高速列车转向架故障信号的聚合经验模态分解和 模糊熵特征分析

为了对高速列车转向架关键部件进行状态监测,利用转向架故障振动信号的特点,提出了一种结合聚合经验模态分解和模糊熵的特征提取方法.对故障信号进行聚合经验模态分解,得到一系列具有不同物理意义的简单成分信号,采用相关分析法选取最能反映原信号特征的本征模态函数.对这些本征模态函数和原信号分别计算模糊熵值构成多尺度复杂性度量的特征向量,输入最小二乘支持向量机中进行分类识别,与模糊熵特征相比得到了更好的识别效果,证明了算法的有效性.     

The use of clipped input information in multidimensional cross-correlation for estimating Wiener-like kernels of non-linear systems

While the measurement of Wiener-like kernels by multidimensional input-output cross-correlation is a well-known non-parametric approach to non-linear system identification, we propose here a simplifying kernel estimation scheme; rather than using the white-noise signal that is actually applied to stimulate the system, the present method uses a clipped information (that is, two- or three-level quantization) of the continuous-level test input for computing the cross-correlation. This greatly reduces the computational requirement without disturbing the generality of actual test input which may be gaussian. The statistical variance of the kernel estimation is discussed in comparison with other algorithms. Certain non-statistical errors may be incurred using this approach, but are thought to be minor in most applications. A special emphasis is given to the problem of choosing optimal procedural parameters for ternary quantization in the case of white gaussian input.     

The monotonic increasing relationship between average powers of CMOS VLSI circuits with and without delay and its applications

The authors theoretically describe the monotonic increasing relationship between average powers of a CMOS VLSI circuit with and without delay. The power of an ideal circuit without delay, which can be fast computed, has been used as the evaluation criterion for the power of a practical circuit with delay, which needs more computing time, in such fields as fast estimation for the average power and the maximum power, and fast optimization for the low test power. The authors propose a novel simulation approach that uses delay-free power to compact a long input vector pair sequence into a short sequence and then, uses the compacted one to fast simulate the average (or maximum) power for a CMOS circuit. In comparison with the traditional simulation approach that uses an un-compacted input sequence to simulate the average (or maximum) power, experiment results demonstrate that in the field of fast estimation for the average power, the present approach can be 6-10 times faster without significant loss in accur     

基于邻域相关性多阈值新函数寻优法的小波降噪分析

为了更好地获取噪声影响下的原有信号,在邻域小波系数收缩的NeighCoeff方法基础之上,提出了一种邻域相关性多阈值新函数的小波降噪方法.该方法根据小波系数之间的相关性,将邻域窗口内所有小波系数的平方和的大小划分为邻域硬阈值、邻域窗口阈值和邻域扩张阈值.将这些邻域阈值与修正的通用阈值相比较,来实现窗口尺寸的自适应调节和小波系数的保留或收缩,以此达到消噪的目的.此外新函数的收缩因子能够较好体现与被滤波噪声的相互关系,可以进一步提高消噪的精度.然后将多阈值函数与修正的全局阈值相结合,利用混沌粒子群对邻域扩张阈值参数γ和修正的全局阈值参数α进行寻优,以获取最优小波系数的重构信号.所提方法与其它阈值函数去噪方法相比,其仿真结果表明在信号信噪比、降低有用信号失真和抑制噪声等方面都有一定的提高.     

Non-parametric linear time-invariant system identification by discrete wavelet transforms

We describe the use of the discrete wavelet transform (DWT) for non-parametric linear time-invariant system identification. Identification is achieved by using a test excitation to the system under test (SUT) that also acts as the analyzing function for the DWT of the SUT's output, so as to recover the impulse response. The method uses as excitation any signal that gives an orthogonal inner product in the DWT at some step size (that cannot be 1). We favor wavelet scaling coefficients as excitations, with a step size of 2. However, the system impulse or frequency response can then only be estimated at half the available number of points of the sampled output sequence, introducing a multirate problem that means we have to ‘oversample’ the SUT output. The method has several advantages over existing techniques, e.g., it uses a simple, easy to generate excitation, and avoids the singularity problems and the (unbounded) accumulation of round-off errors that can occur with standard techniques. In extensive simulations, identification of a variety of finite and infinite impulse response systems is shown to be considerably better than with conventional system identification methods.     

High resolution time delay estimation using sliding discrete Fourier transform

In this paper, we propose a novel time delay estimation approach based on sliding the discrete Fourier transform (DFT) analysis window, sample by sample, over the received short continuous wave (CW) pulse signal with the DFT evaluated successively. This approach uses the maximum magnitude of the spectrum and its corresponding phase offset to estimate the time delay (pulse echo mode) of the signal. We use the corresponding time as the first estimate, which is improved on the basis of the related phase. Examples are given of synthetic signals and simulated delays scenario, with and without added white noise. An underwater application, based on distance and speed of sound measurements using this approach in a water tank is demonstrated. The proposed method is shown to significantly outperform standard correlator-based approaches. Furthermore, the algorithm is simple to use and can be easily implemented, being based on phase detection using the sliding DFT.     

一种应用于遥操作系统的机械振动辨识方法

当考虑遥操作系统中轻型机器人的结构振动辨识问题时,由于通讯环节的带宽窄和通讯频率低,传统的振动辨识方法将遭遇瓶颈．因此,研究如何减少振动辨识过程中需要传输的数据量显得尤为重要的．为此,本文借鉴ESPRIT（estimation of signal parameters via rotational invariance techniques）算法,研究了机械结构振动的欠采样辨识问题．提出了ASP（adding sampling points）方法．在ASP中,用平均值矩阵取代了传统ESPRIT算法中的相关矩阵,在保证充分利用采样数据的同时,克服了相关矩阵易产生病态的问题,并减小了计算量;对振动频率和衰减系数分别进行辨识,然后再将它们匹配成对,解决了频率解混叠和衰减系数准确辨识之间的矛盾;利用差异度法匹配振动频率和衰减系数,在匹配过程中可以对辨识出的振动频率进行修正．仿真和实验都证明了该方法不仅可行,而且具有很好的效果．     

2D-discontinuity detection from scattered data

We describe a numerical approach for the detection of discontinuities of a two dimensional function distorted by noise. This problem arises in many applications as computer vision, geology, signal processing. The method we propose is based on the two-dimensional continuous wavelet transform and follows partially the ideas developed in [2], [6] and [8]. It is well-known that the wavelet transform modulus maxima locate the discontinuity points and the sharp variation points as well. Here we propose a statistical test which, for a suitable scale value, allows us to decide if a wavelet transform modulus maximum corresponds to a function value discontinuity. Then we provide an algorithm to detect the discontinuity curves fromscattered and noisy data.     

Wavelet denoising via sparse representation

Wavelet threshold denoising is a powerful method for suppressing noise in signals and images. However, this method often uses a coordinate-wise processing scheme, which ignores the structural properties in the wavelet coefficients. We propose a new wavelet denoising method using sparse representation which is a powerful mathematical tool recently developed. Instead of thresholding wavelet coefficients individually, we minimize the number of non-zero coefficients under certain conditions. The denoised signal is reconstructed by solving an optimization problem. It is shown that the solution to the optimization problem can be obtained uniquely and the estimates of the denoised wavelet coefficients are unbiased, i.e., the statistical means of the estimates are equal to the noise-free wavelet coefficients. It is also shown that at least a local optimal solution to the denoising problem can be found. Our experiments on test data indicate that this new denoising method is effective and efficient for a wide variety of signals including those with low signal-to-noise ratios. Supported by the U.S. National Institutes of Health (Grant No. U01 HL91736), and the National High-Tech Research & Development Program of China (Grant No. 2007AA01Z175)     

基于稀疏表征的话者识别

近年来,随着信号的稀疏性理论越来越受到人们的关注,稀疏表征分类器也作为一种新型的分类算法被应用到话者识别系统中。该模型的基本思想是:只要超完备字典足够大,任意待测样本都能够用超完备字典进行线性表示。基于信号的稀疏性理论,未知话者的向量系数,即稀疏解可以通过L1范数最小化获取。超完备字典则可视为语音特征向量在高斯混合模型-通用背景模型(GMM-UBM)上进行MAP自适应而得到的大型数据库。采用稀疏表征模型作为话者辨认的分类方法,基于TIMIT语料库的实验结果表明,所采用的话者辨认方法,能够大大提高说话人识别系统的性能。     

Soft-competitive learning of sparse codes and its application to image reconstruction

We propose a new algorithm for the design of overcomplete dictionaries for sparse coding, neural gas for dictionary learning (NGDL), which uses a set of solutions for the sparse coefficients in each update step of the dictionary. In order to obtain such a set of solutions, we additionally propose the bag of pursuits (BOP) method for sparse approximation. Using BOP in order to determine the coefficients of the dictionary, we show in an image encoding experiment that in case of limited training data and limited computation time the NGDL update of the dictionary performs better than the standard gradient approach that is used for instance in the Sparsenet algorithm, or other state-of-the-art methods for dictionary learning such as the method of optimal directions (MOD) or the widely used K-SVD algorithm. In an application to image reconstruction, dictionaries trained with this algorithm outperform not only overcomplete Haar-wavelets and overcomplete discrete cosine transformations, but also dictionaries obtained with widely used algorithms like K-SVD.     

QoS for real-time reliable multicasting in wireless multi-hop networks using a Generation-Based Network Coding

Most of the works on Generation-Based Network Coding (GBNC) consider a fixed generation size. A large generation size maximizes the Network Coding benefits but leads to a long delay while a small generation size reduces the delay but decreases the throughput. This paper presents the DYnamic GEneration Size (DYGES) approach. Our network-aware method adjusts the generation size according to the network variations (network size, congestion, losses) for multicast flows to keep the delay steady. Since Network Coding and redundancy cope with data packet loss, we propose an enhancement of DYGES with ACK recovery. This method, named RDYGES, uses the opportunistic listening feature of nodes to recover the lost ACK. Our goal is to guarantee a Quality of Service (QoS) in terms of delay. The simulation results show the accuracy of our approach.     

Semi-parametric classification of noisy curves

We propose a novel semi-parametric modeling strategy for classifying noisy curves. This strategy uses a family of non-linear parametric models to describe known aspects of the signal and its propagation, with a non-parametric component incorporating unmodeled characteristics. We propose a novel multi-record model building strategy and assess its scope in classifying acoustic and radar signals. Our experiments suggest that the semi-parametric approach generally out performs the parametric approach, and in certain circumstance gives better performance than the non-parametric approach. In all cases, it is close to the best approach considered, with the added advantage of interpretable coefficients in the parametric component.     

中医脉诊信号的无监督聚类分析研究

随着中医客观化工作的推进,脉诊技术也越来越走向客观化和仪器化。然而,如何对仪器所检测和收集到的信息进行解读,却还是回到了原来脉诊诊断主观化的问题上。因为传统的机器学习方法,依赖于对大量的脉诊数据进行标注。但是在临床诊断和教学中,医生与医生之间对于脉象的体会不同,会导致他们对病人脉象的区分标注不同。在对比了多种特征提取方法和聚类方案之后,提出了一个较好的无监督脉诊客观化方法,在双树复小波变换(DTCWT)对数据进行预处理的基础上,以梅尔倒谱系数(MFCC)进行特征提取,在中医专家对数据进行标注之前,先根据信号的特征,使用Fuzzy c-means (FCM)聚类算法进行粗线条的分类,使得在此基础之上,可以开展进一步的细化分类研究。实验结果表明:该方法可取得较好的分类效果,为中医脉诊提供了进一步客观化的依据。     

等间隔分数时延滤波器及其线性插值设计

针对宽带数字信号精确可变时延需求,研究等间隔分数时延滤波器及其线性插值方法。现有分数时延滤波器设计方法不能应用于可变时延,基于多采样率信号处理理论,提出等间隔分数时延滤波器设计方法。将所需时延近似为等间隔分数时延,选取对应时延滤波器组对信号进行时延,满足工程中变分数时延滤波需求。为提高时延滤波精度,提出对相邻滤波器系数线性插值的分数时延滤波器设计方法。对线性调频信号仿真结果表明,所提方法与常规分数时延滤波方法运算量相当,且适用于宽带信号的精确可变时延应用。     

Online Algebraic Identification Approach of Continuous Linear Time Delay Systems

In this paper, we propose an online algebraic method to identify linear continuous time delay process from step response, in presence of unknown initial state and constant load disturbance. The identification mechanism is split into two sequential steps. The time delay and dynamic of the plant is estimated firstly through a spectral formulation. In the next stage, the transient regime is deduced from a linear regression. To improve the robustness of the proposed method, a local convolution by sigmoid function is proposed. Simulation results are provided, at last, to show the effectiveness of our conceived approach.     

A frequency adaptive control for multidimensional systems

An adaptive control is designed for a multidimensional system with unknown constant coefficients under bounded polyharmonic disturbances containing an infinite number of harmonics of unknown amplitudes and frequencies. It uses a very small test signal. The control aim is to ensure given bounds for the forced oscillations in the output of the system and controller. Adaptation is based on finite-frequency identification of the system and a closed-loop system. By way of example, an adaptive control of a real physical system is given.     

Regularized variational Bayesian learning of echo state networks with delay&sum readout

In this work, a variational Bayesian framework for efficient training of echo state networks (ESNs) with automatic regularization and delay&sum (D&S) readout adaptation is proposed. The algorithm uses a classical batch learning of ESNs. By treating the network echo states as fixed basis functions parameterized with delay parameters, we propose a variational Bayesian ESN training scheme. The variational approach allows for a seamless combination of sparse Bayesian learning ideas and a variational Bayesian space-alternating generalized expectation-maximization (VB-SAGE) algorithm for estimating parameters of superimposed signals. While the former method realizes automatic regularization of ESNs, which also determines which echo states and input signals are relevant for "explaining" the desired signal, the latter method provides a basis for joint estimation of D&S readout parameters. The proposed training algorithm can naturally be extended to ESNs with fixed filter neurons. It also generalizes the recently proposed expectation-maximization-based D&S readout adaptation method. The proposed algorithm was tested on synthetic data prediction tasks as well as on dynamic handwritten character recognition.     

Spoofing Signal Generation Based on Manipulation of Code Delay and Doppler Frequency of Authentic GPS Signal

We propose an efficient spoofing signal generation method that uses the processing results of a global positioning system (GPS) receiver for authentic GPS signals. Conventional methods of generating spoofing signals use expensive GPS simulators because the structures of the spoofing signals must be almost identical to those of the GPS signals. Simulators require GPS ephemeris at a specific time and target position. Subsequently, a complicated process is used to generate navigation data using the ephemeris and model error sources such as the satellite clock bias and ionospheric delay. In contrast, the proposed method can generate spoofing signals for the desired target position without requiring GPS simulators; it does so by adjusting the signal processing results of the receiver. Using the navigation results of the receiver, such as position and velocity, the pseudorange delay and spoofing Doppler frequency between the estimated position of the receiver and the target spoofing position are obtained; these are then applied to shift the signal-tracking results of the receiver to create a new signal for the target spoofing position. Our experimental results indicate that the proposed algorithm can effectively generate spoofing signals with characteristics highly similar to those of authentic GPS signals. In addition, we confirmed that the spoofing signals generated by the proposed method are difficult to be detected using conventional spoofing detection techniques.

     

Self-Tuning of a Neural Network Controller with an Integral Estimate of Contradictions between the Commands of the Learning Algorithm and Memory

We propose an approach to organizing self-tuning for a controller based on an artificial neural network that uses information on the contradictions arising in the creation of the value for the control signal between accumulated memory of the neural network and the learning algorithm based on backpropagation. The activity of neural network memory is estimated as its reaction to changing the state of the control system. Self-tuning is done by controlling the learning rate coefficient with an integral controller in order to stabilize the integral criterion for estimating the contradictions. Based on this modeling, we show a conceptual possibility for the operation of the self-tuning system with constant tuning parameters in a wide range of changes of the control object’s dynamical properties.