This paper introduces a robust identification solution for the linear parameter varying Autoregressive Exogenous systems with outlier-contaminated outputs. The Laplace distribution with heavy tails and the expectation...
详细信息
This paper introduces a robust identification solution for the linear parameter varying Autoregressive Exogenous systems with outlier-contaminated outputs. The Laplace distribution with heavy tails and the expectationmaximizationalgorithm are combined to build the robust system identification framework. To overcome the obstacles brought by the outliers, the Laplace distribution which can be decomposed into infinite Gaussian components, is applied to mathematically model the system noise. The problem of parameter estimation is solved using the expectationmaximizationalgorithm, and the equations to infer the system model and noise parameters are simultaneously provided in the developed identification method. Finally, the verification tests performed on a numerical example and a mechanical unit are used to prove the validity of the developed identification method.
Bayesian methods have been extended for the linear system identification problem in the past ten years. The traditional Bayesian identification selects a Gaussian prior and considers the tuning of kernels, i.e., the c...
详细信息
Bayesian methods have been extended for the linear system identification problem in the past ten years. The traditional Bayesian identification selects a Gaussian prior and considers the tuning of kernels, i.e., the covariance matrix of a Gaussian prior. However, Gaussian priors cannot express the system information appropriately for identifying a positive finite impulse response (FIR) model. This paper exploits the truncated Gaussian prior and develops Bayesian identification procedures for positive FIR models. The proposed parameterizations in the truncated Gaussian prior can reflect the decay rate and the correlation of the impulse response of the system to be identified. The expectation-maximization (EM) algorithm is tailored to the hyperparameter estimation problem of positive system identification with the truncated Gaussian prior. Numerical experiments compare the truncated Gaussian prior to the traditional Gaussian prior for positive FIR system identification. The simulation results demonstrate that the truncated Gaussian prior outperforms the Gaussian prior. (C) 2020 Elsevier B.V. All rights reserved.
The study proposed a new crop water stress indicator - the mean value of Gaussian distribution of excess green index for maize canopy (MGDEXG) within an RGB image. A series of RGB images were collected in a maize fiel...
详细信息
The study proposed a new crop water stress indicator - the mean value of Gaussian distribution of excess green index for maize canopy (MGDEXG) within an RGB image. A series of RGB images were collected in a maize field under varying levels of deficit irrigation during 2013, 2015 and 2016 growth seasons in northern Colorado. To evaluate the sensitivity of MGDEXG to maize water status, canopy temperature, canopy-to-air temperature difference, crop water stress index (CWSI), leaf water potential, and sap flow were used as water status references. The results show that MGDEXG distinguished different levels of deficit irrigation treatments well and responded to the release and reimposition of deficit irrigation. The MGDEXG showed a significant correlation (p < 0.01) to different water stress references. Especially, the coefficient of determination (R2) with CWSI was 0.63 (n = 59) for 2013, 0.80 (n = 90) for 2015, and 0.80 (n = 50) for 2016. In addition, among the three Tc-based water stress indicators, the relationship between MGDEXG and CWSI was the most robust with the least annual changes of slope and intercept. The robust relationship between MGDEXG and CWSI could also show that MGDEXG was resistant to the micro-meteorological conditions within the field. Significant correlations (p < 0.01) were found between MGDEXG and leaf water potential with R2 of 0.85 and 0.87 for 2013 and 2015, and between MGDEXG and sap flow in 2015 (R2 = 0.62). MGDEXG relies only on the distribution of crop pixels within an RGB image and could be calculated easily, so it could be cheaper or easier to popularize than other crop water stress indicators in practice. Overall, our results show that MGDEXG could be successfully used as a maize water stress indicator. In the future, more field experiments are needed to further explore the changes of MGDEXG with different scale and spatial resolution of RGB images, and to evaluate MGDEXG for specific climate and crop varieties.
An improved Gaussian mixture model (GMM)- based clustering method is proposed for the difficult case where the true distribution of data is against the assumed GMM. First, an improved model selection criterion, the ...
详细信息
An improved Gaussian mixture model (GMM)- based clustering method is proposed for the difficult case where the true distribution of data is against the assumed GMM. First, an improved model selection criterion, the completed likelihood minimum message length criterion, is derived. It can measure both the goodness-of-fit of the candidate GMM to the data and the goodness-of-partition of the data. Secondly, by utilizing the proposed criterion as the clustering objective function, an improved expectation- maximization (EM) algorithm is developed, which can avoid poor local optimal solutions compared to the standard EM algorithm for estimating the model parameters. The experimental results demonstrate that the proposed method can rectify the over-fitting tendency of representative GMM-based clustering approaches and can robustly provide more accurate clustering results.
For a dynamic process identification throughout the whole operating range under diverse operating conditions, it is difficult to capture the process dynamics by a single process model in which the traditional identifi...
详细信息
For a dynamic process identification throughout the whole operating range under diverse operating conditions, it is difficult to capture the process dynamics by a single process model in which the traditional identification method can be adopted to implement parameter estimation. By using the multiple dual-rate state-space models to approach the parameter-varying time- delay systems with different operating conditions, this paper explores an EM algorithm to simultaneously estimate the hidden variable, the parameter vector, the state variable and the time-delay by introducing hidden variables and by using a Kalman smoother.
With the aid of blanking nonlinearity, the low density parity check (LDPC) coded bit-interleaved coded modulation (BICM) has been jointly considered as a robust mitigation for the impulsive interference in orthogonal ...
详细信息
With the aid of blanking nonlinearity, the low density parity check (LDPC) coded bit-interleaved coded modulation (BICM) has been jointly considered as a robust mitigation for the impulsive interference in orthogonal frequency division multiplexing (OFDM) systems. However, the Gaussian assumption for the nonlinear channel conditional probability induces the mismatched L-values in the conventional MAP demodulator. In this paper, combined with the pulse blanking optimization via the PEXIT analysis, we propose a novel MAP demodulator based on the Gaussian mixture model (GMM) and estimate the parameters with the expectation-maximization (EM) algorithm. Taking the L-band Digital Aeronautical Communication System Type1 (L-DACS1) as an example, the GMM-based MAP demodulator can obtain the PEXIT thresholds that match the decoding curves well and provide the better BER performance in the interference-limit channel environment.
Background: Knowledge of HLA haplotypes is helpful in many settings as disease association studies, population genetics, or hematopoietic stem cell transplantation. Regarding the recruitment of unrelated hematopoietic...
详细信息
Background: Knowledge of HLA haplotypes is helpful in many settings as disease association studies, population genetics, or hematopoietic stem cell transplantation. Regarding the recruitment of unrelated hematopoietic stem cell donors, HLA haplotype frequencies of specific populations are used to optimize both donor searches for individual patients and strategic donor registry planning. However, the estimation of haplotype frequencies from HLA genotyping data is challenged by the large amount of genotype data, the complex HLA nomenclature, and the heterogeneous and ambiguous nature of typing records. Results: To meet these challenges, we have developed the open-source software Hapl-o-Mat. It estimates haplotype frequencies from population data including an arbitrary number of loci using an expectation-maximization algorithm. Its key features are the processing of different HLA typing resolutions within a given population sample and the handling of ambiguities recorded via multiple allele codes or genotype list strings. Implemented in C++, Hapl-o-Mat facilitates efficient haplotype frequency estimation from large amounts of genotype data. We demonstrate its accuracy and performance on the basis of artificial and real genotype data. Conclusions: Hapl-o-Mat is a versatile and efficient software for HLA haplotype frequency estimation. Its capability of processing various forms of HLA genotype data allows for a straightforward haplotype frequency estimation from typing records usually found in stem cell donor registries.
In this paper, a generalized autoregressive conditional heteroskedasticity model under skew-normal distributions is studied. A maximum likelihood approach is taken and the parameters in the model are estimated based o...
详细信息
In this paper, a generalized autoregressive conditional heteroskedasticity model under skew-normal distributions is studied. A maximum likelihood approach is taken and the parameters in the model are estimated based on the expectation-maximization algorithm. The statistical diagnostics is made through the local influence technique, with the normal curvature and diagnostics results established for the model under four perturbation schemes in identifying possible influential observations. A simulation study is conducted to evaluate the performance of our proposed method and a real-world application is presented as an illustrative example.
A new hybrid evolutionary algorithm (EA) for Gaussian mixture model-based clustering is proposed. The EA is a steady-state method that, in each generation, selects two individuals from a population, creates two offspr...
详细信息
A new hybrid evolutionary algorithm (EA) for Gaussian mixture model-based clustering is proposed. The EA is a steady-state method that, in each generation, selects two individuals from a population, creates two offspring using either mutation or crossover, and fine-tunes the offspring using the expectationmaximization (EM) algorithm. The offspring compete with their parents for survival into the next generation. The approach proposed uses a random swap, which replaces a component mean with a randomly chosen feature vector as a mutation operator. In the crossover operator, a random component is copied from the source mixture into a destination mixture. Copying in crossover favors components of the source mixture located away from the components of the destination mixture. In computational experiments, the approach was compared to a multiple restarts EM algorithm, a random swap EM method, and a state-of-the-art hybrid evolutionary algorithm for Gaussian mixture model learning on one real and 29 synthetic datasets. The results indicate that, given the same computational budget, the proposed method usually learns mixtures with higher log-likelihoods than other benchmarks competing algorithms. The partitions of data obtained by the method correspond best to the original divisions of datasets into classes.
In this paper, the expectation-maximization (EM) algorithm and adaptive type-II progressive censoring (A-II-PC) scheme are discussed for the first time under ramp-stress accelerated life testing (RS-ALT) experiment ba...
详细信息
In this paper, the expectation-maximization (EM) algorithm and adaptive type-II progressive censoring (A-II-PC) scheme are discussed for the first time under ramp-stress accelerated life testing (RS-ALT) experiment based on extended Weibull (EW) distribution. Under the assumptions of cumulative exposure (CE) model and inverse power law relationship, the maximum likelihood estimators (MLEs) of the unknown parameters and acceleration factors are obtained using the EM algorithm, then compared with the scoring algorithm. Furthermore, the observed information matrix based on the missing value principle is computed and used to construct the asymptotic confidence intervals for the parameters and acceleration factors. Bootstrap techniques are also considered in the interval estimation. Moreover, a numerical example is presented to illustrate the application of both the scoring and EM algorithms. Finally, a Monte Carlo simulation study is carried out to assess the performance of the suggested algorithms.
暂无评论