Search

In the present article, we derive a new Modified-X (NM-X) family of probability distributions by using a popular approach known as the T-X method. For the NM-X family, some of the main distributional (or statistical) properties such as quantile function, identifiability property, order statistics, residual life and reverse residual life, moments, and moments generating function are determined. A well-known method of estimation such as the Maximum Likelihood estimation method is used for estimating the model parameters of the introduced family of distributions. A Monte Carlo simulation method is incorporated to judge the performance of the estimators with different selected values of parameters. To check the tails behaviors of the proposed model, the risk measurements such as value-at-risk (VaR), tailed-value-at-risk (TVaR), tailed-variance (TV), and tailed-variance-premium (TVP) are investigated. For illustrative purposes, a special sub-model of the proposed class is derived and named a New Modified Weibull distribution (abbreviated as the NM-Weib). Furthermore, four applications from the field of medicine are analyzed to demonstrate the application, and effectiveness of the NM-Weib distribution. By adopting different diagnostic criteria, the NM-Weib distribution is compared with Weibull (Weib) distribution, Alpha Power Transformed-Weib (APTra-Weib) distribution, new Exponent-Weib (NExpo-Weib) distribution, Flexible Reduced Logarithmic-Weib (FRLog-Weib) distribution, Marshal Olkin-Weib (MO-Weib) distribution, and Kumaraswamy-Weib (K-Weib) distribution. Finally, based on the diagnostic criteria, it is observed that the NM-Weib distribution provides a suitable and best fit for each considered biomedical data set. [ABSTRACT FROM AUTHOR]

Extracting high-level structural information from 3D point clouds is challenging but essential for tasks like urban planning or autonomous driving requiring an advanced understanding of the scene at hand. Existing approaches are still not able to produce high-quality results consistently while being fast enough to be deployed in scenarios requiring interactivity. We propose to utilize a novel set of features describing the local neighborhood on a per-point basis via first and second order statistics as input for a simple and compact classification network to distinguish between non-edge, sharp-edge, and boundary points in the given data. Leveraging this feature embedding enables our algorithm to outperform the state-of-the-art technique PCEDNet in terms of quality and processing time while additionally allowing for the detection of boundaries in the processed point clouds. [Display omitted] [ABSTRACT FROM AUTHOR]

In this paper, we introduce weighted bootstrap algorithms for both non-degenerate and degenerate two-sample U -statistics with arbitrary degrees. For the non-degenerate case, weighted bootstrap with dependent weights is introduced as a generalization of Efron's conventional bootstrap. In addition, two weighted bootstrap procedures with independent productive weights and independent additive weights are proposed under non-degeneracy. More importantly, we extend the weighted bootstrap method to two-sample U -statistics under the degeneracy of order 2 with a novel construction of random weights. Theoretical supports of the proposed weighted bootstrap procedures under non-degeneracy and degeneracy of order 2 are established. Numerical studies illustrate that the proposed approaches are feasible and effective for statistical inferences. • Weighted bootstrap for non-degenerate and degenerate two-sample U -statistics. • Weighted bootstrap with independent productive/additive weights under nondegeneracy. • Weighted bootstrap involves a novel construction of random weights under degeneracy. • Proposed weighted bootstrap methods are supported by theoretically and numerically. [ABSTRACT FROM AUTHOR]

Due to advancements in Bayesian modeling, the likelihood-free posterior estimate is now feasible. These estimation methods are crucial for a deeper understanding of simulation-based systems since it might be difficult, if not impossible, to estimate probability values. This compares the effectiveness of such updated estimate approaches to earlier maximum likelihood forecasting models and offers some adjustments to maximum likelihood estimation for estimating the parameters of the Bayesian analysis in this work. Recent improvements in Bayesian modeling have made it feasible to obtain the likelihood-free posterior estimate. These estimate methods are essential for evaluating simulation-based theories since it might be difficult, if not impossible, to determine probability values. Simulation-based concepts such as the Leaky Competing Accumulator (LCA) theory and Feed-Forward Inhibition (FFI) theory have not yet benefited from Bayesian techniques. As assessment criteria, total relative deviation (TRD), total mean square error (TMSE), and Stein Loss Function (SLF) are used. Maximum likelihood (ML) estimates are modified because the original statistic's Cumulative Distribution Function (CDF) is better than traditional ML estimations and other modified estimators emphasize average and coefficient variability. Prior estimator performance was unaffected by response rate or real parameter settings. These systems have not been formally evaluated by the Bayesian factor. [ABSTRACT FROM AUTHOR]

This paper introduces a new three-parameter distribution called the modified Kies Gompertz (MKGO) distribution to identify an effective statistical distribution in a real-life modeling data set. The distribution's probability density function may take numerous forms and graphs, and it could be negatively and positively skewed and uni-modal. We established several statistical properties of the MKGO distribution, such as moments, skewness, kurtosis, moment generating function, distribution of order statistic, quantile function, stress strength reliability and two cases of entropy measures. Further, we proposed six methods for estimating the unknown parameters of the suggested model. A brief Monter Carlo (MC) simulation study is illustrated in terms of average estimate (AE) and mean square error (MSE). to show the efficiency of these estimators. Finally, we analyzed four real data sets in the engineering sector to evaluate the potential of the proposed model with that of various existing distributions. [ABSTRACT FROM AUTHOR]

We investigate extreme value theory of a class of random sequences defined by the all-time suprema of aggregated self-similar Gaussian processes with trend. This study is motivated by its potential applications in various areas and its theoretical interestingness. We consider both stationary sequences and non-stationary sequences obtained by considering whether the trend functions are identical or not. We show that a sequence of suitably normalised k th order statistics converges in distribution to a limiting random variable which can be a negative log transformed Erlang distributed random variable, a Normal random variable or a mixture of them, according to three conditions deduced through the model parameters. Remarkably, this phenomenon resembles that for the stationary Normal sequence. We also show that various moments of the normalised k th order statistics converge to the moments of the corresponding limiting random variable. The obtained results enable us to analyse various properties of these random sequences, which reveals the interesting particularities of this class of random sequences in extreme value theory. [ABSTRACT FROM AUTHOR]

Data-driven soft sensor technology has been widely used in process monitoring, quality prediction, etc. However, there are dynamic time-varying and concept drift problems in industrial processes, making the accurate modeling of the soft sensor is still a challenging task. To solve the above problems, this paper proposes a concept drift adaptive dynamic partial least squares method. The method maps high-dimensional process data into a low-dimensional latent variable subspace. Dynamic information is introduced by establishing the dynamic regression relationship between process latent variables and quality latent variables. At the same time, transfer learning is used to align second-order statistics between latent variables of data from different distributions to solve the concept drift problem. The experiments on multiple industrial datasets show that the method proposed in this paper can effectively reduce prediction errors and improve the generalization ability of the model. [Display omitted] • Distribution alignment and dynamic system modeling are realized under NIPALS. • The objective function is organized into the form of Rayleigh quotient. • There are few hyperparameters and they have better interpretability. • Experiments show that our method can adapt to the change of data distribution well. [ABSTRACT FROM AUTHOR]

We study the family of distributions characterised by an increasing odds rate (IOR), showing that this provides an alternative interpretation of the "adverse ageing" notion, comparable to the well known increasing hazard rate (IHR) model. We prove some preservation properties of this class under several transformations that are often considered in reliability and life testing problems, including formation of order statistics. Moreover, the IOR assumption enables the derivation of survival bounds and tolerance limits, extending the scope of applicability of some known results, which are based, for instance, on the IHR assumption. Finally, we propose a test for the IOR null hypothesis, establishing its consistency and providing a table of simulated p -values. • We propose the IOR condition as an alternative interpretation of the "adverse ageing" notion. • The IOR is shown to satisfy some closure properties, which are of interest in reliability theory. • Bounds and tolerance limits are obtained assuming the IOR condition, and their application is discussed. • A nonparametric test for IOR is proposed and characterised. [ABSTRACT FROM AUTHOR]

• Introduction of a new robust structure similarity index based on p th order statistics. • A simple and low cost special case: l1-norm based robust structure similarity index. • Extension of this measure to colour images. • Demonstration of the success of new index on various types of distortions. Structural Similarity Index Measure (SSIM) has been a very successful tool in image processing and computer vision. After almost two decades, it is still the most popular method for quantifying the similarity between original and distorted images with applications in image quality assessment, image retrieval, video coding, computer vision, image encryption and data-hiding. Despite its general success, there are some types of distortions, such as when the distortion is of non-Gaussian character, where SSIM does not sustain its success. We provide a generalized version of SSIM utilizing l p -norm (1 ≤ p ≤ 2) based sample moments as opposed to the classical SSIM which uses l 2 -norm based sample moments. In particular, we study the l 1 -norm special case and our simulations on well-known image-databases show superior performance compared to SSIM and related measures. [ABSTRACT FROM AUTHOR]

• Prediction of the emotional states in children with ASD using ECG signals. • Customized elicitation protocol using audio and video clips for data acquisition. • Analysis of geometrical and time domain features of positive and negative emotions. • KNN and Ensemble classifier were used for the classifying the two emotional states. • Geometrical features resulted in good accuracy of 84.8% and 74.7% in both states. Autism Spectrum Disorder (ASD) is a neurodevelopmental condition that is characterized by various social impairments. Children with ASD have major difficulties in expressing themselves, resulting in stress and meltdowns. Understanding their hidden feelings and needs may help in tackling and avoiding such strenuous behaviors. This research aims to aid the parents and caretakers of children with ASD to understand the hidden and unexpressed emotional state by using physiological signals obtained from wearable devices. Here, electrocardiogram (ECG) signals pertaining to two valence states ('like' and 'dislike') were recorded from twenty children (10 Control and 10 children with ASD). The heart rate variability (HRV) signals were then obtained from the ECG signals using the Pan-Tompkins's algorithm. The statistical, higher order statistics (HOS) and geometrical features which were statistically significant were trained using the K Nearest Neighbor (KNN) and Ensemble Classifier algorithms. The findings of our analysis indicate that the integration of major statistical features resulted in an overall average accuracy of 84.8% and 75.3% using HRV data for the control and test population, respectively. Similarly, geometrical features resulted in a maximum average accuracy of 84.8% and 74.2% for control and test population respectively. The decreased HRV in the test population indicates the presence of autonomic dysregulation in children with ASD when compared to their control peers. [ABSTRACT FROM AUTHOR]

Conceptually, the alignment of fibers in cementitious matrices leads to an appropriate choice of orientation indices, orientation coefficient, orientation tensors, etc. methods that do not have the advantages offered by descriptive or inferential statistics. The work presented here uses directional statistics as an efficient measure of the alignment of a set of fibers that initially have random orientations. The randomly oriented fibers are immersed in a fluid metaphor (hydrogel) transparent that reproduces some rheological characteristics of cementitious matrices. Magnetic alignment carried out by discharging an Resistor–Capacitor-Inductor-Circuit (RCL circuit) that causes pulses of homogeneous magnetic fields of intensity ranging approximately from 20 mT to 90 mT (peak) changing the final orientation distribution of the set of fibers. These changes in the orientation distribution give rise to their directional statistical analysis, revealing the relationships between the magnetic properties of the fibers and the rheological parameters of the matrix in which they are immersed. Among the studies presented here is the possibility of planning various fiber alignment strategies using a circular diagram as an abacus (Circles of Active/Dead Fibers) that allow us to go beyond the application of a simple pulse and/or a single alignment direction, seeking a process fiber alignment that is optimal and controlled. • Directional statistical method as a procedure for alignment analysis of steel fibers. • Integration of the magnetic properties of fibers with the rheological of the matrix. • Description of circular active/dead orientation sectors. • Eccentric alignment directions coincide with the average circular orientation. • Circular method for the optimal planning of successive alignment processes. [Display omitted] The alignment of reinforce steel fibers in fluid matrices requires knowledge of two opposite properties: the magnetic driving torque provides an alignment character to the steel fibers, this driving torque requires the knowledge of ferromagnetic properties of the fiber material. On the other hand, the fluid matrix opposes the movement with an opposition torque whose characteristic is found in its rheological properties. The study presented here uses directional statistics in order to study which steel fibers within the material are active fibers (they are capable of aligning) and which are dead fibers (incapable of moving due to being subjected to static yield stress). Both types of fibers in the matrix characterize either active sectors with a mean circular direction or dead sectors with fiber that does not participate in the alignment process. Knowing the amplitude of both sectors allows you to plan and manage optimal fiber alignment processes. [ABSTRACT FROM AUTHOR]

We consider the binomial decomposition of ordered weighted averaging (OWA) functions proposed by Calvo and De Baets (1998) in the framework of Choquet integration. Our aim in the paper is to further investigate the equivalence between the two representations of OWA functions involved in the binomial decomposition: the usual canonical representation in terms of the order statistics, and the binomial representation in terms of the binomial OWA functions. We describe and discuss in detail the linear transformations that relate the coefficients of these two equivalent representations: the original expression of the weights in terms of the coefficients of the binomial representation, and its inverse, the expression of those coefficients in terms of the weights. In both cases simple and direct proofs are presented. Moreover, we consider the linear transformations between the two representations in the general linear algebra framework of unconstrained linear combinations of binomial OWA functions. In this perspective we obtain compact matrix expressions for the linear transformations, which also offer new insight on the geometry of the coefficient constraints in the binomial representation of OWA functions. [ABSTRACT FROM AUTHOR]

Traditionally, Convolutional Neural Networks make use of the maximum or arithmetic mean in order to reduce the features extracted by convolutional layers in a downsampling process known as pooling. However, there is no strong argument to settle upon one of the two functions and, in practice, this selection turns to be problem dependent. Further, both of these options ignore possible dependencies among the data. We believe that a combination of both of these functions, as well as of additional ones which may retain different information, can benefit the feature extraction process. In this work, we replace traditional pooling by several alternative functions. In particular, we consider linear combinations of order statistics and generalizations of the Sugeno integral, extending the latter's domain to the whole real line and setting the theoretical base for their application. We present an alternative pooling layer based on this strategy which we name "CombPool" layer. We replace the pooling layers of three different architectures of increasing complexity by CombPool layers, and empirically prove over multiple datasets that linear combinations outperform traditional pooling functions in most cases. Further, combinations with either the Sugeno integral or one of its generalizations usually yield the best results, proving a strong candidate to apply in most architectures. [ABSTRACT FROM AUTHOR]

Ordered random vectors are frequently encountered in many problems. The generalized order statistics (GOSs) and sequential order statistics (SOSs) are two general models for ordered random vectors. However, these two models do not capture the dependency structures that may be present in the underlying random variables. In this paper, we study the developed sequential order statistics (D-SOSs) and developed generalized order statistics (D-GOSs) models that incorporate dependency structures among ordered random vectors. We then study various univariate and multivariate ordering properties of D-SOS and D-GOS models under Archimedean copula. We develop corresponding results for both one-sample and two-sample situations. We also present some simulational results and a real data analysis for illustrative purpose. [ABSTRACT FROM AUTHOR]

There are some situations in reliability studies in which a middle part of the sample is missing. For example, in life-testing experiments, the experimenter may not enable to observe the middle lifetimes. This may be due to negligence, removal of test units (censoring) and or equipment failure and technical problems in the middle of the experiment. In this paper, the problem of reconstructing unobserved middle lifetimes from the exponential distribution is discussed and reconstruction regions (or sets) are obtained. First, it is shown how to find balanced reconstruction regions for the unobserved middle lifetimes. Then, a constrained minimization problem is used to find the reconstruction region with the smallest area. For illustrative and comparative purposes, a simulation study and two numerical examples are presented. • Exact reconstruction regions for unobserved middle lifetimes are proposed. • Constrained optimization problems are stated to determine the minimum-size reconstruction regions. • Efficient procedures based on minimizing Lagrangian functions are proposed. • Two numerical examples and a simulation study are presented for illustrative and comparative purposes. [ABSTRACT FROM AUTHOR]

[Display omitted] • Detailed lithological and structural mapping using textural analysis. • The approach allows for the depiction of structural features and their reverberations within the host rocks. • Pan-sharpening with PRISM data (2.5 m) greatly assists in constructing a primitive microstructural pattern. Detailed geological mapping is the decisive key for mineral deposit prospecting, deciphering tectonic models, and outlining the main framework for most development programs and constructions. Without a doubt, various remote sensing datasets have introduced reliable lithological and structural mapping solutions. The main defect with remote sensing data is the curse of dimensionality, especially with hyperspectral data, where a lot of time is spent handling and selecting representative bands for various geological analyses. Consequently, and for the first time, our research is an attempt to resolve the complicated structural patterns (lineaments, folds, foliations, and cross-cutting relationships) and enhance lithological discrimination using a single band (of Sentinel 2 and ALOS PRISM data) and textural analysis. Through several trials over different pixel sizes (2.5 m and 10 m) and various kernels (3 × 3, 7 × 7, or 11 × 11), reasonable results are obtained, enabling lithological discrimination, in-depth structural analysis (foliations, faults, joints, and folds), shape recognition of systematic rock bodies, and delineation of quaternary deposits using single band analysis rather than time-consuming multiple band processing. Our results have been verified using intensive fieldwork and accurate visual interpretations using different datasets (e.g., previous geological maps, remote sensing data, etc.). Upon field verification and petrographical investigations of this research outcomes, we strongly recommend the adopted approach for the geological community, as it opens the doors for various applications utilizing single-band second-order statistics. We expect that this research could significantly help the geological community by reviving several previous studies and being applicable for future research, besides offering a reasonable approach for minimizing the time and efforts required for detailed field studies by highlighting micro- and mesoscale structures. [ABSTRACT FROM AUTHOR]

Efficient medium access protocols are crucial in leveraging the advancement in the physical layer techniques that have been proposed for the next-generation wireless networks. However, collisions among data packets in random access protocols reduce channel efficiency and hence the throughput. In this paper, we propose a MAC protocol based on both the signal transmission diversity and Successive interference cancellation (SIC) to improve the throughput. In the proposed Medium Access Control (MAC) protocol, the receiver receives two sets of signals from the transmitters over two uncorrelated time slots. Inter slot SIC is first carried out at the receiver over two independent sets of received interfered signals. Signals are decoded by canceling those signals that have already been decoded in one set from the other set. Closed-form expressions for the packet decoding probability, system throughput, average packet transmission delay, and energy efficiency have been derived by considering order statistics of received signal power. Analytical results are validated against simulation. Maximum achievable throughput of the proposed scheme is shown to be four times that of slotted Aloha and twice that of a slotted Aloha with SIC. [ABSTRACT FROM AUTHOR]

• Fates of viruses and F-specific phages were monitored at two WWTPs in Japan. • NoV GII, AiV, PMMoV, and GI/GII-FRNAPHs were frequently detected in wastewater. • Membrane bioreactor reduced viruses more than activated sludge treatment. • Safe water reclamation in both processes requires additional virus reductions. The membrane bioreactor (MBR) process always offers better wastewater treatment than conventional activated sludge (CAS) treatment. However, the difference in their efficacy of virus reduction remains unknown. To investigate this, we monitored virus concentrations before and after MBR and CAS processes over 2 years. Concentrations of norovirus genotypes I and II (NoV GI and GII), aichivirus (AiV), F-specific RNA phage genotypes I, II, and III (GI-, GII-, and GIII-FRNAPHs), and pepper mild mottle virus (PMMoV) were measured by a quantitative polymerase chain reaction (qPCR) method at two municipal wastewater treatment plants (WWTPs A and B) in Japan. Virus concentration datasets containing left-censored data were estimated by using both maximum likelihood estimation (MLE) and robust regression on order statistics (rROS) approaches. PMMoV was the most prevalent at both WWTPs, with median concentrations of 7.5 to 8.8 log 10 copies/L before treatment. Log 10 removal values (LRVs) of all viruses based on means and standard deviations of concentrations before and after treatment were consistently higher following MBR than following CAS. We used NoV GII as a model pathogen in a quantitative microbial risk assessment of the treated water, and we estimated the additional reductions required following MBR and CAS processes to meet the guideline of 10−6 DALYs pppy for safe wastewater reuse. [Display omitted] [ABSTRACT FROM AUTHOR]

Land use and land cover classification from a remote sensing image is a long standing research problem. It ranges from simple classifications like mapping water bodies to complex classifications like crop and forest strands. Crop image classification is complex because of various stages of growth of the same crop, same spectral values for various crops, an other multitude of problems. Crop image classification is very essential for agriculture monitoring, crop yield production, global food security, etc. A new unsupervised algorithm, Spherical Contact Distribution Classification Algorithm (SCDCA) is proposed in this paper which uses mathematical morphology, spherical contact distributions, and first order statistics. Later SCDCA is compared with linear contact distribution classification algorithm (LCDCA). Quantitative analyses prove the efficiency of the algorithm and present that the complexity of SCDCA is very much less when compared to that of LCDCA. [ABSTRACT FROM AUTHOR]

Denote S n (p) = k n − 1 ∑ i = 1 k n log (X n + 1 − i , n ∕ X n − k n , n) p , where p > 0 , k n ≤ n is a sequence of integers such that k n → ∞ and k n ∕ n → 0 , and X 1 , n ≤ ⋯ ≤ X n , n are the order statistics of iid random variables X 1 , ... , X n with regularly varying upper tail of index 1 ∕ γ. The estimator γ ̂ (n) = (S n (p) ∕ Γ (p + 1)) 1 ∕ p is an extension of the Hill estimator. We investigate the asymptotic properties of S n (p) and γ ̂ (n) both for fixed p > 0 and for p = p n → ∞. We prove consistency for γ ̂ (n) and limit theorem for γ ̂ (n) − γ under appropriate assumptions. We obtain both Gaussian and non-Gaussian (stable) limit depending on the growth rate of the power sequence p n. Applied to real data we find that for larger p the estimator is less sensitive to the change in k n than the Hill estimator. [ABSTRACT FROM AUTHOR]

Knowledge of the turbulence in the wakes of wind turbines is important to improve the operation and lifetime of downstream turbines exposed to these wakes. As the investigation of a rotating turbine can be challenging, both experimentally and numerically, static actuator discs can substitute rotating turbines. To investigate the similarity of the turbulence in the wakes and how turbulent inflow alters the result, an experimental wind tunnel study is presented that compares hot-wire measurements in the central wakes of an actuator disc and a model wind turbine (diameter-based Reynolds number Re ≈ 300000). Two turbulent inflows are investigated, an intermittent, i.e. gusty, atmospheric-like inflow generated by an active grid, and a non-intermittent inflow generated by a regular grid. The data is investigated using one-point and higher-order two-point statistics. We find that the wakes of both models have similar properties for all investigated quantities within the far wake. The turbulence in the central wake is independent of the inflow conditions, and both models filter larger scale intermittency in the far wake with regard to the inflow. Also, we find that the turbine generates its own kind of turbulence that dominates the ambient turbulence and has strong features of homogeneous, isotropic turbulence. [Display omitted] • The results indicate where an actuator disc is a good substitute for a turbine. • Findings indicate how to determine a safe spacing for turbines within a wind farm. • One-point statistical properties are alike in the central far wakes of both models. • Two-point statistical properties are alike in the central far wakes of both models. • Both models create a universal turbulent flow independent of the inflow conditions. [ABSTRACT FROM AUTHOR]

In this paper, the phase diagram of (VNbTaMoW) 0.5 C x (0 < x < 0.6) was first calculated. We establish the complete phase diagram in two spatial dimensions from analysis the statistics of order phase parameters. To obtain high-entropy carbides and validate the calculated results, a series of (VNbTaMoW)C were sintered at different temperature. The single-phase carbide of (VNbTaMoW)C were synthesized at 1800 °C, which is higher than the theoretically calculated temperature in phase diagram. The reason is that the estimated temperature was calculated by mixing entropy for experimentally reported ideal solid solution models that simplified the temperature dependent parameters. Phase composition of (VNbTaMoW)C sintered at/below 1800 °C agrees well with the calculation results in phase diagram. Furthermore, we have systematically investigated the single-phase formation ability of four-metal and three-metal composition based on five elements (V, Nb, Ta, Mo, W). A total of 5 types of four-metal compositions and 10 types of three-metal compositions can be generated. However, for each composition, only 4 kinds four-metal and 6 kinds three-metal compositions can form single phase carbides. Those as-prepared single-phase carbides exhibit enhanced mechanical properties compared to the calculated values according to the rule of mixtures from constituent monocarbides. [ABSTRACT FROM AUTHOR]

• Studies on fade duration and fade slope are done at 20.2 GHz over tropical region. • Fade duration statistics are compared with ITU-R and Cheffena-Amaya Model. • Exponent of power law model of fade duration at Ka band is investigated. • A new model on parameters of fade duration statistics is proposed. Some second order rain attenuation statistics such as fade duration and fade slope are investigated on the basis of experimental measurements of received signals using the GSAT-14 satellite beacon signal at 20.2 GHz for three years (2014–2016) over the tropical location Ahmedabad (23.02 0E, 72.510N), India with an Elevation angle of 630. Existing models of fade duration are compared with experimental data in this study and exponent of power law model of fade duration at Ka band is further explored. A new model for fade duration for Ka band for tropical locations is proposed where the constant of exponent of attenuation in the power law is found to be 0.143 instead of 0.055 used in ITU-R. Other relevant parameters for implementation of fade mitigation technique to prevent the link outage like cumulative distribution of signal fade rate, maximum and minimum fade rise and fade fall are also studied. Fade slope asymmetry over tropical region is also investigated. Keeping in view of exploiting the commercial launch of Ka band in Indian region there is an urgent need for validation of the existing models of fade slope (specially looking into fade symmetry) and fade duration. It will help the SATCOM (Satellite Communication) link designer to improve closed loop fade mitigation technique to minimize the possible link failure/link outage over the tropical region. [ABSTRACT FROM AUTHOR]

• The uncertainty theory-based reliability evaluation is proposed to analyze the limited and censored time-to-failure data. • Two types of life testing censoring and two types of time-to-failure data are considered. • The proposed method can quantify epistemic uncertainties and provide stable and accurate mean time to failure results. Limited and censored time-to-failure data are common in practical life testing, which may lead to a lack of knowledge, i.e., epistemic uncertainties, and makes it difficult to evaluate product reliability accurately. Under this situation, the large-sample based probability theory is not appropriate anymore. To deal with such problem, in this paper, the uncertainty theory based reliability evaluation is proposed to cope with the limited and censored time-to-failure data, where two types of life testing censoring (type-I and type-II censoring) and two types of time-to-failure data (precise and interval data) are considered. Firstly, the lifetime model and reliability evaluations based on uncertainty theory are presented. Then, the uncertain statistics method is presented for each combination of censoring types and data types in life testing, which includes four steps: order statistics construction, belief degree construction according to the Laplace principle of indifference, parameter estimation based on the uncertain principle of least squares, and uncertain hypothesis testing. Two simulation studies and two practical cases with comprehensive and in-depth discussion are conducted to illustrate the practicability and effectiveness of the proposed method. The results show that the proposed method can quantify epistemic uncertainties properly and provide more stable and accurate mean time to failure results under limited and censored time-to-failure data compared with probability and Bayesian probability methods. [ABSTRACT FROM AUTHOR]

Multi-view learning deals with data that is described through multiple representations, or views. While various real-world data can be represented by three or more views, several existing multi-view classification methods can only handle two views. Previously proposed methods usually solve this issue by optimizing pairwise combinations of views. Although this can numerically deal with the issue of multiple views, it ignores the higher order correlations which can only be examined by exploring all views simultaneously. In this work new multi-view classification approaches are introduced which aim to include higher order statistics when three or more views are available. The proposed model is an extension to the recently proposed Restricted Kernel Machine classifier model and assumes shared hidden features for all views, as well as a newly introduced model tensor. Experimental results show an improvement with respect to state-of-the art pairwise multi-view learning methods, both in terms of classification accuracy and runtime. • A novel multi-view classification model, called ϱ TMV-RKM, is proposed. • Tensor learning is incorporated to account for higher order correlations. • Experimental results show the merit of using a weight tensor. • Comparisons with other state-of-the-art methods show improvement. • Multiple approaches to handle a large-scale dataset are proposed. [ABSTRACT FROM AUTHOR]

A robust myoelectric control system (MCS) is essential for the design of electromyography (EMG) based human–machine interface (HMI) designs such as prosthetics, exoskeleton, wheelchair and humanoid robots. The functionality of the current pattern recognition (PR) technique in MCS is limited by factors such as variation in the user's limb position. To overcome the effect of this dynamic variation, an invariant higher order statistics – frequency domain feature set (HOS-FD) is proposed in this paper. EMG data from eight hand movements in five limb positions are considered. When trained with three limb position data, the HOS-FD with three hidden layers deep neural network (DNN) achieved a significantly high classification accuracy of 97.84%±0.22 compared to other classifiers viz., single layer artificial neural network (ANN), linear discriminant analysis (LDA), support vector machine (SVM), k nearest neighbor (kNN), decision tree (DT) and Naïve Bayes (NB) classifiers with accuracies of 94.11%±1.63, 95.02%±1.89, 94.63%±2.33, 90.05 ± 4.11, 86.66 ± 4.72 and 78.78%±5.02 respectively. Further, when trained with data from all five limb positions, the proposed feature set with DNN had an accuracy of 99.16%±0.14. The statistical significance of the high classification accuracy obtained using the proposed feature set is also proven using multiple analysis of variance tests (p < 0.001). These results indicate that the proposed method is a promising technique for HMI. [ABSTRACT FROM AUTHOR]

• Generalization of binary sequences to multivariate improves data representation. • The weighted-probability of pattern occurrence is constructed based on complexity. • The MDS based on the novel dissimilarity measure is constructed. • The MDS creates better clustering results when compared with other methods. • The method has an acceptable running time and possesses robustness to noisy data. In this paper, a new dissimilarity measure for more accurate feature extraction and clustering is put forward. The method is proposed from the perspective of the weighted-probability distribution of dispersion patterns and their rank order statistics, where an effective quantization procedure is provided and the loss of information can be reduced. The proposed dissimilarity is applied in the multidimensional scaling (MDS) method to investigate simulated and reality-based signals. The comparative experiment shows that the clustering results of the proposed technique are clearer and more appropriate. State-of-the-art techniques and conventional methods are both included in the comparative experiments. In particular, for the heartbeat signals, it is discovered that the distribution of the weighted-probabilities of the dispersion patterns can discriminate subjects with different physiological conditions and exhibit visible changes of the subject's dynamical features when aging and disease attacks are taken place, which can be regarded as a microscopic insight of the dynamical mechanisms. [ABSTRACT FROM AUTHOR]

• The multi-scale defects activation of Gd-Co-Al series metallic glasses (MGs) was systematically investigated based on their nanoindentation behaviors. • The Gd 18.33 Tb 18.33 Dy 18.34 Co 17.5 Al 27.5 MG with high configurational entropy demonstrates the highest hardness and elastic modulus. • Flow defects derived from the Maxwell-Voigt model associated with long relaxation time are mainly responsible for the large deformability. • The statistics of crystal-like order indicate the activation of shear transformation zones in high-entropy MGs requires overcoming a high energy barrier. Defect activation is of fundamental importance for plastic deformation. However, its mechanism is not yet fully disclosed in metallic glasses (MGs), especially for those with high configurational entropy that demonstrate outstanding physical or chemical properties, more particularly, mechanical characteristics. In this work, multi-scale defects activation of three rare-earth-based MGs with low-, medium-, and high-entropy (LE, ME, and HE) were systematically investigated by tracing their room-temperature nanoindentation behaviors. Among the three alloys, the Gd 18.33 Tb 18.33 Dy 18.34 Co 17.5 Al 27.5 HE MG exhibits the highest hardness and elastic modulus. Unlike the poor deformability of the LE MG, the pronounced nanoindentation displacements under a constant load were observed in HE and ME MGs, especially at high loading rates. To reveal the plastic origin of HE MGs, a generalized physical model (so-called Maxwell-Voigt) was utilized to describe the nanoindentation deformation on the mesoscale. The characteristic relaxation spectra show that the activated defects with longer relaxation time are mainly responsible for the large nanoindentation displacement at high strain rates. Based on the cooperative shearing model, the shear transformation zone (STZ) volume is determined to be around 0.45–2.98 nm3. The statistics of the short-range order indicate that the HE MG possesses a highly ordered configuration, which results in the difficulty of STZ activation for requiring overcoming a high energy barrier, and a high elastic modulus of the HE MG. Our work might provide insight into the underlying plastic deformation mechanism of HE MGs from the characteristics of activated defects as well as their entropy effect. Multi-scale defects activation of three rare-earth-based MGs with low-, medium-, and high-entropy (LE, ME, and HE) were systematically investigated by tracing their room-temperature nanoindentation behaviors. [Display omitted] [ABSTRACT FROM AUTHOR]

Working in the setting of i.i.d. last-passage percolation on R D with no assumptions on the underlying edge-weight distribution, we arrive at the notion of grid entropy — a Subadditive Ergodic Theorem limit of the entropies of paths with empirical measures weakly converging to a given target, or equivalently a deterministic critical exponent of canonical order statistics associated with the Levy-Prokhorov metric. This provides a fresh approach to an entropy first developed by Rassoul-Agha and Seppäläinen as a large deviation rate function of empirical measures along paths. In their 2014 paper, variational formulas are developed for the point-to-point/point-to-level Gibbs Free Energies as the convex conjugates of this entropy. We rework these formulas in our new framework and explicitly link our descriptions of grid entropy to theirs. We also improve on a known bound for this entropy by introducing a relative entropy term in the inequality. Furthermore, we show that the set of measures with finite grid entropy coincides with the deterministic set of limit points of empirical measures studied in a 2020 paper by Bates. We partially answer a directed polymer version of a question of Hoffman which was previously tackled in the zero temperature case by Bates. [ABSTRACT FROM AUTHOR]

The nonlinear, time correlation, and non-Gaussian features in data present significant challenges for effective fault detection. While the Gate Recurrent Unit (GRU) network is renowned for its capacity to manage time correlation, it falls short in capturing non-Gaussian features in process data, which can likely lead to suboptimal monitoring results. To address this limitation, the Enhancement Gate Recurrent Unit (ENGRU) is developed to perfect the fault detection accuracy of the network. Specifically, The ENGRU effectively extracts high order statistics information by employing the overcompleted independent component analysis method, thereby augmenting its ability to capture non-Gaussian properties. The extracted features information are then entered into the ENGRU model further to uncover additional hidden features beyond what the GRU can achieve. The ENGRU network, which is built upon the extracted characteristic information, even farther enhances the accuracy of the fault detection. The merits of the proposed model are demonstrated by comparing it with excellent fault detection algorithms on a benchmark platform. [ABSTRACT FROM AUTHOR]

Recent works have derived and proven the large-population mean-field limit for several classes of particle-based stochastic reaction–diffusion (PBSRD) models. These limits correspond to systems of partial integral–differential equations (PIDEs) that generalize standard mass-action reaction–diffusion PDE models. In this work we derive and prove the next order fluctuation corrections to such limits, which we show satisfy systems of stochastic PIDEs with Gaussian noise. Numerical examples are presented to illustrate how including the fluctuation corrections can enable the accurate estimation of higher order statistics of the underlying PBSRD model. [ABSTRACT FROM AUTHOR]

A signal can be represented by sparse representation with fewer coefficients. Due to this ability, sparse representation is used in research fields such as signal compression, noise elimination, and classification. In this study, sparse coefficients of the signals were obtained by using dictionary learning and sparse representation algorithms. The obtained coefficients were used in the weight update process of three different classifiers, which were created by using AdaBoost, SVM, and LDA algorithms. So, Dictionary learning based AdaBoost classifiers were obtained. The proposed Dictionary Learning (DL) based AdaBoost classifiers classified the ECG (Electrocardiography) signals. Before classification, the feature selection process was applied to ECG signals and six different feature subsets were obtained by Discrete Wavelet Transform (DWT), First Order Statistics (FOS), T-test, Bhattacharyya, Entropy, and Wilcoxon test methods. The feature subsets were used as the new dataset. The classification process was done by the proposed method and satisfying results were obtained. The best classification accuracy was obtained as 99.75% by the proposed dictionary learning based method called as DL-AdaBoost-SVM on feature subsets obtained by DWT and Wilcoxon test methods. [ABSTRACT FROM AUTHOR]

We investigate a class of weighted voting games for which weights are randomly distributed over the standard probability simplex. We provide close-formed formulae for the expectation and density of the distribution of weight of the k -th largest player under the uniform distribution. We analyze the average voting power of the k -th largest player and its dependence on the quota, obtaining analytical and numerical results for small values of n and a general theorem about the functional form of the relation between the average Penrose–Banzhaf power index and the quota for the uniform measure on the simplex. We also analyze the power of a collectivity to act (Coleman efficiency index) of random weighted voting games, obtaining analytical upper bounds therefor. [ABSTRACT FROM AUTHOR]

The constrained ordered weighted averaging (OWA) aggregation problem arises when we aim to maximize or minimize a convex combination of order statistics under linear inequality constraints that act on the variables with respect to their original sources. The standalone approach to optimizing the OWA under constraints is to consider all permutations of the inputs, which becomes quickly infeasible when there are more than a few variables, however in certain cases we can take advantage of the relationships amongst the constraints and the corresponding solution structures. For example, we can consider a land-use allocation satisfaction problem with an auxiliary aim of balancing land-types, whereby the response curves for each species are non-decreasing with respect to the land-types. This results in comonotone constraints, which allow us to drastically reduce the complexity of the problem. In this paper, we show that if we have an arbitrary number of constraints that are comonotone (i.e., they share the same ordering permutation of the coefficients), then the optimal solution occurs for decreasing components of the solution. After investigating the form of the solution in some special cases and providing theoretical results that shed light on the form of the solution, we detail practical approaches to solving and give real-world examples. [ABSTRACT FROM AUTHOR]

• We demonstrate the limits- potential of single-channel speech enhancement algorithms regarding the inter-frequency frequency dependencies. • We derive linear multidimensional short-time spectral amplitude estimator and provide empirical evidence for assumptions forming this model. • We provide empirical evidence for assumptions forming basis of the proposed linear multidimensional short-time spectral amplitude estimator. • The performance of proposed approach is evaluated in terms of speech quality and intelligibility and is compared to relevant benchmark methods. • Results showed that the proposed method achieved improvement in speech quality and intelligibility for various noisy scenarios. In this paper, we investigate single-channel speech enhancement algorithms that operate in the short-time Fourier transform and take into account dependencies w.r.t. frequency. As a result of allowing for inter-frequency dependencies, the minimum mean square error optimal estimates of the short-time Fourier transform expansion coefficients are functions of complex-valued covariance matrices in general. The covariance matrices are not known a priori and have to be estimated from the observed data. This work is dedicated to analyzing how this affects the respective single-channel speech enhancement algorithms. We propose a statistical model that circumvents the need to estimate complex-valued second order statistics and derive a linear multidimensional short-time spectral amplitude estimator that is motivated by these assumptions. Further, we provide empirical evidence for the assumptions that form the basis of this model. We evaluate the potential of taking into account inter-frequency dependencies for single-channel speech enhancement and subsequently compare the estimator resulting from the proposed statistical model to relevant benchmark methods. The results indicate that estimators that consider inter-frequency dependencies are capable of pushing the limits of standard approaches in terms of joint speech quality and intelligibility improvement when the second order statistics are estimated from isolated speech data. The proposed linear multidimensional short-time spectral amplitude estimator preserves this trend in fully blind scenarios. [ABSTRACT FROM AUTHOR]

Appearance based loop closure detection plays an important role in visual simultaneous localization and mapping systems (vSLAM) by measuring similarity of the places and checking loops to reduce the accumulated error. Traditional loop closure methods execute place recognition by image retrieval with Bag-of-Word model, which forms an orderless representation of local feature descriptors. Convolutional neural networks (CNNs) based features have been investigated for place recognition, where the final descriptors usually are generated by first-order pooling, limiting the representation ability in challenging scenarios. To handle above issue, we introduce high-order statistics into place recognition by developing a novel adaptively normalized covariance pooling method for learning place representations in an end-to-end manner. The proposed method provides robust covariance matrix estimation of high-dimensional and small-size deep features by adaptive covariance normalization (AdaCN). Experimental results on place recognition in the urban environment and image retrieval tasks show that second-order representation is effective, especially for discriminating places with confusing objects, changes in viewpoint and illumination. Besides, the proposed adaptive normalization performs favorably against its counterparts based on Log-Euclidean Riemannian metric and Power-Euclidean metric, while our method is superior to the state-of-the-art place recognition approaches. [ABSTRACT FROM AUTHOR]

In this paper, we proposed a new extension of the Rayleigh distribution with two parameter called type I half logistic Rayleigh distribution. Several important mathematical and statistical properties of this new distribution are discussed. Simulation studies and real data applications are also considered for performance of the new distribution. [ABSTRACT FROM AUTHOR]

• A new change-point model for the r -largest order statistics is proposed. • The proposed model can be used for any application that involves an extreme time series with a structural break. • A Bayesian adaptation for the estimation of the optimal number of order statistics is proposed. • Simulations and real data sets shows the good performance of this new model. • The application in three examples verifies the performance of the new model is better than that the usual model. This study makes a new contribution to extreme value theory by proposing a change-point model of the distribution of the r -larger order statistics. In some situations, using only the maxima of grouped data results in a small sample size that may require a larger dataset. In this sense, using the joint distribution of the r -largest order statistics provides more information and, consequently, better estimators. We perform a comprehensive simulation to show the advantage of this method over other competitive models that approach the change-point model in extremes. Finally, the proposed model is fitted to river quota data (environmental data) and NASDAQ daily returns data (financial data) to demonstrate its potential for practical application. [ABSTRACT FROM AUTHOR]

We study the dynamics of erasing randomly chosen letters from words by introducing a certain class of discrete-time stochastic processes, general erased-word processes (GEWPs), and investigating three closely related topics: Representation, Martin boundary and filtration theory. We use de Finetti's theorem and the random exchangeable linear order to obtain a de Finetti-type representation of GEWPs involving induced order statistics. Our studies expose connections between exchangeability theory and certain poly-adic filtrations that can be found in other exchangeable random objects as well. We show that ergodic GEWPs generate backward filtrations of product-type and by that generalize a result by Laurent (2016). [ABSTRACT FROM AUTHOR]

• EMD-DWT two stage analysis of voice samples. • Temporal energy to select robust IMF. • Extraction of HOSs features from robust IMF. • Classification accuracy reached 100% for the classification of "Funktionelle Dysphonia vs. Rekurrensparse". The voice is a prominent tool allowing people to communicate and to change information in their daily activities. However, any slight alteration in the voice production system may affect the voice quality. Over the last years, researchers in biomedical engineering field worked to develop a robust automatic system that may help clinicians to perform a preventive diagnosis in order to detect the voice pathologies in an early stage. In this context, pathological voice detection and classification method based on EMD-DWT analysis and Higher Order Statistics (HOS) features, is proposed. Also DWT coefficients features are extracted and tested. To carry out our experiments a wide subset of voice signal from normal subjects and subjects which suffer from the five most frequent pathologies in the Saarbrücken Voice Database (SVD), is selected. In The first step, we applied the Empirical Mode Decomposition (EMD) to the voice signal. Afterwards, among the obtained candidates of Intrinsic Mode Functions (IMFs), we choose the robust one based on temporal energy criterion. In the second step, the selected IMF was decomposed via the Discrete Wavelet Transform (DWT). As a result, two features vector includes six HOSs parameters, and a features vector includes six DWT features were formed from both approximation and detail coefficients. In order to classify the obtained data a support vector machine (SVM) is employed. After having trained the proposed system using the SVD database, the system was evaluated using voice signals of volunteer's subjects from the Neurological department of RABTA Hospital of Tunis. The proposed method gives promising results in pathological voices detection. The accuracies reached 99.26% using HOS features and 93.1% using DWT features for SVD database. In the classification, an accuracy of 100% was reached for "Funktionelle Dysphonia vs. Rekrrensparese" based on HOS features. Nevertheless, using DWT features the accuracy achieved was 90.32% for "Hyperfunktionelle Dysphonia vs. Rekurrensparse". Furthermore, in the validation the accuracies reached were 94.82%, 91.37% for HOS and DWT features, respectively. In the classification the highest accuracies reached were for classifying "Parkinson versus Paralysis" 94.44% and 88.87% based on HOS and DWT features, respectively. HOS features show promising results in the automatic voice pathology detection and classification compared to DWT features. Thus, it can reliably be used as noninvasive tool to assist clinical evaluation for pathological voices identification. [ABSTRACT FROM AUTHOR]

Deep learning technology has been gradually applied to solve a variety of fault diagnosis problems because of its outstanding feature learning and nonlinear classification abilities. However, few deep learning network models can be applied to both variable speed conditions and unbalanced samples scenarios in fault diagnosis, especially in extreme cases where fault samples are missing. And most of the fault diagnosis models do not have the ability to update automatically as the collected fault data increases. To deal with the above problems, a deep learning model named renewable fusion fault diagnosis network (RFFDN) is proposed. The network has three main parts: improved feature classification network; second order statistics fusion network and unbalanced feature comparison network. Moreover, these three networks are simultaneously organized on a two-branch convolutional neural network (CNN) architecture with fused data input, so as to facilitate the network to learn the depth nonlinear domain invariant features. Finally, the RFFDN model and other mainstream fault diagnosis models are tested on two different datasets. The results show that the RFFDN model not simply achieves high diagnostic accuracy in diagnosis results, but also extracts the domain invariant features at variable speed conditions under unbalanced samples, and accurately classifies the new faults. These prove that the model can not only be applied to a variety of operating modes, but can be updated as more data are collected as well, which is of great significance to the field of fault diagnosis. [ABSTRACT FROM AUTHOR]

Causal inference aims to estimate the treatment effect of an intervention on the target outcome variable and has received great attention across fields ranging from economics and statistics to machine learning. Observational causal inference is challenging because the pre-treatment variables may influence both the treatment and the outcome, resulting in confounding bias. The classic inverse propensity weighting (IPW) estimator is theoretically able to eliminate the confounding bias. However, in observational studies, the propensity scores used in the IPW estimator must be estimated from finite observational data and may be subject to extreme values, leading to the problem of highly variable importance weights, which consequently makes the estimated causal effect unstable or even misleading. In this paper, by reframing the IPW estimator in the importance sampling framework, we propose a Pareto-smoothing method to tackle this problem. The generalized Pareto distribution (GPD) from extreme value theory is used to fit the upper tail of the estimated importance weights and to replace them using the order statistics of the fitted GPD. To validate the performance of the new method, we conducted extensive experiments on simulated and semi-simulated datasets. Compared with two existing methods for importance weight stabilization, i.e., weight truncation and self-normalization, the proposed method generally achieves better performance in settings with a small sample size and high-dimensional covariates. Its application on a real-world heath dataset indicates its utility in estimating causal effects for program evaluation. [ABSTRACT FROM AUTHOR]

It is vital for fault detection technology to extract features of industrial process data effectively. Local kernel principal component analysis (LKPCA) has proved its good performance in preserving global and local structural characteristics. However, it ignored useful high-order statistics of data, so multi-block statistics local kernel principal component analysis (MSLKPCA) algorithm integrating statistics pattern analysis (SPA) into LKPCA is proposed. The correlation coefficient matrix is first calculated and K-means clustering is adopted to divide the original variables into several blocks. Then the weighted SPA, which gives different weights to different samples in each window according to their distributions, is adopted to build statistic spaces containing both low-order and high-order statistics. After that, LKCPA is performed in each statistic space to realize feature extraction. To reduce the noise effect amplified by SPA, PCA is adopted in the residual space to remove noise. Bayesian strategy is used to fuse the results of each block and two monitoring statistics E T and E R are proposed to monitor the feature space and the residual space respectively. The Tennessee–Eastman (TE) process simulation shows the effectiveness and superiority of the proposed algorithm for process monitoring. [ABSTRACT FROM AUTHOR]

Arrhythmia is one of the major cause of deaths across the globe. Almost 17.9 million deaths are caused due to cardiovascular diseases. In order to reduce this much mortality rate, the cardiovascular disease should be properly identified and the proper treatment for the same should be immediately provided to the patients. In this study, a new ensemble based support vector machine (SVM) classifier was proposed to classify heartbeat into four classes from MIT-BIH arrhythmia database. The results were compared with other classifiers that are SVM, Random Forest (RF), K-Nearest Neighbours (KNN), and Long Short Term Memory network. The four features were extracted from the ECG signals that were used by the classifiers are Wavelets, high order statistics, R-R intervals and morphological features. An ensemble of SVMs obtained the best result with an overall accuracy of 94.4%. [ABSTRACT FROM AUTHOR]