蒙特卡罗-牛顿法进行电路产品合格率估计和最优中心设计

本文利用蒙特卡罗法估计电路产品合格率对元件参数中心值的一、二阶偏导数,再与牛顿最优化方法结合得到了一个电路中心设计的算法,为了提高偏导数估计值的精度,文中又给出了利用失败样本点的估计式。为了减少电路分析次数,作者提出了适用于蒙特卡罗分析的采样频率按权排序法,理论分析和算例表明:本文算法是成功的,适用于非凸、非单连通可行域,可进行中等规模电路的产品合格率估计和中心设计。     

光OFDM系统中几种同步算法的对比分析

对光OFDM系统中的三种同步算法进行了理论分析,基于仿真实验,针对定时度量、定时偏差、丢帧率三个性能参数进行了对比分析。仿真分析表明T.Schmidl＆D.Cox算法由于循环前缀（Cyclic Prefix,CP）缘故使得其定位精确性欠佳,偏差较大,无法进行精确定位,但算法稳定性及丢帧率方面具有很好的性能。而H.Minn和Li-xun Huang＆Li-jun Sun两种同步算法定位精确性相对较好,偏差较小,但在稳定性及丢帧率方面,后者优于前者。因而在这三种同步算法中,Li-xun Huang＆Li-jun Sun算法性能较好,适应性较强,更具实际应用价值。     

A quasi-three-dimensional large-signal circuit model for lateral transient analysis of MOS device

A new quasi-three-dimensional large-signal circuit model of the metal-oxide-semiconductor capacitance (MOSC) under non-equilibrium depletion or inversion bias is presented. This model is based on the threedimensional version of Sah's distributed equivalent circuit model. A simple technique of scaling the circuit elements to account for circular symmetry is shown. The circuit elements perpendicular to the MOS interface are integrated to yield a quasi-three-dimensional circuit model. The numerical analysis technique of Green and Schewchun is then used to obtain photocurrent transient responses under arbitrary signal level. Theoretical photocurrent transient calculations are in excellent agreement with experimental curves, both for the fast initial transient as well as the slow long time decay.     

IC优化设计及其成品率预测

IC成品率是与电路性能和制造成本及制造效益紧密相关的一个重要因素,在进行IC优化设计时,可将成品率与制造效益作为协调各性能指标的优化目标.针对这一问题,本文完善了IC成品率效益协调优化设计模型,提出了一种实现该模型的IC优化设计和成品率预测方法,并利用OrCAD/PSpice中的统计分析和电路性能分析的功能和特点,建立了相应的算法.实例表明,该方法及其实现算法是有效的.     

A new channel estimator for fast start-up equalization

A novel noniterative algorithm is proposed for estimating the sampled impulse response of unknown channels. With the proper choice of the training sequence, implementation of the proposed channel estimator requires only additions and subtractions, i.e. no multiplications of the estimator is so simple that it can be easily implemented using an ordinary microprocessor with minimal storage. The channel estimate can serve to reduce the effects of intersymbol interference either by the maximum likelihood sequence estimator using the Viterbi algorithm or by channel equalization with a direct solution to obtain the optimum equalizer taps. A new procedure for the latter case is proposed using the Levinson-Trench algorithm for fast start-up of adaptive equalizers in noisy environments. The performance of the proposed algorithm is evaluated through simulation, and it is compared to some previous techniques. Computational efficiency is also taken into account. The results of the simulation show the superiority of the proposed scheme     

Analytic threshold calculation of frequency estimation for OFDM communication

In this paper, we introduce an improved expression for the normalized variance of a frequency offset estimator, which successfully estimates the actual behavior of the normalized variance of the frequency offset estimator below and above the threshold. Therefore, an orthogonal frequency division multiplexing designer does not need any more to simulate the whole system in order to obtain the actual, normalized variance of the frequency offset estimator for the lower range of signal to noise ratio. We also present the source for the threshold phenomenon and the “threshold region”.     

Ambiguity resolution for permanent scatterer interferometry

In the permanent scatterer technique of synthetic aperture radar interferometry, there is a need for an efficient and reliable nonlinear parameter inversion algorithm that includes estimation of the phase cycle ambiguities. Present techniques make use of a direct search of the solution space, treating the observations as deterministic and equally weighted, and which do not yield an exact solution. Moreover, they do not describe the quality of the estimated parameters. Here, we use the integer least squares estimator, which has the highest probability of correct integer estimation for problems with a multivariate normal distribution. With this estimator, the propagated variance-covariance matrix of the estimated parameters can be obtained. We have adapted the LAMBDA method, part of an integer least squares estimator developed for the ambiguity resolution of carrier phase observations in global positioning systems, to the problem of permanent scatterers. Key elements of the proposed method are the introduction of pseudo-observations to regularize the system of equations, decorrelation of the ambiguities for an efficient estimation, and the combination of a bootstrap estimator with an integer least squares search to obtain the final integer estimates. The performance of the proposed algorithm is demonstrated using simulated and real data.     

一种有效的电路容错结构及其成品率分析

本文提出了一种有效的电路容错结构,给出了一种重组算法.按照这种重组算法,对该结构的成品率进行了分析,并给出了其成品率的表达式.最后,通过电路设计和分析说明了该容错结构的有效性.     

一种基于非参数统计的微波集成电路成品率优化算法

给出了一种新的微波集成电路成品率优化算法——基于非参数统计的成品率快速优化算法。利用该方法，仅需要较少的仿真次数或直接利用非参数统计成品率分析算法的结果，便可直接得出一组或几组参数的成品率优化值，有效地缩短了优化时间。算例表明，该方法对微波集成电路进行快速成品率优化设计及提高电路设计的稳定性具有较好的应用价值。     

Performance evaluation of a new BPSK-OFDM timing estimation algorithm

Previously, the timing estimation in OFDM system employs cyclic prefix has been presented with the assumption of independent identified distributed OFDM symbol. The information used to find the symbol synchronization depends on the length of cyclic prefix. Actually, the data in the BPSK-OFDM modulated symbol, transferred by inverse fast Fourier transform, is with a complex symmetry character. With this character, more information from the whole OFDM symbol could be provided for the symbol timing. In this paper, the proposed algorithm uses the autocorrelation of the received signal to obtain the symbol timing. The simulation of Beek’s algorithm and the proposed scheme are presented. The performances of these algorithms are evaluated based on the missing symbol probability and the estimator mean square error. The simulation results show that the proposed algorithm is suitable to achieve a better symbol synchronization.     

A Fast Implementation of a Minimum Variance Estimator for Computerized Tomography Image Reconstruction

Most present day computerized tomography (CT) systems are based on reconstruction algorithms that produce only approximate deterministic solutions of the image reconstruction problem. These algorithms yield reasonable results in cases of low measurement noise and regular measurement geometry, and are considered acceptable because they require far less computation and storage than more powerful algorithms that can yield near optimal results. However, the special geometry of the CT image reconstruction problem can be used to reduce by orders of magnitude the computation required for optimal reconstruction methods, such as the minimum variance estimator. These simplifications can make the minimum variance technique very competitive with well-known approximate techniques such as the algebraic reconstruction technique (ART) and convolution-back projection. The general minimum variance estimator for CT is first presented, and then a fast algorithm is described that uses Fourier transform techniques to implement the estimator for either fan beam or parallel beam geometries. The computational requirements of these estimators are examined and compared to other techniques. To allow further comparison with the commonly used convolution-back projection method, a representation of the fast algorithm is derived which allows its equivalent convolving function to be examined. Several examples are presented.     

Stochastic Gradient Algorithm Under (<Emphasis Type="Italic">h</Emphasis>,<Emphasis Type="Italic">φ</Emphasis>)-Entropy Criterion

Motivated by the work of Erdogmus and Principe, we use the error (h,φ)-entropy as the supervised adaptation criterion. Several properties of the (h,φ)-entropy criterion and the connections with traditional error criteria are investigated. By a kernel estimate approach, we obtain the nonparametric estimator of the instantaneous (h,φ)-entropy. Then, we develop the general stochastic information gradient algorithm, and derive the approximate upper bound for the step size in the adaptive linear neuron training. Moreover, the (h,φ) pair are optimized to improve the performance of the proposed algorithm. For the finite impulse response identification with white Gaussian input and noise, the exact optimum φ function is derived. Finally, simulation experiments verify the results and demonstrate the noticeable performance improvement that may be achieved by the optimum (h,φ)-entropy criterion.     

逆向统计模拟及在VLSI统计分析中的应用

本文在理论上提出了逆向统计模拟思想,并用以开发了NMOS数字集成电路统计模拟通用软件——STANMOS。应用该软件可定量得到工艺涨落及工艺干扰对电路性能的影响,分析电路性能的工艺灵敏度及成品率,确定主要影响电路性能一致性的工艺步骤。     

Position control of a sensorless synchronous reluctance motor

This paper presents a novel position control for a sensorless synchronous reluctance drive system. By measuring the three-phase currents of the motor, a rotor position estimator is achieved. Then, a velocity estimator is derived from the estimated rotor position by using a state estimating technique. The estimated velocity tracks the real velocity well. Next, a robust position controller is designed to improve the transient and load disturbance responses. By using the proposed estimating techniques and control algorithm, a high-performance sensorless synchronous reluctance drive is obtained. A digital signal processor, TMS-320-C30, is used to execute the estimating and control algorithms. No hardware circuit or external signal is added as compared with the traditional drive system with an encoder or resolver. To evaluate the performance of the position control system, a moving table is connected with the drive system. The drive system can precisely control the moving table. Experimental results show that the proposed system has good performance. Several experimental results validate the theoretical analysis.     

An area-efficient maximum-ratio-combining diversity receiver in 90 nm CMOS for free-space optical links

This paper presents a low-power imaging diversity front-end receiver employing the maximum-ratio-combining algorithm for free-space optical communication. It consists of seven signal channels and an output stage, each channel has a front-end transimpedance amplifier, a signal-to-noise ratio (SNR) estimator and a variable gain amplifier (VGA). The imaging receiver circuit was implemented in a 90 nm CMOS process. The maximum-ratio weighting is achieved with the SNR estimator and variable gain amplifier (VGA), which provides the signal with a gain proportional to the signal amplitude. The maximum ratio combining feature was demonstrated with two channels driven by photodiode emulation circuits for electrical characterization. The power dissipation for the whole chip is 43 mW from a single 1.2 V supply.     

基于OMP的SRAM成品率分析的分组建模方法

静态随机存取存储器(SRAM)电路的失效是极小概率事件,并且不同电路条件下的失效区域边界可能相距很远。因此,为了更高效地获得更精准的SRAM成品率预测结果,提出一种基于正交匹配追踪(OMP)算法的SRAM性能分组建模方法,并应用于典型SRAM电路成品率的预测。此方法主要根据不同SRAM电路条件下失效区域边界距离的差异将仿真数据划分为多组,之后利用OMP算法对不同组的数据分别建立多项式模型,该模型可用于对SRAM电路的成品率进行快速分析预测。与标准蒙特卡洛统计算法及基于OMP的单一建模方法相比,基于OMP的分组建模方法不仅可以缩短建模时间,提高建模准确度,还能够获得更加精确的SRAM成品率预测结果。     

A fast, high-quality inverse halftoning algorithm for errordiffused halftones

Halftones and other binary images are difficult to process with causing several degradation. Degradation is greatly reduced if the halftone is inverse halftoned (converted to grayscale) before scaling, sharpening, rotating, or other processing. For error diffused halftones, we present (1) a fast inverse halftoning algorithm and (2) a new multiscale gradient estimator. The inverse halftoning algorithm is based on anisotropic diffusion. It uses the new multiscale gradient estimator to vary the tradeoff between spatial resolution and grayscale resolution at each pixel to obtain a sharp image with a low perceived noise level. Because the algorithm requires fewer than 300 arithmetic operations per pixel and processes 7x7 neighborhoods of halftone pixels, it is well suited for implementation in VLSI and embedded software. We compare the implementation cost, peak signal to noise ratio, and visual quality with other inverse halftoning algorithms.     

REDUCED COMPLEXITY FREQUENCY ESTIMATOR FOR BURST TRANSMISSION

In burst transmission, carrier recovery is a critical point for synchronization systems. With a feedforward carrier phase recovery algorithm, a small frequency offset can significantly increase the cycle slip rate and then the phase error variance. Therefore, in order to obtain an accurate carrier phase estimation, a precise frequency correction is required. For M-states phase shift keying (M-PSK) modulated signals an unbiased feedforward reduced complexity frequency estimator (RCFE), operating in the non-data aided mode (NDA) is derived from the maximum likelihood (ML) principle. A compromise is realized between noise filtering and estimation slip probability by minimizing the estimator variance. It is optimized to operate at a low signal-to-noise ratio and short bursts. Its performance is compared to that of the ML estimator. The estimator is applied to an all-feedforward synchronization structure with QPSK modulated signals. Global performance of the modem synchronization structure is supplied. © 1996 by John Wiley & Sons, Ltd.     

A robust channel estimator for DS-CDMA systems under multipath fading channels

This paper addresses the problem of channel estimation for direct-sequence code-division multiple-access (DS-CDMA) systems with time-varying multipath fading channels. The multipath fading channels are modeled as autoregressive (AR) models. A method is first proposed to convert the time-varying regression model due to the time-varying nature of users' information symbols into a time-invariant one. Then, a polynomial approach is proposed to obtain the minimum mean square error (MMSE) estimator. The uncertainty of the channel model and decision errors of the DS-CDMA detector are taken into consideration in the design of the MMSE estimator. Compared with the Kalman estimator, the computational complexity of the proposed algorithm is much lower. The simulation results show that the proposed estimator provides a comparable estimation performance with the Kalman estimator and is robust for fast-fading channels.     

Yield optimization using a GaAs process simulator coupled to aphysical device model

A physics-based large-signal GaAs MESFET model and circuit simulator has been developed to predict and optimize the yield of GaAs MESFET designs before fabrication. Device acceptance criteria include both small- and large-signal RF operating characteristics such as small-signal gain, maximum power added efficiency, and output power at 1-dB gain compression. Channel doping details are described on the basis of processing specifications for parameters such as material deposition, ion implantation, and implant annealing. Monte Carlo techniques are used to estimate yield when disturbances in the physical parameters are modeled as multivariate Gaussian distributions. The yield estimator is integrated with an optimizer so that a design can be centered for maximum yield in the presence of process disturbances