Mode Filter Research Articles

Online operational modal decomposition

This paper presents a novel approach for online operational modal decomposition utilizing nonstationary noisy measurements only. Critical problems include the online updating of modal participation factors and the estimation of mode shape components and they are successfully resolved in the proposed online operational modal decomposition framework. The algorithm is performed in a two-phase manner. In the first phase, mode shape components for the observed degrees of freedom are extracted from the nonstationary noisy measurements. In the second phase, modal coordinates are updated recursively by combining a modal extended Kalman filter and a Bayesian probabilistic approach for online updating the modal participation factors. The proposed algorithm requires neither a finite element model of the dynamical system nor the assumptions about stationarity of the measurements and, thus, it provides a computationally efficient framework for online operational modal decomposition of dynamical systems. An illustrative example and a practical application are utilized to demonstrate the applicability and efficiency of the proposed approach.

Arbitrary-Scale Video Super-resolution Guided by Dynamic Context

We propose a Dynamic Context-Guided Upsampling (DCGU) module for video super-resolution (VSR) that leverages temporal context guidance to achieve efficient and effective arbitrary-scale VSR. While most VSR research focuses on backbone design, the importance of the upsampling part is often overlooked. Existing methods rely on pixelshuffle-based upsampling, which has limited capabilities in handling arbitrary upsampling scales. Recent attempts to replace pixelshuffle-based modules with implicit neural function-based and filter-based approaches suffer from slow inference speeds and limited representation capacity, respectively. To overcome these limitations, our DCGU module predicts non-local sampling locations and content-dependent filter weights, enabling efficient and effective arbitrary-scale VSR. Our proposed multi-granularity location search module efficiently identifies non-local sampling locations across the entire low-resolution grid, and the temporal bilateral filter modulation module integrates content information with the filter weight to enhance textual details. Extensive experiments demonstrate the superiority of our method in terms of performance and speed on arbitrary-scale VSR.

On robust boundary treatments for wall‐modeled LES with high‐order discontinuous finite element methods

SummaryTo robustly and accurately simulate wall‐bounded turbulent flows at high Reynolds numbers, we propose suitable boundary treatments for wall‐modeled large‐eddy simulation (WMLES) coupled with a high‐order flux reconstruction (FR) method. First, we show the need to impose an auxiliary boundary condition on auxiliary variables (solution gradients) that are commonly introduced in high‐order discontinuous finite element methods (DFEMs). Auxiliary boundary conditions are introduced in WMLES, where the grid resolution is too coarse to resolve the inner layer of a turbulent boundary layer. Another boundary treatment to further enhance stability with under‐resolved grids, is the use of a modal filter only in the wall‐normal direction of wall‐adjacent cells to remove the oscillations. A grid convergence study of turbulent channel flow with a high Reynolds number () shows that the present WMLES framework accurately predicts velocity profiles, Reynolds shear stress, and skin friction coefficients at the grid resolutions recommended in the literature. It was confirmed that a small amount of filtering is sufficient to stabilize computation, with negligible influence on prediction accuracy. In addition, non‐equilibrium periodic hill flow with a curved wall, including flow separation, reattachment, and acceleration at a high Reynolds number (), is reported. Considering stability and the prediction accuracy, we recommend a loose auxiliary wall boundary conditions with a less steep velocity gradient for WMLES using high‐order DFEMs.

Selecting mode by the complex Berry phase in non-Hermitian waveguide lattices

Bloch oscillations (BOs) in a parity-time (PT)-symmetric Su–Schrieffer–Heeger (SSH) waveguide array are theoretically investigated. We show that the BOs are amplified or damped even for the systems to exhibit entirely real energy bands. The amplified and damped BOs stem from the complex Berry phase and closely relate to the topological properties of the lattice. For the topological nontrivial lattice, the amplification and attenuation of BOs are much more prominent than the trivial case and the output Bloch mode can be selected. Furthermore, we propose an experimental scheme and perform a numerical simulation based on a bent waveguide array. Our work uncovers the impact of the topological properties on the dynamics of the bulk Bloch modes and unveils a horizon in the study of non-Hermitian physics. The mode selection induced by the complex Berry phase may also find application in integrated photonic devices such as the mode filter.

Behavioral evidence of the modulation filterbank in an avian animal model of complex-sound perception

Behavioral detection of amplitude modulation (AM) can be adversely affected by competing AM “noise” present in the acoustic environment. Called AM masking, human studies suggest the existence of a modulation filterbank that separates sounds (signal from noise) based on differences in AM frequency. The modulation filterbank is an important theoretical advancement in hearing science because in addition to explaining AM masking, the model successfully predicts differences in speech perception across noisy listening environments with different AM statistics. Currently, there are no behavioral animal models of the modulation filterbank, presenting a serious impediment to understanding neural underpinnings. We trained budgerigars, a parakeet species, to detect AM frequencies from 64 to 400 Hz in the presence of narrowband AM maskers applied to a 2.8-kHz carrier signal. AM masker center frequencies spanned a >2 octave range centered on the target AM frequency. Behavioral AM sensitivity was evaluated with operant conditioning, a single-interval two-alternative discrimination task, and two-down one-up adaptive threshold tracking procedures. Budgerigar AM masking functions had a band-pass characteristic, consistent with the modulation filterbank and with human AM masking results for the same stimuli. To our knowledge, these are the first behaviorally estimated modulation filter shapes in a nonhuman species. [Funding: NIDCD R01-DC017519.]

Ultra-compact and low-insertion-loss silicon modal bandpass filter using a pixelated region

In this paper, a silicon modal bandpass filter using a pixelated region, in which the TE1 mode passes through while TE0 and TE2 modes are blocked, is proposed, optimized, and experimentally demonstrated. To achieve low insertion loss, ultra-compact footprint, relatively broad bandwidth and large extinction ratio, the direct-binary-search optimization algorithm and finite difference time domain method are employed to optimize the design of the pixelated region. Measurement results reveal that, for the fabricated filter, an extinction ratio of 16.91 dB and an insertion loss of 0.89 dB are realized at 1550 nm. The measured extinction ratio is larger than 16.03 dB and the insertion loss is lower than 1.04 dB from 1538.5 to 1584.5 nm. Additionally, the length of the pixelated region is only 4.8 μm.

Collaborative Encoding Method for Scene Text Recognition in Low Linguistic Resources: The Uyghur Language Case Study

Current research on scene text recognition primarily focuses on languages with abundant linguistic resources, such as English and Chinese. In contrast, there is relatively limited research dedicated to low-resource languages. Advanced methods for scene text recognition often employ Transformer-based architectures. However, the performance of Transformer architectures is suboptimal when dealing with low-resource datasets. This paper proposes a Collaborative Encoding Method for Scene Text Recognition in the low-resource Uyghur language. The encoding framework comprises three main modules: the Filter module, the Dual-Branch Feature Extraction module, and the Dynamic Fusion module. The Filter module, consisting of a series of upsampling and downsampling operations, performs coarse-grained filtering on input images to reduce the impact of scene noise on the model, thereby obtaining more accurate feature information. The Dual-Branch Feature Extraction module adopts a parallel structure combining Transformer encoding and Convolutional Neural Network (CNN) encoding to capture local and global information. The Dynamic Fusion module employs an attention mechanism to dynamically merge the feature information obtained from the Transformer and CNN branches. To address the scarcity of real data for natural scene Uyghur text recognition, this paper conducted two rounds of data augmentation on a dataset of 7267 real images, resulting in 254,345 and 3,052,140 scene images, respectively. This process partially mitigated the issue of insufficient Uyghur language data, making low-resource scene text recognition research feasible. Experimental results demonstrate that the proposed collaborative encoding approach achieves outstanding performance. Compared to baseline methods, our collaborative encoding approach improves accuracy by 14.1%.

Feature interaction and two-stage cross-modal fusion for RGB-D salient object detection

Most existing RGB-D salient object detection (SOD) methods extract features of both modalities in parallel or adopt depth features as supplementary information for unidirectional interaction from depth modality to RGB modality in the encoder stage. These methods ignore the influence of low-quality depth maps, and there is still room for improvement in effectively fusing RGB features and depth features. To address the above problems, this paper proposes a Feature Interaction Network (FINet), which performs bi-directional interaction through feature interaction module (FIM) in the encoder stage. The feature interaction module is divided into two parts: depth enhancement module (DEM) filters the noise in the depth features through the attention mechanism; and cross enhancement module (CEM) effectively interacts RGB features and depth features. In addition, this paper proposes a two-stage cross-modal fusion strategy: high-level fusion adopts the semantic information of high level for coarse localization of salient regions, and low-level fusion makes full use of the detailed information of low level through boundary fusion, and then we progressively refine high-level and low-level cross-modal features to obtain the final saliency prediction map. Extensive experiments show that the proposed model achieves better performance than eight state-of-the-art models on five standard datasets.

Ultrahigh Transverse Mode Purity by Enhanced Modal Filtering in Double‐Clad Single‐Ring Hollow‐Core Photonic Crystal Fiber

AbstractMaintaining high fundamental mode purity in low‐loss single‐ring hollow‐core photonic crystal fiber remains a key challenge. Here a double‐clad anti‐resonant‐reflecting design is reported that, through modal filtering, provides high suppression of higher‐order modes while maintaining a low‐loss fundamental mode. The structure consists of a ring of six thin‐walled capillaries fused to the inner wall of a thicker‐walled capillary that is in turn fused at a single point to the silica outer cladding. Greater than 40 dB m−1 suppression of higher order modes over a 6 nm wide window whose central wavelength can be tuned by adjusting the gas pressure is observed. The novel design paves the way to practical applications of such fibers in high‐quality laser beam delivery and low‐noise optical‐signal transmission.

Effect of flux and shear rate on E. coli recovery in tangential flow filtration through a single hollow fiber.

Pathogenic bacteria which enter a viable but non-culturable (VBNC) state impede efforts to reach detectable concentrations required for PCR methods. This motivated a strategy for tangential flow filtration to concentrate bacteria in aqueous samples while maintaining the bacteria in a viable state, maximizing their recovery and achieving high fluxes through a single hollow fiber membrane. Filtrations were carried out for green fluorescent protein (GFP) E. coli at high shear rates (up to 27,000 sec-1) through 0.2 μm cut-off polyethersulfone (PES) microfilter membranes or 50 kDa polysulfone (PS) ultrafilter membranes. High shear minimized bacterial attachment on membrane surfaces, which would otherwise occur due to forced convection of the particles to the membrane surface at high flux conditions. Single fiber filter modules were constructed to facilitate concentration of Escherichia coli at fluxes ranging from 55 to 4500 L m-2 h-1. The effect of high shear rates on bacterial viability was found to be minimal with bacterial losses during filtration caused principally by their accumulation on the membrane surface. Recoveries of 90% were achievable at high shear rates when the average flux was ≤300 L m-2 h-1. This corresponded to a 3-h filtration time for a 225 mL sample through a single hollow fiber. Detectable bacteria concentrations of 1800 colony-forming unit (CFU)/mL were achieved for starting concentrations of 140 CFU/mL.

A novel piecewise nonlinear companding scaling scheme for FBMC PAPR reduction

The performance of offset quadrature amplitude modulation filter bank multi carrier (OQAM-FBMC) has been noticed to be affected by a high peak-to-average power ratio (PAPR). High PAPR causes the transmit power amplifier to become saturated, improving the bit error rate (BER) and causing out of band (OOB) distortions. In this paper, the PAPR of FBMC systems can be reduced by using shifting and scaling the trigonometric companding technique associated with a reduced BER performance and the power spectral density (PSD) with low OOB distortions. The characteristics of low peak signals are not changed by this companding scheme, while the characteristics of high peak signals are transformed into scaling piecewise nonlinear trigonometric functions. By tuning the parameters the suggested technique maintains the average signal power level unchanged. The theoretical analysis of the proposed scheme is also studied and also compared with the simulation results. The performance of the trigonometric companding scheme is also analysed with the μ-law and A-law existing companding methods and the suggested method obtaining 3 dB and 3.2 dB PAPR improvement respectively. The proposed technique determined to be better in terms of BER performance and spectrum efficiency.

Novel techniques for in situ estimation of shear-wave velocity and damping ratio through MASW testing – I: a beamforming procedure for extracting Rayleigh-wave phase velocity and phase attenuation

SUMMARY A robust, in situ estimate of shear-wave velocity VS and the small-strain damping ratio DS (or equivalently, the quality factor QS) is crucial for the design of buildings and geotechnical systems subjected to vibrations or earthquake ground shaking. A promising technique for simultaneously obtaining both VS and DS relies on the Multichannel Analysis of Surface Waves (MASW) method. MASW can be used to extract the Rayleigh wave phase velocity and phase attenuation data from active-source seismic traces recorded along linear arrays. Then, these data can be inverted to obtain VS and DS profiles. This paper introduces two novel methodologies for extracting the phase velocity and attenuation data. These new approaches are based on an extension of the beamforming technique which can be combined with a modal filter to isolate different Rayleigh propagation modes. Thus, the techniques return reliable phase velocity and attenuation estimates even in the presence of a multimode wavefield, which is typical of complex stratigraphic conditions. The reliability and effectiveness of the proposed approaches are assessed on a suite of synthetic wavefields and on experimental data collected at the Garner Valley Downhole Array and Mirandola sites. The results reveal that, under proper modelling of wavefield conditions, accurate estimates of Rayleigh wave phase velocity and attenuation can be extracted from active-source MASW wavefields over a broad frequency range. Eventually, the estimation of soil mechanical parameters also requires a robust inversion procedure to map the experimental Rayleigh wave parameters into soil models describing VS and DS with depth. The simultaneous inversion of phase velocity and attenuation data is discussed in detail in the companion paper.

On the Generalization and Advancement of Half-Sine-Based Pulse Shaping Filters for Constant Envelope OQPSK Modulation

On the Generalization and Advancement of Half-Sine-Based Pulse Shaping Filters for Constant Envelope OQPSK Modulation

ICNCS: Internal Cascaded Neuromorphic Computing System for Fast Electric Vehicle State-of-Charge Estimation

Accuracy and speed of Lithium-ion battery state of charge (SOC) estimation determine the reliability and stability of electric vehicle (EV), as well as promoting the development of smart home energy management system. Existing SOC estimation approaches embedded in commercial EVs still suffer from limitations of low precision, effectiveness, and relatively low robustness. To address these issues, we propose an internal cascaded neuromorphic computing system (ICNCS) via memristor circuits for EV SOC estimation. Specifically, we use three circuit modules to facilitate the design of the proposed ICNCS. Firstly, a bidirectional gated recurrent unit (Bi-GRU) circuit module is designed, enabling adequate feature extraction from the time-contextual battery information. Secondly, an attention circuit module is proposed to distinguish the useful and unimportant information related to SOC estimation. Thirdly, the Kalman filter module is constructed to eliminate the random noise caused by data transmission and processing. Finally, a series of experiments and analysis demonstrate that the proposed ICNCS has good performances in terms of accuracy (the lowest achievable RMSE and MAE are 0.853 and 0.711, respectively), time efficiency (approximately 16 20 times), and robustness (anti-noise capacity), indicating an advancement in consumer electronics applications.

Systematising clustering techniques through cross-disciplinary research, leading to the development of new methods

Clustering algorithms can help scientists gain valuable insights from data. Thereâ–™s a variety of clustering methods in use, which means there are gaps between the methods used in different fields. Associate Professor Kohei Inoue, Department of Media Design, Kyushu University, Japan, wants to bridge these gaps by investigating the relationships among various clustering methods developed in different fields, in order to systematise the world of clustering. He is bringing two decades of research activities in pattern recognition and image processing to this work. In order to clarify the relationships between different clustering methods, Inoue and the team are conducting an interdisciplinary survey. First, the researchers are working to clarify the relationship between the technologies used across different fields. So far, they have successfully clarified the relationship between the rolling guidance filter and the local mode filter. In a previous study, Inoue and his collaborators proposed a robust K-means clustering al-algorithm. The researchers demonstrated the effectiveness of their technique utilising a BBC dataset originating from BBC News. In their work, the team is collaborating with a laboratory at a university in Japan that is studying non-photorealistic rendering. They have so far published several co-authored papers, as well as having obtained results from their joint research. Ultimately, by systemising clustering technology, Inoue believes that the characteristics of each method, as well as the interrelationships between each method, can be explained and clustering technology enhanced, as well as new clustering techniques developed.

Single line of sight frame camera based on the RadOptic effect of ultrafast semiconductor detector

A new optical beam splitting method is proposed, based on which the optical frame camera capable of capturing multiple frames in a single exposure is designed and experimentally verified. The operation of the frame camera is based on an ultra-fast response semiconductor detector. It is equipped with an optical beam splitter and an optical imaging module. The ultrafast semiconductor detector receives an optical pulse that produces a transient refractive index change, and ultrafast physical processes are recorded by diffracting the probe laser through the transient phase grating. The interaction of an X-ray pulse with a semiconductor detector to produce a phase grating is simulated, based on the Monte Carlo method. The optical beam splitting mode separates a laser into two optical pulses with a certain time difference in the direction of polarization perpendicular to each other. The imaging module filters the diffracted probe laser in the spectral plane and then images multiple frames. The frame camera was used to record the temporal and spatial distribution characteristics of femtosecond laser pulses with a temporal resolution of 4.1 ps. This frame camera has great potential and value for applying to experimental studies of inertial confinement fusion.

Multiple Moving Vehicles Tracking Algorithm with Attention Mechanism and Motion Model

With the acceleration of urbanization and the increasing demand for travel, current road traffic is experiencing rapid growth and more complex spatio-temporal logic. Vehicle tracking on roads presents several challenges, including complex scenes with frequent foreground–background transitions, fast and nonlinear vehicle movements, and the presence of numerous unavoidable low-score detection boxes. In this paper, we propose AM-Vehicle-Track, following the proven-effective paradigm of tracking by detection (TBD). At the detection stage, we introduce the lightweight channel block attention mechanism (LCBAM), facilitating the detector to concentrate more on foreground features with limited computational resources. At the tracking stage, we innovatively propose the noise-adaptive extended Kalman filter (NSA-EKF) module to extract vehicles’ motion information while considering the impact of detection confidence on observation noise when dealing with nonlinear motion. Additionally, we borrow the Byte data association method to address unavoidable low-score detection boxes, enabling secondary association to reduce ID switches. We achieve 42.2 MOTA, 51.2 IDF1, and 364 IDs on the test set of VisDrone-MOT with 72 FPS. The experimental results showcase our approach’s highly competitive performance, attaining SOTA tracking performance with a fast speed.

A High Throughput Tunable Filter Module for Multispectral Imaging

AbstractTunable filters are vital not only for dynamic display but also in the miniaturized spectral imaging system. Yet, the present tunable devices possess low optical throughput, bulky volume, slow response, or complex configuration. In this paper, the study proposed and experimentally demonstrated a compact tunable filter module based on metagratings combined with liquid crystals (LC). By the synergetic effect between the birefringent LC and the dichroic metagratings, the module possesses distinct transmissive features by electrically tuning the LC material, resulting from the polarization interference effect. A simple nanoimprinting method, which is suitable for the mass production of nanostructures is adopted for low‐cost fabrication. These unique features endow the present tunable filter module with low cost, high optical throughput, and compact volume. As a proof of concept, a multispectral imager is built with this filter based on a commercial microscope. Experimental results show the system response is five times higher in optical throughput across the whole visible region than the traditional liquid crystal tunable filter and the signal‐to‐noise ratio can be improved by 7 dB. Considering the feasibility of nanoimprinting, this module paves the way for spectral imaging in low‐light conditions and on‐site scenarios such as tongue diagnosis and food quality inspection.

An FPGA-Based Hardware Architecture for P + M Class PMU Using Accuracy-Aware O-Spline Filter Selection and Modulation Detection

An FPGA-Based Hardware Architecture for P + M Class PMU Using Accuracy-Aware O-Spline Filter Selection and Modulation Detection

Signal Integrity Analysis of Modal Filters Formed by Modification of Microstrip Lines and Coplanar Waveguides

Signal Integrity Analysis of Modal Filters Formed by Modification of Microstrip Lines and Coplanar Waveguides