基于人工神经网络方法的LSI工艺优化实践

本文基于人工神经网络的非线性和容错性，对集成电路生产工艺进行了分析和优化。主要内容有：１．使用人工神经网络方法建立模型，确定生产线上工艺参数和成品率之间的映射关系，构造以工艺参数为输入，成品率为输出的多维函数曲面；２．对该多维函数曲面进行搜索，找出成品率最高的最优点；３．以该最优点的工艺参数值为依据，确定工艺参数的规范值，对工艺参数提出优化建议，提高成品率。结论：神经网络提出的优化建议是合理的，并     

时间序列神经网络预测方法

本文从信息论的角度出发，讨论了利用神经网络理论构造时间序列预测模型的可能性和关键问题，并在此基础上提出３种时间序列神经网络预测方法，它们是：神经网络非线性时间序列模型，神经网络多维时间序列模型和神经网络组合预测模型。将上述模型应用于实例的结果表明，在非线性信息的处理能力和预测精度方面都有很大提高。进一步，对今后智能信息预测方法的发展方向进行了探讨，提出了智能信息预测系统的结构模型。     

局部散射源参数估计的非线性算子方法

本文利用单次快摄数据的两个非线性算子估计局部散射源中心波达方向和角扩展,与多维参数搜索和其他低复杂性的局部散射源参数估计算法不同,非线性算子方法给出了单次快摄的局部散射源参数估计的闭式解,无需进行多维搜索和子空间分解。     

一种基于高阶矩的神经网络参数估计器

本文提出了一种可控学习的两级多层神经网络模型,由此设计出一种基于高阶矩匹配的神经网络参数估计器;并对该神经网络模型的学习算法进行了研究,提出了一种自适应并行学习算法。仿真结果表明,这种利用神经网络进行模型参数估计的方法是可行的。     

时间序列神经网络预测方法

本文从信息论的角度出发,讨论了利用神经网络理论构造时间序列预测模型的可能性和关键问题,并在此基础上提出3种时间序列神经网络预测方法,它们是:神经网络非线性时间序列模型、神经网络多维时间序列模型和神经网络组合预测模型,将上述模型应用于实例的结果表明,在非线性信息的处理能力和预测精度方面都有很大提高。进一步,对今后智能信息预测方法的发展方向进行了探讨,提出了智能信息预测系统的结构模型。     

自适应谱估计的神经网络方法

提出了一种基于自适应ＡＲ过程参数求解而进行谱估计的神经网络方法，采用神经网络完成的ＡＲ参数估计回避了以往各种算法中的递推公式计算，实现了无乘法器的求解，该神经网络模型结构简单，便于硬件实现，计算机模拟结果表明，这种方法对平稳和非平稳随机过程进行的ＡＲ／ＭＥ谱分析结果令人满意。     

一种新的信号多参数估计方法

本文基于圆阵列及其输出时延提出了一种空间信号频率、方位角和仰角的联合参数估计新方法。此方法回顾了多维谱峰搜索和参数配比，克服了在短数据时离散傅立叶变换测频分辨率低和性能差的不足。计算机模拟证实了该方法的有效性。     

一种新的双基地MIMO雷达目标参数联合估计算法

摘 要：为了实现双基地多输入多输出（MIMO）雷达目标参数估计,本文建立了双基地MIMO雷达信号模型,提出了一种目标方向角和多普勒频率联合估计方法,该方法通过构造空时矩阵,只进行一次特征值分解,即可得到雷达多目标多参数的闭式解,避免了多维非线性的谱峰搜索,并可实现自动配对。仿真结果证明了所提算法的有效性。     

调幅信号的参数循环累量估计

本文利用循环累量讨论讽调幅混合相位信号的参数估计方法。首先给出了调幅信号的基于任意阶循环累量的参数方程。对于调幅混合相位ＡＲ，ＭＡ和ＡＲＭＡ模型，提出了相应的参数估计方法。模拟实验表明了本文方法的有效性。     

基于DDC模型的机载雷达多普勒参数估计研究

基于多普勒分布式杂波(DDC)模型,本文给出了机载雷达杂波多普勒参数估计的性能分析.并针对多普勒扩展较小的DDC模型,提出了一种基于非线性算子的多普勒参数快速估计算法,该算法通过采样数据直接给出多普勒参数的闭式解,能够显著地减少运算量并满足实际机载雷达的应用需求.     

基于自适应波束形成的高维数据挖掘算法

提出一种基于自适应波束形成的高维声传感器网络数据挖掘算法.进行多通道声传感器网络信号的高维信息数据采集和相空间重组,进行信号模型构建,对高维数据信息流进行子空间降维和自适应陷波器降噪滤波处理,采用自适应波束形成方法进行数据的谱峰聚焦和特征提取,实现数据准确挖掘.仿真结果表明,采用该算法进行数据挖掘的准确检测概率较高,抗干扰性能较好,波束旁瓣得到有效抑制.     

基于神经网络与最小二乘法在故障诊断中的应用研究

首先介绍BP神经网络与非线性最小二乘法在故障诊断中的应用,然后融合两者的技术特点,提出一种用于分析时序冗余信号,基于参数估计的故障诊断技术。随后,较详细地介绍了用于故障诊断神经网络样本的生成和网络的训练,通过以不依赖系统模型的神经网路的输出作为初始估计,利用依赖系统模型的最小二乘法完成对故障参数的最终估计。最后,通过在不同输入激励下的输出响应进行校验的方式,以仿真的方式验证了该故障诊断的应用具有可行性。     

一种基于神经网络的微波器件快速建模方法北大核心CSCD

随着微波器件结构复杂度的增长和产品性能要求的提高,微波器件建模不仅要能够描述其理想电磁特性,还要能快速准确反映多物理参数对器件性能的影响。虽然神经网络已经被引入到微波器件领域,但是将其应用于器件的多物理特性建模的研究还比较少。文章提出了一种基于人工神经网络的多物理参数建模方法来表示输入输出变量之间的非线性关系。提出了一种高效的神经网络多物理参数模型,并针对该模型引入了一种新的训练算法。所提出的模型可以快速准确地预测微波器件的多物理响应,如滤波器的S参数特性曲线、离子敏感场效应晶体管的输出特性曲线等。与有限元方法相比,此方法可以节省约98%的计算成本与99%的计算时间,为实现快速高效的微波器件行为级建模提供一种可行方法。     

Hopfield neural networks for state estimation: parameters, efficient implementation and results

State estimation processes measurements and other information to find the network state vector. In this paper, state estimation is considered as an optimization problem to be solved with a Hopfield neural network. Several activation models for this network are simulated and compared. A new method is proposed that calculates the integration step parameter for this network in an autonomous way, eliminating the need for determining it in a manual way for each particular problem. This algorithm has been successfully tested for a wide range of electrical nets. Neural and classic analytical methods are compared.     

基于一种杂交学习算法的自适应复信道均衡技术

本文提出了一种基于多层前馈神经网络杂交学习算法的自适应复信道均衡的新方法。该学习算法用来训练一个输入、输出、权值和激活函数均为复数的神经网络。神经网络的训练利用了监督和非监督相结合的杂交技术,而权值的调整是基于TLS（total least square)准则进行的。计算机仿真结果表明,无论是在线性还是在非线性信道中,所提出的方法都表现出了很好的性能,这为自适应复信道均衡提供了一种新方法。     

Wavelet Neural Network Approach for Testing of Switched-Current Circuits

Combining the time and frequency location and multiple-scale analysis of wavelet transform with the nonlinear mapping and generalizing of neural network, an efficient defect-oriented parametric test method using Wavelet Neural Network (WNN) for switched-current integrated circuits is proposed. Contraposing to the fully compatible digital CMOS technology and current scaling calculation of SI circuits, parameter cohort of switched current elements is used to compute the sensitivity and gain tolerance and is applied for selecting the test models. The selecting of the appropriate wavelet function based on particular switched current fault signal is discussed, and the number of network input and output nodes are determined by the circuit status and dimension of eigenvector which is the energy of wavelet decomposition coefficient. To simplify configuration of the neural network, the sampled data was preprocessed by wavelet transform. Illustrative examples show that the proposed wavelet neural network method for testing of switched current circuits is effective.     

Neural Network Based Adaptive Tracking of Nonlinear Multi-Agent System

In this paper, the problems of robust consensus tracking control for the second-order multi-agent system with uncertain model parameters and nonlinear disturbances are considered. An adaptive control strategy is proposed to smooth the agent’s trajectory, and the neural network is constructed to estimate the system’s unknown components. The consensus conditions are demonstrated for tracking a leader with nonlinear dynamics under an adaptive control algorithm in the absence of model uncertainties. Then, the results are extended to the system with unknown time-varying disturbances by applying the neural network estimation to compensating for the uncertain parts of the agents’ models. Update laws are designed based on the Lyapunov function terms to ensure the effectiveness of robust control. Finally, the theoretical results are verified by numerical simulations, and a comparative experiment is conducted, showing that the trajectories generated by the proposed method exhibit less oscillation and converge faster.     

Nonlinear Compensation of High Power Amplifier Distortion for Communication Using a Histogram-Based Method

A probabilistic approach to compensate the nonlinear distortion caused by a high power amplifier (HPA) in a communication system is proposed here. This is a nonparametric method that involves estimating two probabilistic cumulative distribution functions without any explicit parameter estimation as in the conventional compensation techniques. It is shown analytically that the maximum compensation error of the proposed method is bounded and small. An adaptive implementation is developed. Experiments are setup to collect the HPA data. Real data analysis shows that the proposed predistorter has an improved performance compared to the conventional polynomial and neural network predistortion techniques.     

Generating optimal adaptive fuzzy-neural models of dynamicalsystems with applications to control

The paper describes an approach to generating optimal adaptive fuzzy neural models from I/O data. This approach combines structure and parameter identification of Takagi-Sugeno-Kang (TSK) fuzzy models. We propose to achieve structure determination via a combination of modified mountain clustering (MMC) algorithm, recursive least squares estimation (RLSE), and group method of data handling (GMDH). Parameter adjustment is achieved by training the initial TSK model using the algorithm of an adaptive network based fuzzy inference system (ANFIS), which employs backpropagation (BP) and RLSE. Further, a procedure for generating locally optimal model structures is suggested. The structure optimization procedure is composed of two phases: 1) locally optimal rule premise variables subsets (LOPVS) are identified using MMC, GMDH, and a search tree (ST); and 2) locally optimal numbers of model rules (LONOR) are determined using MMC/RLSE along with parallel simulation mean square error (PSMSE) as a performance index. The effectiveness of the proposed approach is verified by a variety of simulation examples. The examples include modeling of a nonlinear dynamical process from I/O data and modeling nonlinear components of dynamical plants, followed by tracking control based on a model reference adaptive scheme (MRAC). Simulation results show that this approach is fast and accurate and leads to several optimal models     

Robust H/sub /spl infin// controller design with recurrent neural network for linear synchronous motor drive

In this paper, a robust controller design with H/sub /spl infin// performance using a recurrent neural network (RNN) is proposed for the position tracking control of a permanent-magnet linear synchronous motor. The proposed robust H/sub /spl infin// controller, which comprises a RNN and a compensating control, is developed to reduce the influence of parameter variations and external disturbance on system performance. The RNN is adopted to estimate the dynamics of the lumped plant uncertainty, and the compensating controller is used to eliminate the effect of the higher order terms in Taylor series expansion of the minimum approximation error. The tracking performance is ensured in face of parameter variations, external disturbance and RNN estimation error once a prespecified H/sub /spl infin// performance requirement is achieved. The synthesis of the RNN training rules and compensating control are based on the solution of a nonlinear H/sub /spl infin// control problem corresponding to the desired H/sub /spl infin// performance requirement, which is solved via a choice of quadratic storage function. The proposed control method is able to track both the periodic step and sinusoidal commands with improved performance in face of large parameter perturbations and external disturbance.