Active RC networks using a parallel T

A method of independent control over the positions of the poles and zeros of an active RC network is described, Secondorder transfer functions can be realised by using these circuits. The adjustments of their characteristics after the construction are very easy. A simple experiment shows good agreement with the theory.     

InGaAs-InP P-I (MQW)-N surface-normal electroabsorption modulatorsexhibiting better than 8:1 contrast ratio for 1.55-μm applicationsgrown by gas-source MBE

We report growth of In_0.53Ga_0.47 As-InP multiple quantum well (MQW) modulators operating at 1.55 μm for fiber-to-the-home applications. By employing a 200-period InGaAs-InP MQW stack in the intrinsic region of a p-i-n structure and working in reflection, we have been able to realize surface-normal modulator devices that exhibit better than an 8:1 contrast ratio. This is the highest contrast ratio reported to date for this type of device working at this wavelength     

Cut-off frequency in radially inhomogeneous single-mode fibre

The limit for single-mode operation in a graded-index fibre has been obtained by calculating the normalised cut-off frequency of the TE01 mode, The affect of diffusion at the core-cladding boundary has been estimated.     

Effect of dip in the referctive index on the cut-off frequency of a single-mode fibre

The normalised frequency for single-mode cut-off in an optical fibre has been calculated as a function of the dip in the refractive index at the centre of the core. For a particular class of profiles, dip width up to 40% have a negligible effect on the propagation characteristics.     

Theoretical analysis of a coated single-mode fiber termination

The antireflection efficiency of a single layer coating in front of an abruptly terminated monomode optical fiber is treated analytically by means of an integral equation, that is appropriate to solve the corresponding boundary value problem. Both guided and radiation modes are taken into account, when describing the field inside the optical fiber region. Then the integral equation is solved approximately by a Neumann series procedure giving highly accurate results when the weakly guidance condition holds, which is a case of practical interest in the optical communications area. The reflection coefficient and the radiation pattern of this configuration are computed. Numerical results are given for several values of the parameters of the geometry under consideration.     

High temperature superconducting transresistance amplifiers for farinfrared detectors

Far-infrared (IR) focal plane arrays are becoming increasingly important for terrestrial and space-based imaging applications. For noise reasons, it is desirable to have transresistance preamplifiers at the focal plane, and, therefore, they need to operate over wide temperature ranges. High-temperature superconducting (HTS) amplifiers that have a high gain, relatively low noise, and response times of <200 ps over at least a 10-80-K temperature range have been developed     

Optical properties of undoped n-AlGaN/GaN superlattices as affected by built-in and external-electric field and by ar-implantation-induced partial disordering

High-resolution x-ray diffraction patterns and 90 K microcathodoluminescence (MCL) spectra were taken for undoped, symmetric AlGaN/GaN superlattices (SLs) with GaN quantum-well (QW) widths of 35 Å and 80 Å. The short-period SL spectra were blue shifted by about 60 meV compared to the GaN substrate, and the magnitude of the blue shift was increased by about 20 meV by application of a reverse bias of ?3 V (electric field of about 4 · 10⁵ V/cm) to a Schottky diode prepared on this SL. A small red shift of about 40 meV compared to GaN was observed for the long-period SL. The two latter observations were interpreted as manifestations of the presence of a strong built-in piezoelectric field, giving rise to the quantum-confined Stark effect (QCSE). Partial disordering of the short-period SL was observed after Ar ion implantation (energy 150 keV, dose 8·10¹³ cm^?2 and 80 keV, 2·10¹³ cm^?2) and subsequent annealing at 1000°C for 3 h under the protective layer of Si₃N₄. However, it was observed that this partial disordering was accompanied by strain relaxation via formation of misfit dislocations or cracks.     

Noise analysis of injection-locked semiconductor injection lasers

The noise of injection-locked semiconductor lasers is analyzed by rate equations including the spontaneous emission noise. The side mode suppression and the relative intensity noise (RIN) of the locked laser (slave laser) are given for different wavelengths detuning between the master and slave laser and for different linewidth enhancement factors α. For large α, locking is difficult to achieve, whereas extremely low noise may be obtained for injection-locked lasers with a low linewidth enhancement factor.     

PLL-based BiCMOS on-chip clock generator for very high-speedmicroprocessor

Described is a phase-locked loop (PLL)-based BiCMOS on-chip clock generator (PCG), which is used to generate an internal clock synchronized to a reference clock from outside the chip. In order to obtain a very wide operation bandwidth, it is proposed that the PCG include a compensation circuit for voltage-controlled oscillator (VCO) operation. The compensation circuit varies the oscillation bandwidth of the VCO according to the reference clock frequency, preventing the expected oscillation frequency from being outside the oscillation bandwidth. The PCG is designed and fabricated with 1.0 μm BiCMOS technology, and it achieves an operation bandwidth of 3 to 90 MHz     

Feature-based detection of the K-complex wave in the humanelectroencephalogram using neural networks

The main difficulties in reliable automated detection of the K-complex wave in EEG are its close similarity to other waves and the lack of specific characterization criteria. We present a feature-based detection approach using neural networks that provides good agreement with visual K-complex recognition: a sensitivity of 90% is obtained with about 8% false positives. The respective contribution of the features and that of the neural network is demonstrated by comparing the results to those obtained with i) raw EEG data presented to neural networks, and ii) features presented to Fisher's linear discriminant.