基于随机森林算法的酒精浓度在线测量系统

酒精度是酒类生产过程中重要的技术指标，目前对于酒精度的测量方法过程繁琐，耗时较长，容易污染浪费，本文提出一种基于微波测量和随机森林算法的在线酒精浓度测量系统。首先基于传输反射法和有限元法(FEM)设计双端口测量装置，优化腔体结构，使S₁₁和S₂₁与酒精浓度呈单调变化；然后基于FEM算法计算腔体内不同介电常数的S₁₁和S₂₁,得到4081组训练数据；基于随机森林算法，训练反演网络；最后引入非线性回归校准算法，依据浓度与介电实部的关系获得待测样品的酒精浓度，测试准确度高达99%。本系统结构简单，可以用于酒精浓度的非接触式在线测量，测量精度高，避免酒精的污染浪费，有效地提高了生产效率。     

基于CMOS的新型人工表面等离激元滤波器设计

本文基于0.18μm CMOS工艺提出了一种新型微波人工表面等离激元(Spoof Surface Plasmon Polariton,SSPP)耦合式滤波器结构并进行加工和测量，同时设计了一种太赫兹SSPP四边形滤波器并进行了全波仿真.新型微波SSPP耦合式滤波器通带为11～12.3 GHz(S₁₁<-10 d B,S₂₁>-3.5 d B)，结构紧凑，电尺寸仅为0.018 4λ_g×0.008 4λ_g，远小于其他基于集成电路工艺设计的无源滤波器.通过优化与调整，可以将新型微波SSPP耦合式滤波器的性能进一步优化到要求范围内.太赫兹SSPP四边形滤波器通带为210.8～241.3 GHz(S₁₁<-10 d B,S₂₁>-4.7 d B)，带内插入损耗仅为2.7 d B，带外抑制良好.两种SSPP滤波器均采用非接触式电磁耦合的新型能量传递方式，结构设计新颖，并且微波段SSPP耦合式滤波器小型化优势明显，电尺寸仅为0.019λ_...     

一种小型化超宽带MIMO天线设计

提出了一种基于槽天线的小型化、高隔离度的超宽带(Ultra Wideband, UWB)多入多出(Multiple-Input Multiple-Output, MIMO)天线.该MIMO天线由两个槽天线单元构成, 为了增加天线阻抗带宽, 每个槽天线单元由末端带有圆形贴片的微带线和末端为圆形的槽线两部分耦合馈电.采用在地板上开槽和方向图分集方法, 减少地板表面波和空中电磁波影响, 达到提高天线隔离度的目的.数值仿真和实验结果表明:该天线在3.1~11 GHz频段内满足端口反射系数|S₁₁| < -10 dB, 隔离度|S₁₂|在7~11 GHz频段内小于-25 dB, 在3.1~7 GHz频段内小于-16 dB, 并根据仿真和测试S参数计算了包络相关系数.     

一种新颖的宽频共缆微带分路器设计

随着有源相控阵天线技术的发展,天线设备同系统中的收发终端需要通过采用各类射频、数字和电源类型的线缆进行互联,采用宽频共缆传输技术可以有效降低工程成本和系统重量。设计了一种宽频共缆微带分路器,通过采用传输线的多级阻抗变换原理,在不用额外的滤波电路情况下,在射频工作频段内,总输入端口的电压驻波比小于1.8,通往射频分端口2的S₂₁大于-0.6 dB,通往控制分端口3和电流分端口4的S₃₁小于-45 dB和S₄₁小于-105 dB,实现了良好的隔离度和传输效果。该分路器具有设计简单、隔离度高、信号能量损耗小和易于调试的优点。     

基于人工磁导体的芯片内/芯片间无线互连单极子天线传输特性研究

提出了一种新的用于芯片内/芯片间无线通信系统中的基于人工磁导体（Artificial Magnetic Conductor,AMC）结构的单极子管脚天线阵列模型.印刷电路板（Printed Circuit Board,PCB）采用厚度3mm的FR4介质,形状为边长50mm正方形.4个单极子天线分布在PCB基板中心15mm×15mm的正方形各顶点上,代表4个芯片的4个管脚.天线为铜材质,长度2.6mm,直径1.5mm.在PCB介质中嵌入了一个周期性人工磁导体铜质平面,4个单极子天线形成一个4端口网络.在仿真基础上,进了实物加工测试.实测结果表明,具有AMC的天线阵列回波损耗（S₁₁）-10dB频带为13.02GHz~15.73GHz;在14.22GHz处,S₁₁的幅度达到最小值-27.25dB,S₂₁、S₃₁、S₄₁分别为-26.26dB、-19.23dB、-21.14dB.与不含AMC结构的天线阵列相比,S参数得到了有效改善,其中S₁₁改善了约3.63dB,S₂₁、S₃₁、S₄₁分别提高了2.05dB、7.21dB、5.28dB.验证了在PCB介质中嵌入AMC结构,可以有效提高单极子天线间的电压传输系数,增加信号的功率传输增益.     

太赫兹矩形波导与共面波导耦合结构设计

太赫兹（THz）波位于微波与红外光波之间,现有微波和光波段波导技术应用正在向THz波段拓展。但是,由于水汽对THz波的强吸收及制造工艺等原因,THz器件主要是平面结构,而THz源及其传输需要用矩形波导。因此,矩形波导与共面波导之间的转换结构成为决定元件和系统性能的关键部分。该设计利用脊波导进行阻抗匹配及电磁场模式转换,实现THz波矩形波导到共面波导的高效率耦合。结果表明,在0.2~0.4 THz频段内,该转换结构的传输系数(S₂₁)高于?3 dB,可以对THz电磁场进行高效率转换。该结果可用于太赫兹分子探测、太赫兹通信等领域,为0.2 THz以上太赫兹的模式转换提供了一种可行方案。     

收扩喷管设计对双S形二元排气系统气动与红外特征的影响

为了研究双S形二元收扩排气系统的气动与红外辐射特性,基于一个基准加力型轴对称排气系统,设计了三种带正尾向全遮挡收扩喷管的双S形二元排气系统模型,通过数值模拟研究了喉道与出口的中心线偏径比差,(S₈?S₉)/D=0.26~0.3,和喉道至出口的宽度扩张比,(W₉?W₈)/D=0.1~0.36,对排气系统气动与红外特征的影响。结果表明:所设计的三种双S形二元收扩排气系统,相比基准轴对称排气系统,在尾向0°~15°角域内红外辐射强度平均降幅在73.4%以上,在上方、下方和侧方90°探测角降幅在60.3%以上,红外辐射强度降幅随(S₈?S₉)/D的减小而上升,随(W₉?W₈)/D的增大而上升,且对(S₈?S₉)/D的敏感较高。三种排气系统的推力系数随(S₈?S₉)/D与(W₉?W₈)/D的降低而上升。     

工艺优先的低成本太赫兹模块研制技术

基于探针设计、性能评价和互连方式对模块研制成本的影响，提出了一种W波段单面图形探针设计，在75～110 GHz频带内，实现S₁₁参数小于-20 dB，S₂₁参数小于-0.05 dB；提出了一种利用冷热源响应的波形图的性能优劣评价方法，提高模块的调试效率；还提出了一种传输线T结构设计，实现了在基板与芯片（或基板）拼接缝隙大于30 μm的情况下，明显改善了W波段高频传输通道的性能指标。根据仿真数据和实物装配后的性能测试验证结果，实现了低成本高性能的太赫兹模块研制。     

引入射频诱导效应的BCT PD-SOI MOSFET建模技术研究

为了精确模拟高漏源电压(V_DS)条件下体接触(Body Contact, BCT)部分耗尽型(Partial Depletion, PD)绝缘体上硅(Silicon-on-Insulator, SOI)MOSFET器件的异常射频诱导效应，建立了一个基于碰撞电离和寄生BJT机理的RL网络模型。首先分析了SCBE模型的体区输出导纳参数，并利用该参数的解析式推导出一种电阻与电感串联的网络拓扑；然后基于直接提取法准确提取RL网络模型的参数；最后，通过仿真得出S₂₁、S₂₂参数分别在史密斯圆图的下半圆和上半圆按顺时针旋转的现象，同时在0.01～20 GHz范围内模型模拟的S参数与实测的S参数的相对误差为2.1%,验证了RL网络模型的有效性和准确性。     

基于新型蝶形单元结构人工表面等离子体激元低通陷波滤波器的设计

为降低滤波器的插入损耗以及实现滤波器的小型化,该文提出一种新型具有陷波功能的人工表面等离子体激元(SSPPs)低通滤波器,该滤波器主要由新型蝶形单元结构、过渡结构以及用于实现陷波功能的叉指电容环路谐振器(IDCLLR)结构组成。新型蝶形单元结构是由一个椭圆形贴片向左右方向旋转30°构成,经过镂空处理后可以显著降低插入损耗,相比传统的矩形和椭圆形结构具有更好的色散特性,大大提高了滤波器的带内的平坦度和带外抑制能力。该文对矩形、椭圆形、梯形以及新型蝶形等不同单元结构的色散曲线进行了分析,并仿真分析了滤波器的S₂₁和S₁₁曲线,验证了新型蝶形单元结构在色散特性、插入损耗、低截止频率和带外抑制方面所具有的优势。最后,对该滤波器进行了加工和测试,测试结果表明,该滤波器仿真结果和测试结果吻合较好,表现出较好的带外抑制和带内平坦度,可以实现对特定干扰频段陷波抑制。滤波器尺寸为0.98λ₀×0.17λ₀。该人工表面等离子体激元滤波器从设计新型单元结构的角度出发,实现了良好性能的同时,实现了滤波器的小型化。     

Development of Measurement and Extraction Technique of Complex Permittivity Using Transmission Parameter <Emphasis Type="Italic">S</Emphasis><Subscript>21</Subscript> for Millimeter Wave Frequencies

This study provides an overview of measured S-parameters and its processing to extract the dielectric properties of materials such as Teflon, PMMA, and PVC which are preferred for materials characterization process. In addition, a correction model is presented for transmission parameter (S ₂₁) to obtain the dielectric constant with high accuracy. A non-destructive and non-contact free space measurement method has been used to measure S-parameters of thin samples in the low THz frequency range. S-parameters are measured in free space by vector network analyzer supported with two frequency extenders. Additionally, the parabolic mirrors are used to collimate the generated beam in wide frequency range. Furthermore, a standard filter process is performed to remove the undesired ripples in signal using singular spectrum analyzer before the implementation of extraction process. Newton-Raphson extraction technique is used to extract the material complex permittivity as a function of the frequency in Y-band (325–500 GHz).     

Development of a permittivity extraction method for ultra low k dielectrics integrated in advanced interconnects

An extraction method to determine the permittivity of ultra low k (ULK) dielectrics on real integrated structures is presented. It is a two-step method based on a comparison between measured and simulated capacitance. A best-estimate value of the k_ULK value is first extracted with optimization software coupled to capacitance extraction software. Secondly, uncertainties on material and process parameters are considered to determine an error margin on the best-estimate extracted k value. The uncertainty on the best-estimate value is approximated by a function of the uncertainties on material and process variables. This function is calculated using a multi-linear approximation model and a numerical design of experiments. The same method is applied for the extraction of a ULK material k value (k_ULK) value and an effective k value (k_eff) but with two different simulation structures. In the simulation structure used for k_eff extraction, an equivalent dielectric layer including the ULK layer, the etch-stop and capping layers is used. This method was applied to metal 1 single damascene structures. First results of extraction are presented for two different ULK dielectrics. With the estimated uncertainty used for the parameters in this work, the uncertainties obtained for the best-estimate value of k_ULK and k_effective are significant. Due to the linearity of the model, the method is still applicable with different values for parameters uncertainty. An analysis work will be realized to improve the parameters uncertainty estimation. Future work will also include extraction of ULK permittivity for more complex structures like double damascene structures.     

Adaptive neuro-fuzzy inference system for gap discontinuities in coplanar waveguides

A new approach based on an adaptive neuro-fuzzy inference system (ANFIS) is presented for gap discontinuities in coplanar waveguides (CPWs). The proposed ANFIS model combines the neural network adaptive capabilities and the fuzzy qualitative approach. The ANFIS is presented so as to produce a good approximation to the nonlinear relationship between the geometrical parameters and the frequency-dependent equivalent circuit parameter (the S ₂₁ parameter of the gap). The ANFIS results for the S ₂₁ parameters are compared with the results available in the literature obtained by using the conformal-mapping technique (CMT), and the results confirm that the proposed ANFIS model can provide an accurate computation of the S ₂₁ parameters of the gaps in CPWs.     

New Measuring Apparatus for Complex Permittivity at Millimeter-Wave Frequencies

A dielectric rod resonator excited by a nonradiative dielectric waveguide is used for measuring complex permittivity of low loss dielectric materials. The complex permittivties of single crystal sapphire, polycrystalline Ba (Mg_1/2 W_1/2) O₃ and Mg₂ Al₄ Si₅ O₁₈ (cordierite) have been obtained at 60 and 77 GHz by the new apparatus. The first time the measurement results of complex permittivity of brain grey and white matters from 15 to 50GHz utilizing a two-port microstrip test fixture is presented. S-parameters of Test fixture are simulated employing the finite-element method. A new spectrometer for the precision measurement of dielectric permittivity and loss tangent, which is capable of providing high resolution data for the first time over an extended W-band (68-118 GHz) frequency for specimens with a large range of absorption values, including highly absorbing specimens that otherwise would not be possible.     

High-Temperature Thermoelectric and Microstructural Characteristics of Cobalt-Based Oxides with Ga Substituted on the Co-Site

The effects of Ga substitution on the Co-site on the high-temperature thermoelectric properties and microstructure are investigated for the misfitlayered Ca₃Co₄O₉ and the complex perovskite-related Sr₃RECo₄O_10.5 (RE = rare earth) cobalt-based oxides. For both systems, substitution of Ga for Co results in a simultaneous increase in the Seebeck coefficient (S) and the electrical conductivity (σ), and the influence is more significant in the high temperature region. The power factor (S ² σ) is thereby remarkably improved by Ga substitution, particularly at high temperatures. Texture factor calculations using x-ray diffraction pattern data for pressed and powder samples reveal that the Ga-doped samples are highly textured. Microstructure observed by scanning electron microscopy shows very well-crystallized grains for the samples with Ga substitution for Co. Among the Ga-doped samples, Ca₃Co_3.95Ga_0.05O₉ shows the best ZT value of 0.45 at 1200 K, which is about 87.5% higher than the nondoped one, a considerable improvement.     

Effects of electromagnetic near-field stress on SiGe HBT’s reliability

In this paper, a reliability damage mechanism was presented in SiGe Heterojunction Bipolar Transistor (HBT). This new stress methodology differs from conventional SiGe, HBT device reliability associated with other stresses, since it was obtained by applying electromagnetic near-field aggression. The near-field set-up is used to disturb with electromagnetic field the Device Under test (DUT) on a localized area. Degradations in the base current and the current gain are identified. They are induced by a large base current leakage due to hot carrier which introduces generation/recombination trap centers at the silicon–oxide interface of the emitter–base spacer. By using the S-parameters measurements, we find that both forward transmission scattering parameter (S₂₁) and the input scattering parameter (S₁₁) are affected by this stress. In addition the power characteristics of DUT are also affected by stress. A Direct Power Injection (DPI) method is used to understand the near-field stress behaviour.     

Investigation of the Dielectric Permittivity and Electrical Conductivity of Ce<Subscript>2</Subscript>S<Subscript>3</Subscript>

The rare-earth semiconductor β-Ce₂S₃ compound samples were synthesized and their dielectric permittivity and electrical conductivity were measured in the temperature range 90–400 K. The energy-band structure has been determined. It is shown that the long-known large electrical parameter spread of semiconductor compounds close in composition to Ce₂S₃ is explained by the structure of impurity donor levels formed by cerium atoms and ions with different ionization degrees.     

Electrical characterization of HfO2 films obtained by UV assisted injection MOCVD

HfO₂ films were deposited at low temperature (400 °C) by UV assisted injection metal-organic chemical vapor deposition (UVI-MOCVD). A three-step process was used for this study, consisting of (A) Pre-deposition anneal for nitridation; (B) Deposition step; (C) Post-deposition annealing in oxygen. Special attention was paid to the effect of UV exposure during these steps. Films were characterized by physical, optical and electrical techniques. Thickness was determined by different methods (X-ray Reflectrometry (XRR), spectroscopic ellipsometry and transmission electron microscopy) and a good agreement was found for all samples. The HfO₂ permittivity, equivalent oxide thickness (E_OT), flat-band voltage (V_fb) and total charge (Q_t) were extracted from the CV response at high frequency taking into account the HfO₂ and SiO₂ thicknesses obtained by XRR. The calculated permittivity values were in the range 7–13, i.e. lower than theoretical values for the monoclinic phase. Explanations are suggested in the context of the other characterizations. JE_eff characteristics were constructed taking into account the E_OT values (E_eff = V/E_OT). Effective breakdown fields range between 8.7 and 16.9 MV/cm. No dependence of E_eff with UV exposure was found.     

An Easy Rake Receiver for TD-SCDMA Smart Antenna

The application research on smart antenna (SA) for CDMA system is consists of two parts in general. One part is the research on mathematical methods in SA technology, i.e., how to isolate the array manifold parameter e^ja_m {e^{ja_m }} and then to get the weighting coefficient w _m for the SA receiving signal S(t)e^ja_m {S(t)e^{ja_m }} . The estimation of DOA is included in the above handling process. The other is how to extract the signal S(t)e^ja_m {S(t)e^{ja_m }} from SA receiver, which should be the focus of the research of SA application. Utilizing the feature of mid-amble sequence of TD-SCDMA system, a novel DOA estimation method based on mid-amble sequence (MDOA) is presented in this paper, by which the array manifold e^ja_m {e^{ja_m }} can be extracted directly from the SA receiving signal. Making use of the array manifold parameter e^ja_m {e^{ja_m }} obtained by the MDOA method, the simulation of downlink beam-forming shows that the proposed MDOA method is feasible. This method will result in the research of mathematical methods used in SA being completely abandoned. Also, a usable SA receiver for TD-SCDMA with beam-forming by amplitude weighting in base-band is given as well as the correctness proof by using spectrum searching method in base-band.     

Models for reliability and fail-safety evaluation of microcomputer control systems

This paper deals with the problem of developing models for microcomputer systems related to the safety of control (abbreviated as MSS). Two actual MSS systems are discussed, one (system S_mic¹ is non-redundant and the other (system S_mic²) is dual-redundant (1-out-of-2:F). A method to categorize system failure states as fail-safe or fail-dangerous is proposed and a general model for MSS is developed. Based on these results, analytical models for reliability and fail-safety evaluation of the above two actual systems, S_mic¹ and S_mic², are established.