无端点检测汉语识别算法的实现及改进——动态时间规整和隐马尔可夫统一模型的应用

语音识别算法中，动态时间规整（ＤＴＷ）和隐马尔可夫模型（ＨＭＭ）是最有效的识别算法，并且两者之间有着本质的联系和内在的统一［１］，据此前期工作中，已经建立了ＤＴＷ和ＨＭＭ的统一模型（ＤＨＵＭ）［２、３］。本文对ＤＨＵＭ进行了改进，在ＤＨＵＭ中引进寂静段自环，并根据汉语语音的特点，提出了一种无端点检测的语音识别算法。在识别过程中，该算法无需确定语音信号起止点位置，而是从寂静段开始，直接按帧提取特征（帧长２０ｍｓ，帧间重叠５０％），特征向量由１５阶倒谱系数和帧平均能量组成。实验中，用ＤＨＵＭ实现了该算法，对９９个相似汉语单字的识别测试结果表明：无端点检测的识别正识率为９４．９５％，正识率下降很少，但不作端点检测却降低了算法的复杂程度。为进一步改善识别性能，特征向量采用一种听觉模型特征，识别器具有更好的鲁棒性，识别率会略有提高。     

基于Legendre多项式的随机连续系统的Markov参数估计

本文在讨论连续Ｗｉｓｅｎｒ过程的Ｌｅｇｅｎｄｒｅ多项式逼近值的相关性和Ｗｉｅｎｅｒ过程扰动下连续线性系统基于该正交多项的最小二乘估计有偏性后，提出了无偏一致的且数估计误差方差最小的Ｍａｒｋｏｖ估计（最小方差估计）算法，并给出本文方法的仿真结果。     

An actor-critic algorithm for constrained Markov decision processes

An actor-critic type reinforcement learning algorithm is proposed and analyzed for constrained controlled Markov decision processes. The analysis uses multiscale stochastic approximation theory and the envelope theorem' of mathematical economics.     

马尔可夫链在产品可靠性预测中的应用

事物未来的状态仅仅受事物现状的影响，而与过去的状态无关，也就是具有马尔可夫性。用马尔可夫链的理论与方法，对具有马尔可夫性产品的可靠性进行预测，既为产品的可靠性设计和产品售后服务的经济性分析提供了参考，也为马尔可夫模型的实际应用拓广了范围。通过实例说明在数学软件的帮助下这种预测方法的简单可行性。     

语音帧间相关信息对基于HMM系统识别精度的影响

本文探讨经短时信号处理后的语音信号帧间相关信息对基于隐马尔可夫模型（ＨＭＭ）的语音识别系统识虽精度的影响，鉴于ＨＭＭ的输出独立假设导致语音帧间相关信息的损失，本文提出了一种描述帧间相关信息的统计模型－马尔可夫链（ＭＣＭ）用来弥补ＨＭＭ在这方面的缺陷；经非特定人和多话者孤立字实验表明，用ＭＣＭ作为ＨＭＭ的辅助模型，可将原有ＨＭＭ系统的识别率提高约１～６个百分点。     

A note on the first-passage problem and VanMarcke’s approximation — short communication

Given a scalar, stationary, Markov process, this short communication presents a closed-form solution for the first-passage problem for a fixed threshold b. The derivation is based on binary processes and the general formula of Siegert [Siegert AJF. On the first-passage time probability problem. Physical Review 1951; 81:617–23]. The relation for the probability density function of the first-passage time is identical to the commonly used formula that was derived by VanMarcke [VanMarcke E. On the distribution of the first-passage time for normal stationary random processes. Journal of Applied Mechanics ASME 1975; 42:215–20] for Gaussian processes. The present derivation is based on more general conditions and reveals the criteria for the validity of the approximation. Properties of binary processes are also used to derive a hierarchy of upper bounds for any scalar process.     

Markov不等式与H(?)lder不等式的差异性

通过对(李亚普诺夫)不等式的证明,论证了Markov不等式及Holder不等式的差异性,同时讨论了2个不等式的适应范围。     

Driving intention recognition and behaviour prediction based on a double-layer hidden Markov model

We propose a model structure with a double-layer hidden Markov model (HMM) to recognise driving intention and predict driving behaviour. The upper-layer multi-dimensional discrete HMM (MDHMM) in the double-layer HMM represents driving intention in a combined working case, constructed according to the driving behaviours in certain single working cases in the lower-layer multi-dimensional Gaussian HMM (MGHMM). The driving behaviours are recognised by manoeuvring the signals of the driver and vehicle state information, and the recognised results are sent to the upper-layer HMM to recognise driving intentions. Also, driving behaviours in the near future are predicted using the likelihood-maximum method. A real-time driving simulator test on the combined working cases showed that the double-layer HMM can recognise driving intention and predict driving behaviour accurately and efficiently. As a result, the model provides the basis for pre-warning and intervention of danger and improving comfort performance.     

Transit smart card data mining for passenger origin information extraction

The automated fare collection(AFC) system,also known as the transit smart card(SC) system,has gained more and more popularity among transit agencies worldwide.Compared with the conventional manual fare collection system,an AFC system has its inherent advantages in low labor cost and high efficiency for fare collection and transaction data archival.Although it is possible to collect highly valuable data from transit SC transactions,substantial efforts and methodologies are needed for extracting such data because most AFC systems are not initially designed for data collection.This is true especially for the Beijing AFC system,where a passenger’s boarding stop(origin) on a flat-rate bus is not recorded on the check-in scan.To extract passengers’ origin data from recorded SC transaction information,a Markov chain based Bayesian decision tree algorithm is developed in this study.Using the time invariance property of the Markov chain,the algorithm is further optimized and simplified to have a linear computational complexity.This algorithm is verified with transit vehicles equipped with global positioning system(GPS) data loggers.Our verification results demonstrated that the proposed algorithm is effective in extracting transit passengers’ origin information from SC transactions with a relatively high accuracy.Such transit origin data are highly valuable for transit system planning and route optimization.