Low-power performance of 0.5-μm JFET for low-cost MMIC's inpersonal communications

The low-power microwave performance of an enhancement-mode ion-implanted GaAs JFET is reported. A 0.5-μm×100-μm E-JFET with a threshold voltage of V_th=0.3 V achieved a maximum DC transconductance of g_m=489 mS/mm at V _ds=1.5 V and I_ds=18 mA. Operating at 0.5 mW of power with V_ds=0.5 V and I_ds =1 mA, the best device on a 3-in wafer achieved a noise figure of 0.8 dB with an associated gain of 9.6 dB measured at 4 GHz. Across a 3-in wafer the average noise figure was F_min=1.2 dB and the average associated gain was G_a=9.8 dB for 15 devices measured. These results demonstrate that the E-JFET is an excellent choice for low-power personal communication applications     

Cryogenic temperature performance of modulation-doped field-effect transistors

Reports S-parameter measurements of AlInAs/GaInAs/InP modulation-doped field-effect transistors (MODFETs) at cryogenic temperatures. The current gain at 80 K is 3 dB higher than at 300 K, and the current gain cutoff frequency f/sub T/ increases from 32 GHz at 300 K, to 42 GHz at 80 K, which is the first observation of higher f/sub T/ by direct measurement.<>     

Frequency response of sub-micrometre pseudomorphic AlGaAs/InGaAs/GaAs MODFETs at cryogenic temperatures

This paper reports on the d.c. and high frequency characteristics of pseudomorphic AlGaAs/InGaAs/GaAs MODFETs with gate lengths from 1.73 to 0.35 μm at 300 and 110 K. Significant device improvement is shown at 110 K, with current gain cut-off frequency, F_T, increasing by 80% for gate length L_g = 1.73 μm and 15% for L_g = 0.35 μm. The d.c. and high frequency behaviour is analysed and it is shown that electron velocity, not mobility, controls transport in these devices.     

Direct extraction of an empirical temperature-dependent InGaP/GaAs HBT large-signal model

A new empirical InGaP/GaAs heterojunction bipolar transistor (HBT) large-signal model including self-heating effects is presented. The model accounts for the inherent temperature dependence of the device characteristics due to ambient-temperature variation as well as self-heating. The model is accompanied by a simple extraction process, which requires only dc current-voltage (I-V) and multibias-point small-signal S-parameter measurements. All the current-source model parameters, including the self-heating parameters, are directly extracted from measured forward I-V data at different ambient temperatures. The distributed base-collector capacitance and base resistance are extracted from measured S-parameters using a new technique. The extraction procedure is fast, accurate, and inherently minimizes the average squared-error between measured and modeled data, thereby eliminating the need for further optimization following parameter extraction. This modeling methodology is successfully applied to predict the dc, small-signal S-parameter, and output fundamental and harmonic power characteristics of an InGaP/GaAs HBT, over a wide range of temperatures.     

Temperature-dependent bit-error-rate characterization ofultralow-noise GaAs MESFET's for 3-Gb/s operation

Noise generated in a system establishes the fundamental limitation on the performance of all communication systems and can be characterized with both minimum noise figure (NF_min) and bit error rate (BER). The development of data processing and transmission into the gigabit/second range requires a detailed understanding of the correlation between NF_min and BER. The authors report on the cryogenic microwave measurements of NF_min, current gain cutoff frequency (F_t), and BER at 3 Gb/s of 0.6-μm GaAs MESFETs. The noise characterization of GaAs-based MESFET devices and circuits is significant because GaAs-based MESFETs are clearly the key industry device for both digital and analog applications     

Integrated VCO With Up/Down Converter for Si-Based 60 GHz WPAN Applications

This letter presents the design and implementation of the largest reported bandwidth of a 60 GHz up/down converter with an integrated voltage controlled oscillator (VCO) in a low-cost 0.18 mum silicon-germanium process. The up/down conversion is achieved using the 2X sub-harmonic passive mixing with anti-parallel diode pairs. A 30 GHz cross-coupled VCO is designed, optimized and integrated with the sub-harmonic mixer through a cascode amplifier to meet the local oscillator power requirements. The fully integrated chip takes only 1.5 mm² of silicon die area and consumes only 40 mW of dc power for a measured conversion loss of 12 dB at 61.5 GHz. The integrated up/down converter is measured to have greater than 9 GHz double-sided 3-dB RF bandwidth suitable for wideband high data-rate WPAN transceiver requirements. The VCO and VCO-amplifier test structures are separately fabricated and measured to have a phase noise as low as -105 dBc/Hz at 1 MHz offset with a tuning range of 2.3 GHz.     

Music genre recognition using convolutional recurrent neural network architecture

The genre is an abstract feature, but still, it is considered to be one of the important characteristics of music. Genre recognition forms an essential component for a large number of commercial music applications. Most of the existing music genre recognition algorithms are based on manual feature extraction techniques. These extracted features are used to develop a classifier model to identify the genre. However, in many cases, it has been observed that a set of features giving excellent accuracy fails to explain the underlying typical characteristics of music genres. It has also been observed that some of the features provide a satisfactory level of performance on a particular dataset but fail to provide similar performance on other datasets. Hence, each dataset mostly requires manual selection of appropriate acoustic features to achieve an adequate level of performance on it. In this paper, we propose a genre recognition algorithm that uses almost no handcrafted features. The convolutional recurrent neural network‐based model proposed in this study is trained on melspectrogram extracted from 3‐s duration audio clips taken from GTZAN dataset. The proposed model provides an accuracy of 85.36% on 10‐class genre classification. The same model has been trained and tested on 10 genres of MagnaTagATune dataset having 18,476 clips of 29‐s duration. The model has yielded an accuracy of 86.06%. The experimental results suggest that the proposed architecture with melspectrogram as input feature is capable of providing consistent performances across the different datasets     

A new analytical model for switching time of a perforated MEMS switch

In this paper, the design of a low-k meander based MEMS shunt capacitive switch with perforated beam meander has been presented. A closed form analytical model to calculate the switching time of designed structure is proposed. The model is based on modified Mejis and Fokkema’s capacitance model and linearization of non-linear electrostatic force on the switch beam. The model is utilized in evaluating the switching time for uniform as well as non-uniform serpentine meander designs, considering different values of actuation voltage and a wide variation of switching parameters. This work takes into account the beam perforation, fringing field and stiffness effect simultaneously altogether. The results obtained for both the meander designs under every design specifications has been found out to be less than or approximately equal to 100 µs. These model based results are then compared with 3D FEM simulated values. Comparative Analysis indicated that the model results and simulation results are in close agreement with each other.

     

Erratum to: Malaria infected erythrocyte classification based on a hybrid classifier using microscopic images of thin blood smear

Multimedia Tools and Applications -     

Isolation,Amino Acid composition and molecular weight distribution ofEucalyptus kirtoniana seed proteins

Proteins were extracted from the deoiled seeds ofEucalyptus kirtoniana in aqueous solutions of various pHs or by different concentrations of NaCl, Na₂SO₃, CaCl₂ and MgCl₂ at pH 8.0. Amino acid analysis of the isolated protein identified 16 amino acids, nine of which were essential. Gel permeation chromatography on Sephadex G-200 revealed the presence of four components in the protein fraction, and their molecular weights were determined by two comparable standard methods. SDS-PAG electrophoresis demonstrated that each protein isolate from different salt solutions was composed of six fractions whose molecular weights were estimated to be 131,800; 108,300; 93,300; 51,300; 38,000 and 25,700 daltons.