrally efficient but exhibits time-varying amplitude and phase characteristics. For acceptable linearity in handling such multicarrier signals with high peak-to-average envelope ratios, the amplifier output power must be backed off from its peak value. This has the disadvantages of reducing amplifier efficiency, increasing energy consumption, and reducing the operating time of portable power sources.     

An accurate on-wafer deembedding technique with application to HBTdevices characterization

An accurate deembedding technique for on-wafer measurements of an active device's S-parameter is presented in this paper. This deembedding technique accounts in a systematic way for effect of all parasitic elements surrounding the device. These parasitic elements are modeled as a four-port network. Closed-form equations are derived for deembedding purposes of this four-port network. The proposed deembedding technique was used to extract small-signal model parameters of a 2×25 μm emitter GaInP/GaAs heterojunction bipolar transistor device, and excellent agreement between measured and model-simulated S-parameter was obtained up to 30 GHz     

Symptomatic Helicobacter pylori infection: prospective study of epidemiological, diagnostic and therapeutic aspects in children in Tunisia]

CASE REPORT: The patient was a boy born in June, 1990. The proband's father had a history of nonspherocytic hemolytic anemia. The patient was anemic at birth (Hb 11.9 g/dl) and had a hemolytic attack on postnatal day 2. His hemolysis became well compensated, and his second hemolytic episode occurred at three years of age. CLINICAL AND LABORATORY FINDINGS: The patient's mental development had so far been normal and he has no neurological symptoms. His only clinical manifestation has been compensated hemolytic anemia with a hemoglobin concentration of about 11.0 g/dl and a reticulocyte count of 3-6%. He was positive on the Heinz body formation test, and target cells were seen on his peripheral blood smear. The osmotic fragility test yielded slightly increased value. Decreased reduced glutathione (GSH) was observed (4.4 mg/dlRBC) (normal range: 63.9 +/- 9.6), and he also had decreased glutathione synthetase (GS) activity of 0.03 U/gHb (0.38 +/- 0.08 U/gHb). A diagnosis of GS deficiency was made. Decreased glutathione S-transferase (GST) activity was also found (0.57 U/gHb) (normal range: 6.65 +/- 1.20). DISCUSSION: GS deficiency has been reported in about 30 families all over the world. This patient was the first Japanese patient with red cell GS deficiency.     

Joint Estimation of MIMO Channel Parameters Using Space–Time Correlation Matrix for Different Angular Distributions

In this paper, we propose a new low-complexity method to jointly estimate the multiple-input multiple-output (MIMO) channel parameters namely the mean angle of arrival (AoA), the angular spread (AS) and the maximum Doppler spread (DS). We consider Gaussian and Laplacian angular distributions for the incoming AoAs in the case of a Rayleigh channel model. Our estimator is based on the magnitudes and phases of the space–time correlation functions of the received signals. To this end, closed-form expressions of the required functions were derived. Two different approaches are studied using these cross-correlation functions; first at a non zero time lag and second at two different time lags. To evaluate the robustness of the proposed estimator, the two Stage approach and the improved maximum likelihood method based on the Gauss Newton algorithm are taken as benchmarks for the mean AoA and the AS estimation. For the maximum DS, the two Rays and the auto-correlation based algorithms are chosen. Simulation results show that the proposed estimator offers more accurate estimates in almost all considered scenarios. We also compare our work to a recent joint estimator which exploits the Derivatives of the cross-correlation function (DCCF). Our method outperforms the DCCF algorithm at a lower computational cost.     

Direct parameter-extraction method for HBT small-signal model

An accurate and broadband method for the direct extraction of heterojunction bipolar transistor (HBT) small-signal model parameters is presented in this paper. This method differs from previous ones by extracting the equivalent-circuit parameters without using special test structures or global numerical optimization techniques. The main advantage of this method is that a unique and physically meaningful set of intrinsic parameters is extracted from the measured S-parameters for the whole frequency range of operation. The extraction procedure uses a set of closed-form expressions derived without any approximation. An equivalent circuit for the HBT under a forward-bias condition is proposed for extraction of access resistances and parasitic inductances. An experimental validation on a GaInP/GaAs HBT device with a 2×25 μm emitter was carried out, and excellent results were obtained up to 30 GHz. The calculated data-fitting residual error for three different bias points over 1-30 GHz was less then 2%