Two-axis-scanning laser Doppler vibrometer for precision engineering

Laser Doppler vibrometer (LVD) has been the most favorite instrument for precision dynamics measurement due to its non-contact, high accuracy and high resolution. However, LDV can only give the dynamic data of a particular location on the entire feature. In order to get the whole field data, a laser beam-scanning mechanism has to be implemented. Currently, motor-driven scanning mirror is used to move the measurement probe from one point to another. The mechanical vibrations of the scanning mirror will reduce the measurement accuracy. This paper introduces a novel scanning LDV optical system embodied in an acousto-optic deflector scanning mechanism. It can improve the measurement accuracy since there is no mechanical motion involved. One main advantage of this system is that it generates a laser scanning beam in parallel that is different from the beam scanning in the conventional scanning laser Doppler vibrometer (SLDV). The new system has a board scanning range. The measurement target size ranges from few tens of millimeters down to 10 μm. We have demonstrated the capability of the novel system on scanning measurements of features as big as ultra-precision cutting tool to features as tiny as AFM cantilever. We believe that the novel SLDV will find profound potential applications in the precision engineering field.     

利用莫尔条纹的准正弦特性的三维轮廓术

分析了两个矩形光栅迭合产生的莫尔条纹的光强分布特性，通过选择适当的光栅参数，可得到一个近似的正弦分划板，并把它用于三维面形测量中，实验结果表明，这种方法简单，易于自动处理，有广泛的实用价值。     

基于移动Agent技术的QoS保障机制

介绍了移动Agent的基本特性。移动Agent可在异构的网络各个节点间自由地移动,它能感知网络的状态,监控系统并与其他Agent进行交互。IETF的IS模型采用了RSVP(ResourcereSerVationProtocol)协议对系统资源进行预留,从而达到一定程度上保障所需QoS的目的。QoS保障机制将移动agent技术与RSVP协议结合起来,通过资源预留和agent动态移动的实现可以达到更好地改善QoS的目的。     

A neural network control system with parallel adaptive enhancementsapplicable to nonlinear servomechanisms

In this paper, we present a technique for using an additional parallel neural network to provide adaptive enhancements to a basic fixed neural network-based nonlinear control system. This proposed parallel adaptive neural network control system is applicable to nonlinear dynamical systems of the type commonly encountered in many practical position control servomechanisms. Properties of the controller are discussed, and it is shown that if Gaussian radial basis function networks are used for the additional parallel neural network, uniformly stable adaptation is assured and the approximation error converges to zero asymptotically. In the paper, the effectiveness of the proposed parallel adaptive neural network control system is demonstrated in real-time implementation experiments for position control in a servomechanism with asymmetrical loading and changes in the load     

Enhancement of oxide break-up by implantation of fluorine inpoly-Si emitter contacted p+-n shallow junctionformation

In this study, it is demonstrated that the incorporation of fluorine can enhance poly-Si/Si interfacial oxide break-up in the poly-Si emitter contacted p⁺-n shallow junction formation. The annealing temperature for breaking up the poly-Si/Si interfacial oxide has been found to be as low as 900°C. As a result, the junction depth of the BF₂-implanted device is much larger than that of the boron-implanted device     

Pd-Ge contact to n-GaAs with the TiW diffusion barrier

Pd-Ge based ohmic contact to n-GaAs with a TiW diffusion barrier was investigated. Electrical analysis as well as Auger electron spectroscopy and the scanning electron microscopy were used to study the contact after it was subjected to different furnace and rapid thermal annealing and different aging steps. All analyses show that TiW can act as a good barrier metal for the Au/Ge/Pd/n-GaAs contact system. A value of 1.45 × 10⁻⁶ Ω-cm² for the specific contact resistance was obtained for the Au/TiW/Ge/Pd/n-GaAs contact after it was rapid thermally annealed at 425°C for 90 s. It can withstand a thermal aging at 350°C for 40 h with its ρ_c increasing to 2.94 × 10⁻⁶Ω-cm² and for an aging at 410°C for 40 h with its ρ_c increasing to 1.38 × 10⁻⁵ Ω-cm².     

Surface leakage currents in SiN/sub x/ passivated AlGaN/GaN HFETs

A novel guarded surface leakage test structure is used to isolate the surface and bulk leakage contributions to gate current in AlGaN/GaN HFETs. Passivation with various recipes of SiN/sub x/ always resulted in the commonly observed increase in gate leakage, which was found to be dominated by bulk leakage through the AlGaN. However, high temperature deposited SiN/sub x/ recipes gave a 1-2 orders reduction in surface leakage, whereas low temperature deposition gave an increase. Gate lag measurements were found to correlate closely with the surface leakage component, giving direct evidence that the key device problem of current slump is associated with current flow at the AlGaN surface.     

Field dependence of impact ionization coefficients in In/sub 0.53/Ga/sub 0.47/As

Electron and hole ionization coefficients in In/sub 0.53/Ga/sub 0.47/As are deduced from mixed carrier avalanche photomultiplication measurements on a series of p-i-n diode layers, eliminating other effects that can lead to an increase in photocurrent with reverse bias. Low field ionization is observed for electrons but not for holes, resulting in a larger ratio of ionization coefficients, even at moderately high electric fields than previously reported. The measured ionization coefficients are marginally lower than those of GaAs for fields above 250 kVcm/sup -1/, supporting reports of slightly higher avalanche breakdown voltages in In/sub 0.53/Ga/sub 0.47/As than in GaAs p-i-n diodes.     

Winograd矩阵乘法算法用于任意阶矩阵时的一种新处理方法

摘要t矩阵乘法StraSsen算法及其变形winograd算法用分而治之的方法把矩阵乘法时间复杂性由传统的D(n。)改进到0(佗kg。n.但是对于奇数阶矩阵,在划分子矩阵时,要作特殊处理才能继续使用此算法.本文提出了一种非等阶“十”字架划分方法,可以最少化填零,最大化性能,使得奇数阶矩阵乘法的时间复杂性更加接近偶数阶矩阵乘法的效果.计算实例显示该方法是有效的.