A low-power readout circuit for nanowire based hydrogen sensor

This paper presents a fully integrated lock-in amplifier intended for nanowire gas sensing. The nanowire will change its conductivity according to the concentration of an absorbing gas. To ensure an accurate nanowire impedance measurement, a lock-in technique is implemented to attenuate the low frequency noise and offset by synchronous demodulation or phase-sensitive detection (PSD). The dual-channel lock-in amplifier also provides both resistive and capacitive information of the nanowire in separate channels. Measurement results of test resistors and capacitors show a 2% resolution in the resistance range 10-40 kΩ and a 3% resolution in the capacitance range 0.5-1.8 nF. Moreover, a 28.7-32.1 kΩ impedance variation was measured through the lock-in amplifier for a single palladium nanowire that was exposed to a decreasing hydrogen concentration (10% H₂ in N₂ to air). The chip has been implemented with UMC 0.18 μm CMOS technology and occupies an area of 2 mm². The power consumption of the readout circuit is 2 mW from a 1.8 V supply.     

A comparison of MOSFETs aging under d.c., a.c. and alternating stress conditions

The effects of static, alternating and dynamic stress conditions on the degradation rate of 0.8μm n- and p-channel LDD MOS transistors have systematically been investigated and compared. The shifts of the threshold voltage, transconductance, linear drain current and charge pumping current were used to monitor the transistor degradation. The results suggest that the aging induced by a dynamic stress cannot be directly explained with static stress models, essentially because it is highly dependent on a larger number of parameters (biases and durations of the top and bottom levels of the pulses, transient times). The correlation of static and dynamic stresses also clarifies the degradation mechanisms, in particular the role of hot holes in the generation of interface states.     

A survey on industrial communication networks

Industrial communication networks differ from other types of networks because these applications are responsible for the control and monitoring of a physical process. In this paper, we will summarise the requirements that users specify for these networks, present the state of art on field buses, factory and cell communication networks and indicate current research directions.     

Occurrence and significance of anomalous chloride waters at the Orakei Korako geothermal field, Taupo Volcanic Zone, New Zealand

Water with Cl concentrations from 15 to almost 10,000 mg/kg, and molecular SO₄/Cl ratios ranging from 0.003 to 1.87, drips periodically from the roof of Ruatapu cave and a side chamber, Rahu Rahu, located in the Orakei Korako geothermal field, Taupo Volcanic Zone, New Zealand. Pools in the bottom of both Ruatapu and Rahu Rahu contain sulfate–chloride waters with pH values ranging from 2.5 to 3.0; their Cl contents have varied temporally from 120 to 240 mg/kg and their molecular SO₄/Cl ratios from 0.86 to 1.30. The Cl in the water dripping from the cave roof cannot come directly from the alkali chloride–bicarbonate water that circulates in the reservoir at Orakei Korako since the modern and historic piezometric surfaces are several meters below the cave roof. Nor does all the Cl in the cave pool waters derive from the reservoir fluid as the volume input required is incompatible with their Na and K contents. A more likely source for the Cl is one whereby rain water, percolating through the fractured rhyolitic country rock, dissolves Cl present either in glass shards or halite deposited in prehistoric times when trapped alkali chloride water boiled to dryness. Given that Cl in the cave pool waters is therefore supplied from a source above rather than, as previously assumed, below, the axiom that Cl present in acid sulfate–chloride ± bicarbonate waters is necessarily a signature of a deep water or magmatic input needs qualification.     

Un nouveau matériau pour le multiplexage holographique sur réseau de fibres optiques

Frequency multiplexing, using gratings, in optical fiber systems is now well known. The potentials of holography have already been validated for this kind of application. However, the silver halide photosensitive materials limited the diffraction efficiency and also the achievable angular discrimination. The material we have developed, called optrigelac, is a gelatin doped with ammonium dichromate. This material allows high diffraction efficiencies, up to the theoretical limit of 100 %, with low noise and adjustable angular and wavelength selectivity. The coating, sensitizing and development procedures, and a model relating the final hologram properties to a practical working procedure are described.     

Quantification et détection

The basic detection problem consists in deciding on an optimal way between two possible hypotheses. This decision is made by comparing to an appropriate threshold the output of a processing system. The purpose of this paper is to study the consequences of a threshold quantization on the detection performance. Using various possible models of quantization we show that it creates in general a degradation of the performance. From this fact we introduce a new detection criterion using the concept of contrast and we show on some examples how this criterion is adapted to quantized detection problems.     

Les dispositifs de lutte contre les effets des évanouissements sélectifs dans les faisceaux hertziens numériques

An overview of the countermeasure techniques used in digital radio systems for combatting the effects of selective fading due to multipath propagation is presented. Basic principles are analyzed. Their performances, based upon experimental measurements carried out throughout the world are compared. Frequency diversity techniques and channel interference cancellation techniques specific to a multichannel environment are not investigated. Future trends in the field are finally discussed.     

Un algorithme précis de synthèse en échelle

An algebraic approach to the problem of LC ladder filter synthesis is presented by use of a polynomial formalism introduced first for polynomial low-pass filters and then generalized to the case of filters with infinite peaks of attenuation. An algorithm based upon this formalism is given which provides a much better accuracy than a continued fraction expansion. Numerical results are given.