Filtering Scheme Research Articles

Exponential [formula omitted] filtering for complex-valued uncertain discrete-time neural networks with time-varying delays

The purpose of this paper is to design a compatible filter for a class of classical discrete-time neural networks (DTNNs) having uncertain complex-valued weighting parameters and time-varying delayed responses subject to the H∞ performance measure. For this notion, the complex-valued filter scheme is designed for the proposed uncertain DTNNs with regard to the available output measurements. At first, some novel complex-valued weighted summation inequalities (WSIs) are put forth to establish a more precise linearized lower bound for the quadratic summing terms resulting from the forward difference of the assigned Lyapunov–Krasovskii functional (LKF). In what follows, an attempt has been made to propose the linear matrix inequality (LMI) based sufficient conditions for designing the robust H∞ filter from the filtering error system attains exponential stability with the appropriate filtering gain matrices. Eventually, the theoretical conclusion is substantiated through a numerical example and the simulation outcomes reveal the applicability and efficiency of the proposed filter scheme.

Зависимость стохастического резонанса в нелинейной системе от характеристик каждого слагаемого аддитивной смеси

A nonlinear system was investigated using a modeling approach, constructed based on the scheme of a digital filter using the Lyapunov characteristic function. The phenomenon of stochastic resonance was observed in the nonlinear system when there was an additive mixture of a quasi-deterministic signal and noise at the system input. Stochastic resonance improves the signal-to-noise ratio at the system output by more than 15400 times compared to its value at the system input. The recorded number may vary and represents the main characteristic of the resonance. The dependence of the stochastic resonance characteristic on the variations of the probabilistic characteristics of the quasi-deterministic signal and noise was studied. An analysis of experimental data describing the values of the characteristic and the peculiarities of the stochastic resonance effect in the nonlinear system was performed. Additionally, a connection between the value of the resonance characteristic and the total energy of the periodic signal was discovered for the first time.

Зависимость стохастического резонанса в нелинейной системе от характеристик каждого слагаемого аддитивной смеси

A nonlinear system was investigated using a modeling approach, constructed based on the scheme of a digital filter using the Lyapunov characteristic function. The phenomenon of stochastic resonance was observed in the nonlinear system when there was an additive mixture of a quasi-deterministic signal and noise at the system input. Stochastic resonance improves the signal-to-noise ratio at the system output by more than 15400 times compared to its value at the system input. The recorded number may vary and represents the main characteristic of the resonance. The dependence of the stochastic resonance characteristic on the variations of the probabilistic characteristics of the quasi-deterministic signal and noise was studied. An analysis of experimental data describing the values of the characteristic and the peculiarities of the stochastic resonance effect in the nonlinear system was performed. Additionally, a connection between the value of the resonance characteristic and the total energy of the periodic signal was discovered for the first time.

Design and performance evaluation of a novel metamaterial broadband THz filter for 6G applications

Terahertz (THz) radiation, which has applications in the imaging of objects, non-destructive testing, satellite communication, medical diagnostics, and biosensing, has generated a great deal of attention due to its remarkable properties. This paper proposes a novel broadband filter for THz applications. The main idea is to overcome the insertion loss and bandwidth issues by modeling a frequency-domain finite difference method and guided-mode resonance (GMR). The optimal design scheme of the wideband pass filter based on the circular resonant ring is discussed by comparing the transmission parameters under various parameters. This scheme overcomes the restriction of the narrow passband bandwidth of the prior THz filters and achieves approximately 3 dB bandwidth of 0.54 THz. The proposed THz filter paper also has the advantages of a straightforward structure, low processing costs, and ease of conformal with other structures, and it can be used for stealth fighters, new communication technology, and precise instruments. In addition, when compared to existing models, the suggested filter offers higher 3 dB BW operation, increased transmittance, low insertion loss, and stable performance at various oblique angles.

Automatic Image-Based SBFE-BESO Approach for Topology Structural Optimization

This paper presents automatic image-based topology optimization with the highest mechanical stiffness. A so-called bi-evolutionary structural optimization is encoded within the scaled boundary finite element (SBFE) framework, incorporating digital image-based quadtree (2D) and octree (3D) decomposing mesh constructions. The information on both geometry and material distribution is transparently transferred using the standard tessellation language (STL) format of digital images, which can be directly modelled for stress analyses. The polytope SBFE method allows for a computationally advantageous model construction with hanging nodes. A convolution filter scheme is applied for colour intensity adjustment between solid and void regions, leading to multi-levels in the quadtree/octree hierarchy. Combined with the convolution filter method, the matrix precomputation technique, within the polytope SBFE decomposition framework enables the automatic adaptive (digital image-based) model construction from the limited number of master cells with balanced quadtree/octree meshes. Hence, the highly efficient analysis-ready design scheme for 2D and 3D structures is carried out. The proposed approach significantly reduces the number of degrees of freedom required, giving thus the computationally efficient efforts that are seen to enhance the high-resolution optimal topology of the practical-scale structures.

A semi-automated coding scheme for occupational injury data: An approach using Bayesian decision support system

IntroductionOver the past few years, classic Machine Learning approaches such as Multinomial Naïve Bayes, Support Vector Machine as well as regularized Logistic Regression have been adapted to autocode injury narratives collected by the Bureau of Labour Statistics to reduce the manual effort needed to assign codes to these narratives. However, the effectiveness of these algorithms is yet to be explored on severe injury reports collected by the Occupational Safety and Health Administration (OSHA). This study aims to explore the performances of two Bayesian models for autocoding these reports, segregate narratives that require manual reviewing, and analyse the usefulness of presenting top k predictions for such reviews to human coders. MethodThe severe injury reports collected by OSHA from January 2015 to February 2021 were used in this study. Firstly, Unigram (UNB) and Bigram (BNB) Naïve Bayes models are used to classify the injury narratives, and their performance is analyzed. Furthermore, two filtering strategies are used a) Instances where the UNB and BNB models agree are autocoded b) Only cases where the two models agree and whose prediction probability is above a minimum threshold are autocoded. The remaining cases are filtered out to be reviewed and coded by manual coders. The sensitivity of top k predictions for the UNB, BNB, and UNB-BNB models are also compared and analyzed to aid human coders in assigning codes to the narratives that are filtered out. ResultsFor fully autocoded data, the sensitivity of the UNB model is 75.21%, and that of the BNB model is 75.17%. The filtering approach has an overall sensitivity of 88.17%, flagging 31% of the injury narratives for manual review. The UNB model performs slightly better than the BNB model, and the accuracy increases as cases where the two models agree are considered and a prediction probability threshold is set. For the top 5 predictions, a maximum F1-score of 55% is achieved by the UNB-BNB model.

Analysis of a new adaptive time filter algorithm for the unsteady Stokes/Darcy model

In this article, we propose a new adaptive time filter algorithm for the unsteady Stokes/Darcy model. Firstly, we present a first order θ-scheme with variable time step which is one parameter family of Linear Multi-step method. Furthermore, using time filter scheme, the convergence order is increased to the second order without almost increasing the extra computation. Moreover, we construct decoupled and coupled adaptive algorithms to improve the computational efficiency. Theoretically, we analyze stability and the second-order accuracy of Linear Multi-step method plus time filter with non-increasing variable time step, respectively. Finally, numerical experiments are given to verify theoretical results, including effectiveness, convergence and efficiency.

A GRNN informed ECMS-ALPF operational management strategy for reducing fuel consumption of diesel/battery/supercapacitor hybrid vehicles

The fuel consumption and pollution emission of ships have become the main focuses in ocean environment management. This paper proposes a new marine vehicular hybrid propulsion system (VHPS), which is composed of a diesel generator and a hybrid energy storage system consisted of batteries and supercapacitors. However, the distinct power and energy characteristics of the three power sources result in complex operational management and a negative impact on energy conservation and emission reduction. Therefore, a general regression neural network informed equivalent consumption minimization strategy and adaptive low pass filter (GRNN informed ECMS-ALPF) operational management strategy (OMS) is proposed to optimize the fuel consumption and energy loss. Finally, a hardware in the loop experimental platform is utilized to verify the performance of proposed OMS. From experimental results, the GRNN informed ECMS-ALPF strategy has accurate load predictive capability, the predictive average root mean square error value is only 2.1633 kW, whose fuel consumption is 14.9% lower compared with a conventional rule-based strategy. And the energy loss of the given hybrid energy storage system can reduce 0.0178 kWh (about 14.55%) compared with a conventional low pass filter strategy. The proposed VHPS and OMS offer insights for the study of ocean and coastal management.

An improved partial differential equation filter scheme for topology optimization of additively manufactured coated structure

In this paper, an improved partial differential equation (PDE) filter is presented for density-based topology optimization of coated structures with both regular and irregular design domains. The proposed filter efficiently addresses the boundary effect problems that causes identification failure of the material interface for modeling coating layer during topology optimization by replacing the original Dirichlet boundary conditions with non-homogeneous Neumann boundary conditions. One attractive feature of the proposed approach is that it can serve as an ideal alternative to the common domain extension approach, particularly for designing geometrically complex engineering structures with irregular design domain meshed by unstructured mesh. Based on the improved PDE filter, a new topology optimization design framework for coated structures with complete coating layer and controllable coating thickness is achieved. Examples of designing regular and irregular two-dimensional and three-dimensional coated structures are provided to demonstrate the effectiveness of the proposed approach.

Wind and Actuator Fault Estimation for a Quadrotor UAV Based on Two-Stage Particle Filter

This paper investigates the problems of wind and actuator fault estimation for a quadrotor unmanned aerial vehicle (UAV). To effectively assess the safety and reliability of a quadrotor UAV in the presence of unknown wind disturbances, a two-stage particle filter (TSPF) scheme is proposed to obtain the simultaneous estimation of winds and actuator faults that may degrade the performance of the vehicle. In this scheme, the first-stage particle filter is used to estimate the states of the quadrotor UAV, and the second-stage particle filter is designed to produce estimates of unknown parameters, including the wind disturbances and actuator faults. To mitigate the degeneracy and impoverishment issues, the second-stage particle filter admits a parallel implementation of increased particle samplings for the wind and actuator fault estimation. Finally, simulation results are presented to demonstrate the effectiveness of the proposed scheme.

CACO: A core-attachment method with cross-species functional ortholog information to detect human protein complexes.

Protein complexes play an essential role in living cells. Detecting protein complexes is crucial to understand protein functions and treat complex diseases. Due to high time and resource consumption of experiment approaches, many computational approaches have been proposed to detect protein complexes. However, most of them are only based on protein-protein interaction (PPI) networks, which heavily suffer from the noise in PPI networks. Therefore, we propose a novel core-attachment method, named CACO, to detect human protein complexes, by integrating the functional information from other species via protein ortholog relations. First, CACO constructs a cross-species ortholog relation matrix and transfers GO terms from other species as a reference to evaluate the confidence of PPIs. Then, a PPI filter strategy is adopted to clean the PPI network and thus a weighted clean PPI network is constructed. Finally, a new effective core-attachment algorithm is proposed to detect protein complexes from the weighted PPI network. Compared to other thirteen state-of-the-art methods, CACO outperforms all of them in terms of F-measure and Composite Score, showing that integrating ortholog information and the proposed core-attachment algorithm are effective in detecting protein complexes. CACO is available at https://csuligroup.com/CACO/ and https://github.com/CSUBioGroup/CACO.

SFyNCS detects oncogenic fusions involving non-coding sequences in cancer.

Fusion genes are well-known cancer drivers. However, most known oncogenic fusions are protein-coding, and very few involve non-coding sequences due to lack of suitable detection tools. We develop SFyNCS to detect fusions of both protein-coding genes and non-coding sequences from transcriptomic sequencing data. The main advantage of this study is that we use somatic structural variations detected from genomic data to validate fusions detected from transcriptomic data. This allows us to comprehensively evaluate various fusion detection and filtering strategies and parameters. We show that SFyNCS has superior sensitivity and specificity over existing algorithms through extensive benchmarking in cancer cell lines and patient samples. We then apply SFyNCS to 9565 tumor samples across 33 tumor types in The Cancer Genome Atlas cohort and detect a total of 165,139 fusions. Among them, 72% of the fusions involve non-coding sequences. We find a long non-coding RNA to recurrently fuse with various oncogenes in 3% of prostate cancers. In addition, we discover fusions involving two non-coding RNAs in 32% of dedifferentiated liposarcomas and experimentally validated the oncogenic functions in mouse model.

Study of Humidity During Sand Dewatering Using a Cone-Shaped Installation

The issue of moisture content of sand, which is used in construction as an aggregate for asphalt concrete mixtures, in the production of silicate building materials, roofing roll materials, and various types of glass, is considered. The results of an experimental study of the process of dewatering construction sand samples in conical warehouses under the conditions of using vacuum systems, depending on the time and placement scheme of needle filters, are presented.

Hybrid‐triggered based reliable filter design for interval type‐2 fuzzy system under deception attacks and missing measurements

AbstractThis study inspects the problem of fault‐tolerant control based filter design for interval type‐2 (IT‐2) fuzzy systems with deception attacks and missing measurements, under hybrid‐triggered mechanism. At first, hybrid‐triggered mechanism is commenced to preserve the network bandwidth and also to maintain the competent system performance. Next, the existence of deception attack in the filter infuses unwanted data in the measurement output in order to limit the transmitted signal in the communication networks. The behaviors of triggering mechanism and attack signals are governed by stochastic Bernoulli distribution. Moreover, the missing measurements phenomenon is also characterized by stochastic random variables, which observes the data loss between system and the filter during the transference of information. The sensor failure, deception attack and missing measurement are the key factors that will destruct the network security completely. Further, by endowing Lyapunov stability theory and linear matrix inequality (LMI), a set of adequate conditions is fabricated to ensure the mean‐square asymptotic filtering of the proposed IT‐2 fuzzy system. Subsequently, the filter gain matrices are exhibited with desired LMI conditions. Eventually, the capability and usefulness of the proposed fault‐tolerant filter scheme is substantiated via two numerical examples which includes a tunnel diode application.

A genetic particle filter scheme for univariate snow cover assimilation into Noah-MP model across snow climates

Abstract. Accurate snowpack simulations are critical for regional hydrological predictions, snow avalanche prevention, water resource management, and agricultural production, particularly during the snow ablation period. Data assimilation methodologies are increasingly being applied for operational purposes to reduce the uncertainty in snowpack simulations and to enhance their predictive capabilities. This study aims to investigate the feasibility of using a genetic particle filter (GPF) as a snow data assimilation scheme designed to assimilate ground-based snow depth (SD) measurements across different snow climates. We employed the default parameterization scheme combination within the Noah-MP (with multi-parameterization) model as the model operator in the snow data assimilation system to evolve snow variables and evaluated the assimilation performance of the GPF using observational data from sites with different snow climates. We also explored the impact of measurement frequency and particle number on the filter updating of the snowpack state at different sites and the results of generic resampling methods compared to the genetic algorithm used in the resampling process. Our results demonstrate that a GPF can be used as a snow data assimilation scheme to assimilate ground-based measurements and obtain satisfactory assimilation performance across different snow climates. We found that particle number is not crucial for the filter's performance, and 100 particles are sufficient to represent the high dimensionality of the point-scale system. The frequency of measurements can significantly affect the filter-updating performance, and dense ground-based snow observational data always dominate the accuracy of assimilation results. Compared to generic resampling methods, the genetic algorithm used to resample particles can significantly enhance the diversity of particles and prevent particle degeneration and impoverishment. Finally, we concluded that the GPF is a suitable candidate approach for snow data assimilation and is appropriate for different snow climates.

A Tiny Accelerator for Mixed-Bit Sparse CNN Based on Efficient Fetch Method of SIMO SPad

Convolution neural networks (CNNs) have been implemented with custom hardware on edge devices since its algorithms were successful in many artificial intelligence applications. Although lots of unstructured pruning and mix-bit quantization algorithms have been proposed to successfully compress CNNs, there are few hardware accelerators which can support both sparse and mix-bit CNNs. Besides, sparse matrix computation consumes lots of hardware resources such as registers or BRAM to fetch the needed input activations into processing element (PE). This paper presents a tiny accelerator for mixed-bit sparse CNNs featuring a novel scheme of single vector-based compressed sparse filter (CSF) method and single input multiple output scratch pad (SIMO SPad) to effectively compress weight and fetch the needed input activation. SIMO SPad is shared by multiple PEs, which saves 13.34% CLB LUTs, 46.24% CLB Registers. Furthermore, the accelerator supports mixed-bit sparse computation to obtain better accuracy and performance. When tested on VGG16, compared with 8-bit non-sparsity baselines, the performance of mixed-bit sparsity on Cifar10 and ImageNet improved by 4.85× and 3.33×, respectively, with small accuracy decrease degradation. Compared to state-of-the-art accelerators, the accelerator achieves 1.40× to 2.98× greater DSP efficiency, and offers 1.91× greater energy efficiency.

Feature Alignment in Anchor-Free Object Detection

Most anchor-free methods perform object detection using dense recommendation, which assumes that one point can simultaneously conduct accurate category prediction and regression estimation. However, due to different task drivers, valid features for classification and regression may locate at distinct areas in the training phase. This problem is called feature misalignment. To solve it, we propose a new feature alignment method based on anchor-free object detector. Firstly, a global receptive field adaptor (G-RFA) is designed by incorporating the feature pyramid networks (FPN) with the global attention mechanism, and forward features are further fine-tuned with a deformable-subnet (De-Subnet) to remove the influence of redundant contextual information. Then, a new feature filter strategy with a misalignment score is proposed to guide the network to focus on sampling points with aligned features. In addition, we establish mutually independent multi-layer quality distributions to model the priori information of an object on different FPN levels. Equipped with our method, the classification and regression features are aligned, and the generated foreground weight map converges to the centers of classification and regression heatmaps. Experimental results show that without bells and whistles, our method achieves 49.3% AP on MS COCO test-dev under the default 2x training schedule, outperforming related methods. Besides, experiments on PASCAL VOC demonstrate the generalization ability of our method. Code is available at https://github.com/GFENGG/featurealign.

Subspace learning using structure learning and non-convex regularization: Hybrid technique with mushroom reproduction optimization in gene selection

Gene selection as a problem with high dimensions has drawn considerable attention in machine learning and computational biology over the past decade. In the field of gene selection in cancer datasets, different types of feature selection techniques in terms of strategy (filter, wrapper and embedded) and label information (supervised, unsupervised, and semi-supervised) have been developed. However, using hybrid feature selection can still improve the performance. In this paper, we propose a hybrid feature selection based on filter and wrapper strategies. In the filter-phase, we develop an unsupervised features selection based on non-convex regularized non-negative matrix factorization and structure learning, which we deem NCNMFSL. In the wrapper-phase, for the first time, mushroom reproduction optimization (MRO) is leveraged to obtain the most informative features subset. In this hybrid feature selection method, irrelevant features are filtered-out through NCNMFSL, and most discriminative features are selected by MRO. To show the effectiveness and proficiency of the proposed method, numerical experiments are conducted on Breast, Heart, Colon, Leukemia, Prostate, Tox-171 and GLI-85 benchmark datasets. SVM and decision tree classifiers are leveraged to analyze proposed technique and top accuracy are 0.97, 0.84, 0.98, 0.95, 0.98, 0.87 and 0.85 for Breast, Heart, Colon, Leukemia, Prostate, Tox-171 and GLI-85, respectively. The computational results show the effectiveness of the proposed method in comparison with state-of-art feature selection techniques.

Adaptive hybrid pansharpening: a novel approach for combining two methods to achieve superior pansharpening performance

ABSTRACT This study proposes an adaptive strategy to improve the performances of two widely used pansharpening strategies, namely the Generalized Laplacian Pyramid (GLP) with Modulation Transfer Function (MTF)-matched filter and Context-Based Decision (CBD) injection scheme (i.e. MTF-GLP-CBD) and the Sparse Representation of Injected Details (i.e. SR-D). By considering the variance information of the PAN data, the suggested method determines which part of a pansharpened image should be produced by which of these procedures. The effectiveness of the suggested technique was assessed both qualitatively and quantitatively by comparing its performance to 31 other pansharpening strategies across three different test sites with diverse features. The outcomes of the experiments demonstrated the efficacy of the suggested approach. The experiments showed that the suggested approach not only surpassed the standard MTF-GLP-CBD and SR-D algorithms, but also all the other pansharpening algorithms employed. Of all pansharpening results, the results of the suggested approach led to the best quality index values. It can also be concluded that the suggested technique has the potential to enhance the effectiveness of other pansharpening algorithms.

Low-profile dual-polarized differential-fed filtering antenna and its novel array with low sidelobe level

This paper presents a single-layer dual-polarized filtering antenna with good out-of-band suppression. By using inherent characteristic of the patch’s eigenmode and meanwhile merging four filtering strategies, i.e., parasitic patches, slot on the driving patch, and defected groundstructure (DGS), a low-profile filtering antenna design with five radiation nulls has been realized. With the designed filtering antenna, a conventional 1×2 array with a spacing of 0.91λ is first developed. In order to reduce the sidelobe level of the array, a sharing patching scheme is introduced to decrease the interspace between the elements as 0.44λ, with the same physical size as the conventional array. In addition, a concept to measure total isolation between two kinds of the polarized ports in the array, i.e., SddVH, has been derived for the first time. The prototype of the proposed array is fabricated with a profile of 0.038λ. Measured results indicate that the designed novel array has an operational band of 3.53–4.06 GHz, SddVH better than 41 dB, a sidelobe level below −18 dB, and a suppression level above 20 dB with the stopband up to 0.5 GHz and 5 GHz.