Effects of process errors on the diffraction characteristics of binary dielectric gratings

The effects of fabrication errors on the predicted performance of surface-relief phase gratings are analyzed with a rigorous vector diffraction technique. For binary elements, errors in the dimensions of the profile [depth, linewidth (fill factor), and grating period], as well as errors in the shape of the profile, are investigated. It is shown that the dimension errors do not have a significant effect on grating performance when the grating is designed for either maximum or minimum diffraction efficiency. A trapezoid is used to model the shape error of the profile. For the first time, design rules that significantly reduce the effects of any shape error are presented.     

User productivity as a function of AutoCAD interface design

Increased operator productivity is a desired outcome of user-CAD interaction scenarios. Two objectives of this research were to (1) define a measure of operator productivity and (2) empirically investigate the potential effects of CAD interface design on operator productivity, where productivity is defined as the percentage of a drawing session correctly completed per unit time. Here, AutoCAD provides the CAD environment of interest. Productivity with respect to two AutoCAD interface designs (menu, template) and three task types (draw, dimension, display) was investigated. Analysis of user productivity data revealed significantly higher productivity under the menu interface condition than under the template interface condition. A significant effect of task type was also discovered, where user productivity under display tasks was higher than productivity under the draw and dimension tasks. Implications of these results are presented.     

SAF-CNN：面向嵌入式FPGA的卷积神经网络稀疏化加速框架

传统的卷积神经网络加速器及推理框架在资源约束的FPGA上部署模型时,往往面临设备种类繁多且资源极端受限、数据带宽利用不充分、算子操作类型复杂难以适配且调度不合理等诸多挑战.提出一种面向嵌入式FPGA的卷积神经网络稀疏化加速框架（sparse acceleration framework of convolutional neural network, SAF-CNN）,通过软硬件协同设计的方法,从硬件加速器与软件推理框架2个角度进行联合优化.首先, SAF-CNN构建并行计算阵列,并且设计并行编解码方案,实现单周期多数据的传输,有效减少通信代价.其次,设计细粒度结构化块划分剪枝算法,于输入通道维度进行块内裁剪来获得稀疏且规则的权重矩阵,借此显著降低计算规模和DSP乘法器等资源占用.然后,提出一种兼容深度可分离卷积的输入通道维度动态拓展及运行时调度策略,实现输入通道参数灵活适配与逐通道卷积和逐点卷积的资源复用.最后,提出一种计算图重构及硬件算子融合优化方法,提升硬件执行效率.实验采用2种资源受限的低端FPGA异构平台Intel CycloneV与Xilinx ZU3EG,结果表明SAF-...     

Photon density waves scattered from cylindrical inhomogeneities: theory and experiments

We present an analytical solution for the scattering of diffuse photon density waves from an infinite circular, cylindrical inhomogeneity embedded in a homogeneous highly scattering turbid medium. The analytical solution, based on the diffusion approximation of the Boltzmann transport equation, represents the contribution of the cylindrical inhomogeneity as a series of modified Bessel functions integrated from zero to infinity and weighted by different angular dependencies. This series is truncated at the desired precision, similar to the Mie theory. We introduce new boundary conditions that account for specular reflections at the interface between the background medium and the cylindrical inhomogeneity. These new boundary conditions allow the separate recovery of the index of refraction of an object from its absorption and reduced scattering coefficients. The analytical solution is compared with data obtained experimentally to evaluate the predictive capability of the model. Optical properties of known cylindrical objects are recovered accurately. However, as the radius of the cylinder decreases, the required measurement signal-to-noise ratiorapidly increases. Because of the new boundary conditions, an upperlimit can be placed on the recovered size of cylindrical objects with radii below 0.3 cm if they have a substantially different index of refraction from that of the background medium.     

A Simple, rapid preparative method for isolating and purifying oxymyoglobin

A simple and rapid procedure was developed to isolate and purify preparative quantities (up to 200 mg) of reduced myoglobin (98% oxy) of high purity (>96%) from beef and pork muscles. This method involved fractional precipitation of a crude myoglobin extract with ammonium sulfate and purification with a single Chromatographic step on a Sephadex G-100 column. The metmyoglobin level of the myoglobin preparation was minimised by using muscle from freshly slaughtered animals (<48h post-mortem), trimming the muscle immediately before use to remove any oxidised myoglobin on the surface and carrying out all procedures at low temperature (0-5 °C) and alkaline pH (8.0-8.5). The purity of the myoglobin preparations was confirmed by ion exchange HPLC and SDS-polyacrylamide gel electrophoresis. The method was effective for skeletal muscle of both low myoglobin content (pork) and high myoglobin content (beef). The resulting purified myoglobin was very stable and there was little change in metmyoglobin level or autoxidation rate during 3 months' storage at 0 °C.     

Fractality of speckle intensity fluctuations

Coherent-light diffraction on random phase screens with fractal properties leads to the formation of speckle patterns with peculiarities in correlation characteristics in the small-scale region. Such peculiarities are manifested in asymptotic behavior in intensity autocorrelation and structure functions in the vicinity of the zero values of their arguments. Intensity fluctuations in the far and the near diffraction zones are also characterized by values of fractal (Hausdorff-Besicovitch) dimensions D(HB), differing from the corresponding Euclidean dimension. Relationships between the exponential factors of the structure functions of boundary field phase and scattered-light intensity fluctuations as well as between values of D(HB) have been obtained as a result of speckle-formation analysis for different conditions. Their dependencies on the illumination and observation conditions obtained in experiments with fractallike scatterers (rough glass plates) are in satisfactory agreement with theoretical results.     

Detection and confirmation of N-nitrosodialkylamines using liquid chromatography-electrospray ionization coupled on-line with a photolysis reactor

Simultaneous detection and confirmation of several N-nitrosodialkylamines are accomplished by on-line coupling of a photolysis reactor with an HPLC-electrospray ionization mass spectrometer. Several parameters such as irradiation wavelength, irradiation time, mobile-phase composition, and pH, as well as different organic acid modifiers are investigated, and their impact on the detection of the N-nitrosodialkylamine-acid complex and its dissociative photolysis products is presented here. Additionally, the type of structural information obtained from the photolytic processes of N-nitrosodialkylamines is compared to that obtained by using in-source collision-induced dissociation. To demonstrate the potential of this technique, six N-nitrosodialkylamines are studied to determine the linearity of the response, the limits of detection and confirmation, and the reproducibility. The technique's versatility is also exhibited by utilizing negative-ion mode as a complementary means for analysis of the compounds. Finally, an illustrative application for N-nitrosodimethylamine analysis in beer is described.