A new algorithm is proposed for the simultaneous estimation of poles and zeros in speech analysis. The algorithm is based on estimating the unknown system input and improvement of this estimate through the iterations....
详细信息
A new algorithm is proposed for the simultaneous estimation of poles and zeros in speech analysis. The algorithm is based on estimating the unknown system input and improvement of this estimate through the iterations. Thus, the algorithm does not require any type of preliminary deconvolution of the speech waveform, such as synchronization with pitch period or homomorphic deconvolution. Detailed analysis of a simulated system, as well as a preliminary analysis of initial nasal consonants /m/ and /n/, are presented. These analyses have shown that the ITIF algorithm gives a very accurate fit of the spectra of the systems analyzed. The iterative inverse filtering algorithm (ITIF) is a new technique for modeling linear systems having unknown input with a flat spectral envelope, such as pulse train or white noise input, by applying the pole-zero model. The ITIF algorithm in each iteration solves two linear parameter estimation problems: in the first one the unknown system input is estimated; in the second one the estimated input is used to determine the parameters of the pole-zero model. Experiments made so far have shown that the algorithm converges in a small number of iterations.
This paper makes an attempt to develop least squares lattice algorithms for the ARMA modeling of a linear, slowly time-varying, multichannel system employing scalar computations only. Using an equivalent scalar, perio...
详细信息
This paper makes an attempt to develop least squares lattice algorithms for the ARMA modeling of a linear, slowly time-varying, multichannel system employing scalar computations only. Using an equivalent scalar, periodic ARMA model and a circular delay operator, the signal set for each channel is defined in terms of circularly delayed input and output vectors corresponding to that channel. The orthogonal projection of each current output vector on the subspace spanned by the corresponding signal set is then computed in a manner that allows independent AR and MA order recursions. The resulting lattice algorithm can be implemented in a parallel architecture employing one processor per channel with the data flowing amongst them in a circular manner. The evaluation of the ARMA parameters from the lattice coefficients follows the usual step-up algorithmic approach but requires, in addition, the circulation of certain variables across the processors since the signal sets become linearly dependent beyond certain stages. The proposed algorithm can also be used to estimate a process from two correlated, multichannel processes adaptively allowing the filter orders for both the processes to be chosen independently of each other. This feature is further exploited for ARMA modeling a given multichannel time series with unknown, white input.
When the statistics of the system modeling error, the measurement errors, and the communication network uncertainties are unknown, the distributed Kalman filtering in the multicoordinated systems becomes suboptimal or...
详细信息
When the statistics of the system modeling error, the measurement errors, and the communication network uncertainties are unknown, the distributed Kalman filtering in the multicoordinated systems becomes suboptimal or nonoptimal depending upon the prior information about the unknown statistics. An algorithm for estimation of the optimal steady state gains of distributed filtering and adaptive distributed filtering in the multi-coordinated systems is developed using the correlation method. The algorithm starts with the prior information about the unknown statistics, and adaptively adjusts the weights for the best integration of the multiple measurement sequences.
This paper provides the basic tools required for an efficient use of the recently proposed fast FIR algorithms. These algorithms not only reduce arithmetic complexity but also partially maintain the multiply-accumulat...
详细信息
This paper provides the basic tools required for an efficient use of the recently proposed fast FIR algorithms. These algorithms not only reduce arithmetic complexity but also partially maintain the multiply-accumulate structure, thus resulting in efficient implementations. A set of basic algorithms is derived, together with some rules for combining them. Their efficiency is compared with that of classical schemes in the case of three different criteria, corresponding to various types of implementation. It is shown that this class of algorithms (which includes classical ones as special cases) makes it possible to find the best tradeoff corresponding to any criterion.
In this work, we describe an implementation of the 2D Tikhonov regularization filter which scales lin-early with the input signal's size. In the homogeneous case, we propose a novel algorithm to decompose the filt...
详细信息
In this work, we describe an implementation of the 2D Tikhonov regularization filter which scales lin-early with the input signal's size. In the homogeneous case, we propose a novel algorithm to decompose the filter's 2D kernel as a sum of axis-aligned Gaussians. Our algorithm uses symmetries of the kernel to provide a fast computation of the Gaussian decomposition in the frequency domain, where the 2D Tikhonov kernel has a closed-form expression. The convolution with each Gaussian is then computed using linear-time separable recursive filtering. In the non-homogeneous case, we also decompose the 2D problem as a series of iterated linear-time separable recursive filters, which can be combined with the Bi-conjugate Gradient Stabilized method for fast convergence. In this way, a fast solution to the 2D Tikhonov regularization problem is obtained. (c) 2023 Elsevier B.V. All rights reserved.
Dynamic systems of graph signals are encountered in various applications, including social networks, power grids, and transportation. While such systems can often be described as state space (SS) models, tracking grap...
详细信息
Dynamic systems of graph signals are encountered in various applications, including social networks, power grids, and transportation. While such systems can often be described as state space (SS) models, tracking graph signals via conventional tools based on the Kalman filter (KF) and its variants is typically challenging. This is due to the nonlinearity, high dimensionality, irregularity of the domain, and complex modeling associated with real-world dynamic systems of graph signals. In this work, we study the tracking of graph signals using a hybrid model-based/data-driven approach. We develop the GSP-KalmanNet, which tracks the hidden graphical states from the graphical measurements by jointly leveraging graph signal processing (GSP) tools and deep learning (DL) techniques. The derivations of the GSP-KalmanNet are based on extending the KF to exploit the inherent graph structure via designing a graph frequency domain filtering and replacing the Kalman gain (KG) with a graph filter that minimizes the prediction error. Thus, it considerably simplifies the computational complexity entailed in processing high-dimensional signals and increases the robustness to small topology changes. Then, we use data to learn the KG, namely, the graph filter, following the recently proposed KalmanNet framework, which copes with partial and approximated modeling, without forcing a specific model over the noise statistics. Restricting the KG to a graph filter in the proposed GSP-KalmanNet reduces learned parameters, thereby enhancing stability. Our empirical results demonstrate that the GSP-KalmanNet achieves enhanced accuracy and run time performance, and improved robustness to model misspecifications compared with both model-based and data-driven benchmarks.
Fault diagnosis and condition monitoring of rotating machinery has drawn considerable attention. The complex structure of rotating machinery and poor working conditions cause two challenges: weak signature detection (...
详细信息
Fault diagnosis and condition monitoring of rotating machinery has drawn considerable attention. The complex structure of rotating machinery and poor working conditions cause two challenges: weak signature detection (WSD) and weak compound fault separation (WCFS). A superior method should realize these two functions simultaneously. This paper proposed a multidimensional blind deconvolution method based on cross-sparse filtering (Cr-SF) for WSD and WCFS, which can enhance the weak signature and decompose the different components from compound fault adaptively without any preprocessing and priori knowledge. Cross kurtosis pursuit (CKP), a novel filter selection technology, is proposed for determining the final filters. The experimental and simulated signals verified the performance of the proposed algorithm. The robustness is also investigated using the success rate of repeated experiments. The results indicate that Cr-SF can handle different fault compounding modes under the strong noise environment and perform strong robustness and noise adaptability.
The block Z transform (BZT) is presented. It is shown that the BZT, used with the modified Fermat number transform (MFNT), is very efficient for FIR filtering of a long impulse response. The BZT takes advantage of the...
详细信息
The block Z transform (BZT) is presented. It is shown that the BZT, used with the modified Fermat number transform (MFNT), is very efficient for FIR filtering of a long impulse response. The BZT takes advantage of the number theoretic transforms (NTT's), namely, the computational efficiency, and overcomes one of the restrictions on the NTT's, namely, the restriction on the length of the impulse response.
In this paper we discuss the use of digital filtering and spectrum estimation techniques for improving the efficiency of the FD-TD algorithm in solving eigenvalue problems. The great improvement of the efficiency of t...
详细信息
In this paper we discuss the use of digital filtering and spectrum estimation techniques for improving the efficiency of the FD-TD algorithm in solving eigenvalue problems. The great improvement of the efficiency of the method is demonstrated by means of both numerical and measurement results. In addition, several improvements to the present FD-TD method for eigenvalue analysis are presented. These include the analysis of open dielectric resonators and the extraction of the resonant frequencies from the FD-TD results. The result for the open dielectric resonator analysis is validated using measured data.
In this article, a distributed filtering problem is studied for a Markov jump system over sensor networks, where measurements are partially disturbed by outliers. A local multiple model filter is designed based on var...
详细信息
In this article, a distributed filtering problem is studied for a Markov jump system over sensor networks, where measurements are partially disturbed by outliers. A local multiple model filter is designed based on variational Bayesian approaches and interacting multiple model methods, the designed filter is able to identify and exclude outliers automatically, so as to mitigate the impact of outliers. A distributed filter is proposed by combining the designed local filter with consensus on information methods. Furthermore, a sufficient condition is given to guarantee the stability of the designed distributed filter, in which the estimation errors of each sensor are bounded in the mean square sense. Finally, both simulations and experiments of target tracking systems are done to show the effectiveness of the designed distributed filter.
暂无评论