Influence of Photolithography on the Cross-Sectional Shape of Polysiloxane as an Optical Waveguide Material on Printed Circuit Boards

Optical waveguide cross-sectional shapes that deviate from rectangles or squares may cause significant loss of signal. In this study, a photolithography approach was adopted to fabricate waveguides on printed circuit boards, using photo-imageable polysiloxane as a waveguide material. The effects of I-line ultraviolet (UV) lamp exposure, 355-nm Nd:YAG laser direct imaging, and 248-nm excimer laser direct imaging on the cross-sectional shape of waveguides were investigated. For I-line UV lamp exposure, increasing the exposure time could cause changes in the tilt angle of the waveguides from negative (inverted trapezoid) to positive (trapezoid). To obtain rectangular waveguides, the optimum I-line UV lamp exposure time was found to be around 150 s. From the results for 355-nm Nd:YAG laser direct imaging, the width and tilt angle of the waveguides varied with the energy density of the laser beam irradiating the core materials, being controlled by the repetition rate and focus. Lowering the laser energy density could produce waveguides with small widths and tilt angles. Excimer laser direct imaging at 248 nm was found to be unsuitable for waveguide patterning since the core materials could not be cured at this wavelength.     

Z-function convolution in EHV power network electromagnetic transient analysis

Time-convolution in the forms arising in transformation from the frequency-domain to the time-domain has been widely used in the earlier development of electromagnetic transient methods in power systems. Independent work has recently led to new methods based on the z-transform, and the present paper develops methods based on discrete convolution arising from transformation to the time-domain now from the z-plane. The recursive solution sequences to which this leads combines high accuracy with low computing time requirements. Checks and controls in the synthesis of transmission line forward impulse response and surge impedance functions in the z-plane ensure that these are always stable system functions, and that numerical solution procedures which include them have very high inherent stability. The formulations developed are applied to transmission line energization from an equivalent source model, and the electromagnetic transients in short-circuit fault operation. Close comparisons are made between representative solutions from standard time-convolution analysis and from the methods of the present paper.     

Hall effect characterization of LPE HgCdTe P/n heterojunctions

The field and temperature dependence of the Hall coefficient has been used to simultaneously extract information about the p and n layers in very long wave length infrared P/n HgCdTe heterojunctions. The field dependence allows the effects of high mobility electrons to be separated from those of low mobility holes. The higher the magnetic field, the higher the sensitivity to the parameters of the P layer. For a maximum magnetic field of 8000 gauss, the hole sheet concentration must be at least five times the electron sheet concentration to obtain accurate results for the P layer. This criterion is satisfied for typical liquid phase epitaxy (LPE) heterostructures. The analysis determines the hole sheet resistance (concentration times mobility), rather than the hole concentration or mobility separately. Independent knowledge of the P layer thickness and the relationship between hole concentration and resistivity are needed to convert the Hall measurement results to hole concentrations. Analysis of the field-dependent Hall data is complicated by the finding that at least three electrons of different mobilities are needed to fit the field dependence of the Hall coefficient in n-type LPE HgCdTe layers. These results are consistent with previous conclusions that electrons with different mobilities are needed to model bulk n-HgCdTe, and with a range of mobilities in the graded composition interface between the LPE layer and CdTe substrate. Consistent results are obtained for the concentrations and mobilities of the three types of electrons in the n-HgCdTe layer with and without the P layer present. N and P type carrier concentrations are also consistent with dopant concentrations measured by secondary ion mass spectroscopy.     

Concentration-dependent 4I13/2 lifetimes inEr3+-doped fibers and Er3+-doped planar waveguides

We report on the concentration- and pump-dependent lifetimes of the spontaneous emission in Er³⁺-doped fibers and Er³⁺ -doped waveguides. In addition, we measure the concentration dependence of the 550-nm fluorescence due to excited state absorption (ESA)     

Resonance in a superstrate-loaded cylindrical-rectangularmicrostrip structure

The complex resonant frequencies of the cylindrical-rectangular microstrip structure loaded with a dielectric superstrate layer is studied using a rigorous full-wave analysis, and the numerical results are obtained using the Galerkin moment method calculation. The numerical convergence for the selected sinusoidal basis functions with and without the edge singularity condition is also discussed. Numerical results for the dependence of the real and imaginary parts of the complex resonant frequencies on the superstrate permittivity and thickness are calculated and analyzed, and are compared with those obtained for the planar microstrip structure     

Automatic detection of mass-resolved ion conics

A processing algorithm for automatically detecting a specific type of ion distribution (called the ion conic distribution) in data obtained from a space-based mass spectrometer has been devised. Automation of this task is necessary due to the sparseness of conic events within the very large databases typical of space plasma instruments. The algorithm used to perform this automated analysis and the methods used to verify the algorithm are described. Initial results on the characterization of the near-Earth space plasma are summarized     

Bi-2223/Ag-sheathed tape processing studies

The production of high critical current density J_c Bi-2223/Ag sheathed conductors is a complex process involving interactions among many different parameters. The effects of three factors: 1) powder production path, 2) the first sinter temperature, and 3) the subsequent sinter temperatures were investigated. Statistical methods were used to design the experiment and interpret the results. Transport J_c was the main response for the analysis, but microstructural results were also used to assess the physical basis for the differences in performance. The powder variable had the largest main effect with only very weak main effects for the other factors.     

Inverse halftoning and kernel estimation for error diffusion

Two different approaches in the inverse halftoning of error-diffused images are considered. The first approach uses linear filtering and statistical smoothing that reconstructs a gray-scale image from a given error-diffused image. The second approach can be viewed as a projection operation, where one assumes the error diffusion kernel is known, and finds a gray-scale image that will be halftoned into the same binary image. Two projection algorithms, viz., minimum mean square error (MMSE) projection and maximum a posteriori probability (MAP) projection, that differ on the way an inverse quantization step is performed, are developed. Among the filtering and the two projection algorithms, MAP projection provides the best performance for inverse halftoning. Using techniques from adaptive signal processing, we suggest a method for estimating the error diffusion kernel from the given halftone. This means that the projection algorithms can be applied in the inverse halftoning of any error-diffused image without requiring any a priori information on the error diffusion kernel. It is shown that the kernel estimation algorithm combined with MAP projection provide the same performance in inverse halftoning compared to the case where the error diffusion kernel is known.     

On computing undetected error probabilities on the Gilbert channel

Two methods for computing the probability of undetected error on the Gilbert (1960) channel are examined. First, using a method proposed by Kittel (1978), we study some standard cyclic redundancy codes and compare the results with those on the binary symmetric channel. Then we consider a general method of approximate code evaluation, proposed by Elliott, which involves P(m, n), the probability of m errors in a block of length n bits. A nonrecursive technique for computing P(m, n) on the Gilbert channel is described     

Sequential packing algorithm for channel assignment under cochanneland adjacent-channel interference constraint

Generally, the channel-assignment problem (CAP) for mobile cellular systems is solved by graph-coloring algorithms. These algorithms, though sometimes can yield an optimal solution, do not supply any information on whether an optimal solution has been found or bow far away it is from the optimum. In view of these undesirable features, two relevant results are presented. First, a lower bound for the minimum number of channels required to satisfy a given call-traffic demand is derived. This lower bound is tighter than the existing ones under certain conditions and can be used as a supplement for other approximate algorithms. Second, we propose an efficient heuristic algorithm to solve this problem. Although the CAP is nondeterministic polynomial (NP) complete in general, our algorithm provides an optimal solution for a special class of network topologies. For the general case, promising results are obtained, and numerical examples show that our algorithm has a better performance than many existing algorithms