A computational strategy for multiscale systems with applications to Lorenz 96 model

Numerical schemes for systems with multiple spatio-temporal scales are investigated. The multiscale schemes use asymptotic results for this type of systems which guarantee the existence of an effective dynamics for some suitably defined modes varying slowly on the largest scales. The multiscale schemes are analyzed in general, then illustrated on a specific example of a moderately large deterministic system displaying chaotic behavior due to Lorenz. Issues like consistency, accuracy, and efficiency are discussed in detail. The role of possible hidden slow variables as well as additional effects arising on the diffusive time-scale are also investigated. As a byproduct we obtain a rather complete characterization of the effective dynamics in Lorenz model.     

基于HMM的机载设备状态健康评估方法研究

针对机载设备的状态健康评估问题,采用隐马尔科夫模型(HMM)对其进行性能退化程度的评估。首先引入状态条件概率矢量对HMM进行不确定性改进,并推导了其表达式;其次以状态条件概率比值为基础,给出了机载设备状态等级量化分值的计算方法,并据此设计了机载设备状态健康评估流程。最后以飞机发动机温控放大器为例进行仿真验证,结果表明上述方法能够给出直观、准确的状态评估结果。     

基于状态驻留时间下由转移弧产生输出的HMM模型中的参数估计

本讨论了用状态驻留时间来模型化传统的HMM模型。HMM的一个基本假设是它认为语音信号是准平稳的。然而由状态输出yt的HMM模型，并不能很好地表征语音信号中平稳段或平稳段之间的具体特征；由转移弧产生输出的自左向右HMM系统,则对语音特征作更为细致的描述。本主要讨论在[2]的基础上,对新建模型进行参数估计。     

基于系统调用的实时IDS方案

给出一种基于进程系统调用的实时IDS方案.该方案选择网络服务进程(特权进程)作为系统的代表进行监测,使用隐马尔可夫模型(HMM)对系统行为进行模拟和测试,使用实时P(O|λ)值和近期对“某些重要文件”的读、写频率这两个条件作为判断系统是否遭受入侵的标准.实验和分析表明,该方案有较高的预报准确率和较小的时间开销.     

Characterization of fish schooling behavior with different numbers of Medaka (Oryzias latipes) and goldfish (Carassius auratus) using a Hidden Markov Model

Fish that swim in schools benefit from increased vigilance, and improved predator recognition and assessment. Fish school size varies according to species and environmental conditions. In this study, we present a Hidden Markov Model (HMM) that we use to characterize fish schooling behavior in different sized schools, and explore how school size affects schooling behavior. We recorded the schooling behavior of Medaka (Oryzias latipes) and goldfish (Carassius auratus  ) using different numbers of individual fish (10–40), in a circular aquarium. Eight to ten 3 s video clips were extracted from the recordings for each group size. Schooling behavior was characterized by three variables: linear speed, angular speed, and Pearson coefficient. The values of the variables were categorized into two events each for linear and angular speed (high and low), and three events for the Pearson coefficient (high, medium, and low). Schooling behavior was then described as a sequence of 12 events (2×2×3$2\times 2\times 3$), which was input to an HMM as data for training the model. Comparisons of model output with observations of actual schooling behavior demonstrated that the HMM was successful in characterizing fish schooling behavior. We briefly discuss possible applications of the HMM for recognition of fish species in a school, and for developing bio-monitoring systems to determine water quality.     

Implied Filtering Densities on the Hidden State of Stochastic Volatility

Abstract

We formulate and analyse an inverse problem using derivative prices to obtain an implied filtering density on volatility’s hidden state. Stochastic volatility is the unobserved state in a hidden Markov model (HMM) and can be tracked using Bayesian filtering. However, derivative data can be considered as conditional expectations that are already observed in the market, and which can be used as input to an inverse problem whose solution is an implied conditional density on volatility. Our analysis relies on a specification of the martingale change of measure, which we refer to as separability. This specification has a multiplicative component that behaves like a risk premium on volatility uncertainty in the market. When applied to SPX options data, the estimated model and implied densities produce variance-swap rates that are consistent with the VIX volatility index. The implied densities are relatively stable over time and pick up some of the monthly effects that occur due to the options’ expiration, indicating that the volatility-uncertainty premium could experience cyclic effects due to the maturity date of the options.     

Consistency analysis of the spectrum and prosody within a syllable for Mandarin speech

This work presents a study of Mandarin speech focusing on consistency analysis of the spectrum and prosody within syllables. Identified as a result of inspection of the human pronunciation process, this consistency can be interpreted as a high correlation between the warping curves of the spectrum and the prosody intra a syllable. The consistency analysis consisted of three steps. First, the hidden Markov model algorithm was used to decode the hidden Markov model‐state sequences within a syllable, while at the same time dividing them into three segments. Second, based on a designated syllable, the vector quantization (VQ) with the Linde–Buzo–Gray algorithm was employed to train the VQ codebooks of the prosodic vector of each segment. Third, the prosodic vector of each segment was encoded as an index using the VQ codebooks, and then, to analyze the consistency, the probability of each possible path was evaluated as a prerequisite. Finally, two syllables were used as examples to verify the consistency property found in the experiments. It is demonstrated experimentally that there is definitely consistency in the case where the syllable is located in exactly the same word. These results offer a research direction in that the warping process between the spectrum and the prosody intra a syllable must be considered in text‐to‐speech systems to improve the synthesized speech quality. Copyright © 2013 John Wiley & Sons, Ltd.     

重庆方言语音识别系统的设计与实现

语音识别赋予了计算机能够识别出语音内容的功能,是人机交互技术领域的重要研究内容。随着计算机技术的发展,语音识别已经得到了成熟的发展。但是关于方言的语音识别还有很大的发展空间。中国是一个幅员辽阔、人口众多的国家,因此方言种类繁多,其中有3000多万人交流使用的重庆方言就是其中之一。采集了重庆方言的部分词语的文本文件和对应的语音文件建立语料库,根据重庆方言的发音特点,选取重庆方言的声韵母作为声学建模基元,选取隐马尔可夫模型(Hidden Markov Model, HMM)为声学模型设计了一个基于HMM的重庆方言语音识别系统。在训练过程利用语料库中训练集语料对声学模型进行训练,形成HMM模型库;在识别过程利用语料库中的测试集语料进行识别测试。实验结果表明,该系统能够实现重庆方言的语音识别,并且识别的正确率为100%。     

一种基于SSM的HMM训练算法

在Baum-Welch(BW)算法的基础上提出了一种基于态相关方法(State—Specific Method：SSM)的隐马尔可夫模型(Hidden Markov Mode：HMM)参数估计算法(简称SBW算法)．该算法在估计HMM不同状态的概率密度函数(probability density function：PDF)的参数时使用了与状态有关的维数较低的特征集合．与传统的BW算法相比，新算法避免了直接估计高维的PDF参数．仿真实验表明，在训练数据量不足的情况下，采用SBW算法的误识率明显低于BW算法．