The performance of recently developed Large Signal (LS) HBT model was evaluated with extensive LS measurements like Power spectrum, Load pull and Inter-modulation investigations. Proposed model has adopted temperature dependent leakage resistance and a simplified capacitance models. The model was implemented in ADS as SDD. Important feature of the model is that the main model parameters are taken directly from measurements in rather simple and understandable way. Results show good accuracy despite the simplicity of the model. To our knowledge the HBT model is one of a few HBT models which can handle high current & Power HBT devices, with significantly less model parameters with good accuracy.

We propose a noise suppression algorithm with the Kalman filter theory. The algorithm aims to achieve robust noise suppression for the additive white and colored disturbance from the canonical state space models with (i) a state equation composed of the speech signal and (ii) an observation equation composed of the speech signal and additive noise. The remarkable features of the proposed algorithm are (1) applied to adaptive white and colored noises where the additive colored noise uses babble noise, (2) realization of high performance noise suppression without sacrificing high quality of the speech signal despite simple noise suppression using only the Kalman filter algorithm, while many conventional methods based on the Kalman filter theory usually perform the noise suppression using the parameter estimation algorithm of AR (auto-regressive) system and the Kalman filter algorithm. We show the effectiveness of the proposed method, which utilizes the Kalman filter theory for the proposed canonical state space model with the colored driving source, using numerical results and subjective evaluation results.

Software rejuvenation is a preventive and proactive solution that is particularly useful for counteracting the phenomenon of software aging. In this article, we consider periodic software rejuvenation models based on the expected cost per unit time in the steady state under discrete-time operation circumstance. By applying the discrete renewal reward processes, we describe the stochastic behavior of a telecommunication billing application with a degradation mode, and determine the optimal periodic software rejuvenation schedule minimizing the expected cost. Similar to the earlier work by the same authors, we develop a statistically non-parametric algorithm to estimate the optimal software rejuvenation schedule, by applying the discrete total time on test concept. Numerical examples are presented to estimate the optimal software rejuvenation schedules from the simulation data. We discuss the asymptotic behavior of estimators developed in this paper.

Conditional random fields (CRFs) have been successfully applied to various applications of predicting and labeling structured data, such as natural language tagging & parsing, image segmentation & object recognition, and protein secondary structure prediction. The key advantages of CRFs are the ability to encode a variety of overlapping, non-independent features from empirical data as well as the capability of reaching the global normalization and optimization. However, estimating parameters for CRFs is very time-consuming due to an intensive forward-backward computation needed to estimate the likelihood function and its gradient during training. This paper presents a high-performance training of CRFs on massively parallel processing systems that allows us to handle huge datasets with hundreds of thousand data sequences and millions of features. We performed the experiments on an important natural language processing task (text chunking) on large-scale corpora and achieved significant results in terms of both the reduction of computational time and the improvement of prediction accuracy.

This paper presents a methodology based on congruent transformation for distributed interconnects described by state-space time-delays system. The proposed approach is to obtain the passive reduced order of linear time-delays system. The unified formulations are used to satisfy the passive preservation. The details of the mathematical proof and a couple of validation examples are given in this paper.

The purpose of the study is to obtain the automatic and optimal matching between a motion-measurement device such as a data glove and an output device such as a dexterous robot hand, where there are many differences in the numbers of degree of freedom, sensor and actuator positions, and data format, by means of motion imitation by the humans. Through the algorithm proposed here, a system engineer or user need no labor of determining the values of gains and parameters to be used. In the system, a subject with data glove imitated the same motion with a dexterous robot hand which was moving according to a certain mathematical function. Autoregressive models were adapted to the matching, where each joint angle in the robot and data glove data of the human were used as object and explanatory variables respectively. The partial regression coefficients were estimated by means of singular value decomposition with a system-noise reduction algorithm utilizing statistical properties. The experimental results showed that the robot hand was controlled with high accuracy with small delay, suggesting that the method proposed in this study is proper and easy way and is adaptive to many other systems between a pair of motion-measurement device and output device.

There has been significant interest in developing new forms of acoustic model, in particular models which allow additional dependencies to be represented than those contained within a standard hidden Markov model (HMM). This paper discusses one such class of models, augmented statistical models. Here, a local exponential approximation is made about some point on a base model. This allows additional dependencies within the data to be modelled than are represented in the base distribution. Augmented models based on Gaussian mixture models (GMMs) and HMMs are briefly described. These augmented models are then related to generative kernels, one approach used for allowing support vector machines (SVMs) to be applied to variable length data. The training of augmented statistical models within an SVM, generative kernel, framework is then discussed. This may be viewed as using maximum margin training to estimate statistical models. Augmented Gaussian mixture models are then evaluated using rescoring on a large vocabulary speech recognition task.

Since their inception almost fifty years ago, hidden Markov models (HMMs) have have become the predominant methodology for automatic speech recognition (ASR) systems--today, most state-of-the-art speech systems are HMM-based. There have been a number of ways to explain HMMs and to list their capabilities, each of these ways having both advantages and disadvantages. In an effort to better understand what HMMs can do, this tutorial article analyzes HMMs by exploring a definition of HMMs in terms of random variables and conditional independence assumptions. We prefer this definition as it allows us to reason more throughly about the capabilities of HMMs. In particular, it is possible to deduce that there are, in theory at least, no limitations to the class of probability distributions representable by HMMs. This paper concludes that, in search of a model to supersede the HMM (say for ASR), rather than trying to correct for HMM limitations in the general case, new models should be found based on their potential for better parsimony, computational requirements, and noise insensitivity.

In this paper, we describe a parallel decoding-based ASR system developed of ATR that is robust to noise type, SNR and speaking style. It is difficult to recognize speech affected by various factors, especially when an ASR system contains only a single acoustic model. One solution is to employ multiple acoustic models, one model for each different condition. Even though the robustness of each acoustic model is limited, the whole ASR system can handle various conditions appropriately. In our system, there are two recognition sub-systems which use different features such as MFCC and Differential MFCC (DMFCC). Each sub-system has several acoustic models depending on SNR, speaker gender and speaking style, and during recognition each acoustic model is adapted by fast noise adaptation. From each sub-system, one hypothesis is selected based on posterior probability. The final recognition result is obtained by combining the best hypotheses from the two sub-systems. On the AURORA-2J task used widely for the evaluation of noise robustness, our system achieved higher recognition performance than a system which contains only a single model. Also, our system was tested using normal and hyper-articulated speech contaminated by several background noises, and exhibited high robustness to noise and speaking styles.

Genetic algorithms are a general problem-solving technique that has been widely used in computational biology. In this paper, we present a framework to map hierarchical parallel genetic algorithms for protein folding problems onto computational grids. By using this framework, the two level communication parts of hierarchical parallel genetic algorithms are separated. Thus both parts of the algorithm can evolve independently. This permits users to experiment with alternative communication models on different levels conveniently. The underlying programming techniques are based on generic programming, a programming technique suited for the generic representation of abstract concepts. This allows the framework to be built in a generic way at application level and thus provides good extensibility and flexibility. Experiments show that it can lead to significant runtime savings on PC clusters and computational grids.

There is increasing evidence from in vivo recordings in monkeys trained to respond to stimuli by making left- or rightward eye movements, that firing rates in certain groups of neurons in oculo-motor areas mimic drift-diffusion processes, rising to a (fixed) threshold prior to movement initiation. This supplements earlier observations of psychologists, that human reaction-time and error-rate data can be fitted by random walk and diffusion models, and has renewed interest in optimal decision-making ideas from information theory and statistical decision theory as a clue to neural mechanisms. We review results from decision theory and stochastic ordinary differential equations, and show how they may be extended and applied to derive explicit parameter dependencies in optimal performance that may be tested on human and animal subjects. We then briefly describe a biophysically-based model of a pool of neurons in locus coeruleus, a brainstem nucleus implicated in widespread norepinephrine release. This neurotransmitter can effect transient gain changes in cortical circuits of the type that the abstract drift-diffusion analysis requires. We also describe how optimal gain schedules can be computed in the presence of time-varying noisy signals. We argue that a rational account of how neural spikes give rise to simple behaviors is beginning to emerge.

Seasonal ARIMA model is a good traffic model capable of capturing the behavior of a network traffic stream. In this paper, we give a general expression of seasonal ARIMA models with two periodicities and provide procedures to model and to predict traffic using seasonal ARIMA models. The experiments conducted in our feasibility study showed that seasonal ARIMA models can be used to model and predict actual wireless traffic such as GSM traffic in China.

In this paper, a source coding model for learning multiple concept descriptions of data is proposed. Our source coding model is based on the concept of transmitting data over multiple channels, called multiple description (MD) coding. In particular, frame expansions have been used in our MD coding models for pattern classification. Using this model, there are several interesting properties within a class of multiple classifier algorithms that share with our proposed scheme. Generalization of the MD view under an extension of local discriminant basis towards the theory of frames allows the formulation of a generalized class of low-complexity learning algorithms applicable to high-dimensional pattern classification. To evaluate this approach, performance results for automatic target recognition (ATR) are presented for synthetic aperture radar (SAR) images from the MSTAR public release data set. From the experimental results, our approach outperforms state-of-the-art methods such as conditional Gaussian signal model, Adaboost, and ECOC-SVM.

This paper introduces the concept of measuring double directional channels in ultra wideband (UWB) systems. Antenna-independent channel data were derived by doing the measurements in a wooden Japanese house. The data were useful for investigating the impact of UWB antennas and analyzing waveform distortion. Up to 100 ray paths were extracted using the SAGE algorithm and they were regarded as being dominant. The paths were then identified in a real environment, in which clusterization analyses were done using the directional information on both sides of the radio link. Propagating power was found to be concentrated around the specular directions of reflection and diffraction. This led to the observation that the spatio-temporal characteristics of extracted paths greatly reflected the structure and size of the environment. The power in the clusters indicated that the estimated 100 paths contained 73% of the total received power, while the rest existed as diffuse scattering, i.e., the accumulation of weaker paths. The practical limits of path extraction with SAGE were also discussed. Finally, we derived the scattering loss and intra-cluster properties for each reflection order, which were crucial for channel reconstrucion based on the deterministic approach.

This paper describes an accurate unsupervised speaker adaptation method for lecture style spontaneous speech recognition using multiple LVCSR systems. In an unsupervised speaker adaptation framework, the improvement of recognition performance by adapting acoustic models remarkably depends on the accuracy of labels such as phonemes and syllables. Therefore, extraction of the adaptation data guided by confidence measure is effective for unsupervised adaptation. In this paper, we looked for the high confidence portions based on the agreement between two LVCSR systems, adapted acoustic models using the portions attached with high accurate labels, and then improved the recognition accuracy. We applied our method to the Corpus of Spontaneous Japanese (CSJ) and the method improved the recognition rate by about 2.1% in comparison with a traditional method.

This paper presents speech-driven Web retrieval models which accept spoken search topics (queries) in the NTCIR-3 Web retrieval task. The major focus of this paper is on improving speech recognition accuracy of spoken queries and then improving retrieval accuracy in speech-driven Web retrieval. We experimentally evaluated the techniques of combining outputs of multiple LVCSR models in recognition of spoken queries. As model combination techniques, we compared the SVM learning technique with conventional voting schemes such as ROVER. In addition, for investigating the effects on the retrieval performance in vocabulary size of the language model, we prepared two kinds of language models: the one's vocabulary size was 20,000, the other's one was 60,000. Then, we evaluated the differences in the recognition rates of the spoken queries and the retrieval performance. We showed that the techniques of multiple LVCSR model combination could achieve improvement both in speech recognition and retrieval accuracies in speech-driven text retrieval. Comparing with the retrieval accuracies when an LM with a 20,000/60,000 vocabulary size is used in an LVCSR system, we found that the larger the vocabulary size is, the better the retrieval accuracy is.

This paper proposes a novel multi-layer approach to fundamental frequency modeling for concatenative speech synthesis based on a statistical learning technique called additive models. We define an additive F0 contour model consisting of long-term, intonational phrase-level, component and short-term, accentual phrase-level, component, along with a least-squares error criterion that includes a regularization term. A backfitting algorithm, that is derived from this error criterion, estimates both components simultaneously by iteratively applying cubic spline smoothers. When this method is applied to a 7,000 utterance Japanese speech corpus, it achieves F0 RMS errors of 28.9 and 29.8 Hz on the training and test data, respectively, with corresponding correlation coefficients of 0.806 and 0.777. The automatically determined intonational and accentual phrase components turn out to behave smoothly, systematically, and intuitively under a variety of prosodic conditions.

This paper describes a new probabilistic sentence reduction method using maximum entropy model. In contrast to previous methods, the proposed method has the ability to produce multiple best results for a given sentence, which is useful in text summarization applications. Experimental results show that the proposed method improves on earlier methods in both accuracy and computation time.

This paper presents CrtSmile--a CAD tool for the automatic extraction of layout-dependent substrate effects for RF MOSFET modeling. CrtSmile incorporates a new scalable substrate model, which depends not only on the geometric layout information of a transistor (the number of gate fingers, finger width, channel length and bulk contact location), but also on the transistor layout and bulk patterns. We show that this model is simple to extract and has good agreement with measured data for a 0.35 µm CMOS process. CrtSmile reads in the layout information of RF transistors in the CIF/GDSII format, performs a pattern-based layout extraction to recognize the transistor layout and bulk patterns. A scalable layout-dependent substrate model is automatically generated and attached to the standard BSIM3 device model as a sub-circuit for use in circuit simulation. A low noise amplifier is evaluated with the proposed CrtSmile tool, showing the importance of layout effects for RF transistor substrate modeling.

In this paper, we propose a probabilistic feature-based parsing model for head-final languages, which can lead to an improvement of syntactic disambiguation while reducing the parsing cost related to lexical information. For effective syntactic disambiguation, the proposed parsing model utilizes several useful features such as a syntactic label feature, a content feature, a functional feature, and a size feature. Moreover, it is designed to be suitable for representing word order variation of non-head words in head-final languages. Experimental results show that the proposed parsing model performs better than previous lexicalized parsing models, although it has much less dependence on lexical information.