The Levinson algorithm, implemented via a ladder filter, has been widely used in seismic signalprocessing and more recently in speech analysis. This algorithm was originally developed (in 1947) for prediction of stat...
详细信息
The Levinson algorithm, implemented via a ladder filter, has been widely used in seismic signalprocessing and more recently in speech analysis. This algorithm was originally developed (in 1947) for prediction of stationary stochastic processes. It has been adapted to applications where only a single sample function (time-series) is available by the use of appropriate "window functions" to "simulate stationarity". Such windowing is often somewhat artificial and can introduce undesirable artefacts in situations where only a limited amount of data is available. However, more realistic "windowing" will destroy the analogy to the stationary stochastic process case and thus apparently prohibit use of the Levinson-ladder- filter implementations. In this talk, we shall first show how by using a concept of how close a given process is to being stationary, we can develop generalized Levinson algorithms and generalized ladder filters applicable to any stochastic process, stationary or not. The complexity of the algorithm and the filter is proportional to the distance from stationarity of the given process. We shall then show that this structure can include analyses of several windowing strategies for applications where only a single observed record is available, thus including results obtained earlier for such problems by Morf, Vieira and Lee. The present results are based on work with D. Lee and H. Lev-Ari.
Machine learning and related statistical signalprocessing are expected to endow sensor networks with adaptive machine intelligence and greatly facilitate the Internet of Things. As such, architectures embedding adapt...
详细信息
Machine learning and related statistical signalprocessing are expected to endow sensor networks with adaptive machine intelligence and greatly facilitate the Internet of Things. As such, architectures embedding adaptive and learning algorithms on-chip are oft-ignored by the design community and present a new set of design trade-offs. This review focuses on efficient implementation of mixed-signal matrix-vector multiplication as a central computational primitive enabling machine learning and statistical signalprocessing, with specific examples in spatial filtering for adaptive beamforming. We describe adaptive algorithms amenable for efficient implementation with such primitives in the presence of noise and analog variability. We also briefly highlight current trends in high-density integration in emerging memory device technologies and their use in highdimensional adaptive computing.
A comparison is made of three characteristic algorithms for adaptive signalprocessing in ISDN (integrated-services digital network) user interfaces: a fast RLS (recursive least squares) algorithm with transversal FIR...
详细信息
A comparison is made of three characteristic algorithms for adaptive signalprocessing in ISDN (integrated-services digital network) user interfaces: a fast RLS (recursive least squares) algorithm with transversal FIR (finite impulse response) structure, an LMS (least mean squares) algorithm with transversal FIR structure, and a sign algorithm with memory filter structure. The algorithms are compared by using a simulated system for data transmission, which is applied to the study of the characteristics of adaptive algorithms and for the education of the students in the area of telecommunications.< >
A signalprocessing system employing digital arithmetic can be made equivalent to a set of interconnected subsystems each using finite field operations to alleviate effects of round-off noise, particularly in recursiv...
详细信息
A signalprocessing system employing digital arithmetic can be made equivalent to a set of interconnected subsystems each using finite field operations to alleviate effects of round-off noise, particularly in recursive structures. The realization of the subsystems involves the multiplication of polynomials over a finite field. Cyclic convolutions may be represented as polynomial multiplications, in a residue class ring modulo the polynomial (x k -1). Fast algorithms arise when the ring is decomposed into a direct product using its minimal ideals, the orthogonality relationships between the idempotents of the minimal ideals resembling the decomposition properties of the Chinese remainder theorem for real polynomials. The resulting component polynomial multiplications involve elements with smaller degrees. A transform domain, based upon an n th root of unity in an extension field, offers an alternative domain for performing the component operations. Polynomial multiplications are equivalent to the vector inner product of the corresponding transform coefficients. The minimal ideal components, when mapped into the transform domain, have nonzero coefficients only in predetermined subsets, the cyclotomic subsets, which permits one coefficient to generate all others by successively forming its powers.
According to global statistics and the world health organization (WHO), about 17.5 million people die each year from cardiovascular disease. In this paper, the heart sounds gathered by a stethoscope are analyzed to di...
详细信息
According to global statistics and the world health organization (WHO), about 17.5 million people die each year from cardiovascular disease. In this paper, the heart sounds gathered by a stethoscope are analyzed to diagnose several diseases caused by heart failure. This research's primary process is to identify and classify the data related to the heart sounds categorized in four general groups of S 1 to S 4 . The sounds S 1 and S 2 are considered as the heart's normal sounds, and the sounds S 3 and S 4 are the abnormal sounds of the heart (heart murmurs), each expressing a specific type of heart disease. In this regard, the desired features are first extracted after retrieving the data by signal processing algorithms. In the next step, feature selection algorithms are used to select the compelling features to reduce the problem's dimensions and obtain the optimal answer faster. While the existing algorithms in the literature classify the sound into two groups of normal and abnormal, in the final section, some of the most popular classification algorithms are utilized to classify the type of sound into three classes of normal, S 3 and S 4 categories. The proposed methodology obtained an accuracy rate of 87.5% and 95% for multiclass data (3 classes) and 98% for binary classification (normal vs. abnormal) problems.
The separation of noise from speech has always been a necessary requirement and is being demanded in audio signalprocessing as an important factor to achieve a dear message in voice communication. In the literature, ...
详细信息
The separation of noise from speech has always been a necessary requirement and is being demanded in audio signalprocessing as an important factor to achieve a dear message in voice communication. In the literature, some traditional methods that has been employed include machine learning softwares with basic algorithms. However, with the accelerated development in the field of computer technology especially the artificial intelligence, application of machine learning softwares and neural networks in the domain of audio signalprocessing is still a fascinating field for researchers. This paper provides an experimental simulation of speech with and with out noise using different neural networks to solve the filtering problem. In the course of this research, first an FIR digital filter is designed, since an FIR filter provides a fast convergence and give results near the global optimal. The neural networks such as Elman, perceptron and radial base function are then trained with different training algorithms and compared with the performance of FIR digital filter including their computational complexity. The experimental results exhibits that training/learning junctions chosen to train the neural network are very important to the final results.
The mathematical structure associated with the split algorithms for computing the reflection coefficients for a given real symmetric positive-definite Toeplitz matrix is analyzed. A novel form of three-term recurrence...
详细信息
The mathematical structure associated with the split algorithms for computing the reflection coefficients for a given real symmetric positive-definite Toeplitz matrix is analyzed. A novel form of three-term recurrence relation is derived and computationally efficient alternatives to the Levinson-Durbin, Schur, and lattice algorithms are obtained. The computational complexity of the proposed algorithms is the same as those of the split algorithms described in recent literature. These algorithms provide further insight into the mathematical properties of the structurally rich Toeplitz matrices.< >
This paper compares three different approaches currently used in recognizing contact calls made from the North Atlantic Right Whale (NRW), Eubalaena glacialis. We present two new approaches consisting of machine learn...
详细信息
This paper compares three different approaches currently used in recognizing contact calls made from the North Atlantic Right Whale (NRW), Eubalaena glacialis. We present two new approaches consisting of machine learning algorithms based on artificial neural networks (NET) and the classification and regression tree classifiers (CART), and compare their performance with earlier work that employs multi-Stage feature vector testing (FVT) approach. A combined total of over 100,000 noise and NRW up-call events were used in the study. Calls were primarily recorded from two areas, Cape Cod Bay and Great South Channel. Of the three classifiers, the CART had the highest assignment rates, overall 86.45% with highest false positive rates (
To achieve accurate and fast online fault diagnosis of proton exchange membrane fuel cell (PEMFC) under complex operating conditions, this article proposes a diagnostic method based on voltage-transfer learning convol...
详细信息
To achieve accurate and fast online fault diagnosis of proton exchange membrane fuel cell (PEMFC) under complex operating conditions, this article proposes a diagnostic method based on voltage-transfer learning convolutional neural network (V-TCNN). The sensitive analysis for PEMFC is conducted for pattern identifications first, and the output voltage response which can accurately characterize the internal mechanisms of the fuel cell stacks is applied to improve the fault diagnosis accuracy. Besides, to reduce the computation time and improve the portability of the diagnostic algorithm, transfer learning methods are applied to optimize the conventional neural network (CNN) architecture. The comparison tests show that the proposed algorithm can improve the diagnostic accuracy by 7.5% and reduce the computation time by 20% compared with the current existing deeply layered neural networks. Comprehensive analysis shows that the algorithm can reach a fault diagnosis accuracy of 99.56%, which can help to contribute to the development and implementation of fuel cell online fault diagnosis methods in fuel cell electric vehicles (FCEVs).
暂无评论