Multi-channel Framework Research Articles

Optimal Security Control for CPSs Under Hybrid Attacks Over a Multi-Channel Framework

Optimal Security Control for CPSs Under Hybrid Attacks Over a Multi-Channel Framework

EXPLORING SELF-SUPERVISED CONTRASTIVE LEARNING OF SPATIAL SOUND EVENT REPRESENTATION.

In this study, we present a simple multi-channel framework for contrastive learning (MC-SimCLR) to encode 'what' and 'where' of spatial audios. MC-SimCLR learns joint spectral and spatial representations from unlabeled spatial audios, thereby enhancing both event classification and sound localization in downstream tasks. At its core, we propose a multi-level data augmentation pipeline that augments different levels of audio features, including waveforms, Mel spectrograms, and generalized cross-correlation (GCC) features. In addition, we introduce simple yet effective channel-wise augmentation methods to randomly swap the order of the microphones and mask Mel and GCC channels. By using these augmentations, we find that linear layers on top of the learned representation significantly outperform supervised models in terms of both event classification accuracy and localization error. We also perform a comprehensive analysis of the effect of each augmentation method and a comparison of the fine-tuning performance using different amounts of labeled data.

A multi-channel framework based Local Binary Pattern with two novel local feature descriptors for texture classification

Local binary pattern (LBP) has been widely used in various application fields, including object detection, texture analysis, and remote sensing. To improve discrimination performance, many LBP-based methods have been proposed for extracting different local feature information for texture classification. However, current LBP-based algorithms typically describe local features at a single sampling scale, but some significant and discriminative texture feature information is contained between different scales. Therefore, the lack of capability to extract cross-scale features can lead to losing critical texture features. Moreover, low-frequency texture information should be accorded great importance in texture classification. Additionally, if the diversity and validity of local feature extraction are lacking, the effectiveness of classification capability suffers. To address these issues, (1) we propose a completed cross-scale Local binary pattern (ccsLBP) operator to extract cross-scale texture features. (2) A mean-filtered Local binary pattern (LBPmf) operator is presented to highlight the important low-frequency texture information of texture images. (3) We build a high-performing multi-channel framework based Local binary pattern (MC-LBP) for texture classification, which combines complementary features extracted by LBP, ccsLBP, and LBPmf hybridly to form a final feature vector of the texture image. The effectiveness of the proposed MC-LBP framework is verified on 6 representative texture databases, and the experimental results demonstrate its state-of-art texture classification performance.

Heat Transfer and Reaction Characteristics of Steam Methane Reforming in a Novel Composite Packed Bed Microreactor for Distributed Hydrogen Production

The development of efficient and compact reactors is an urgent need in the field of distributed hydrogen production. Steam reforming of methane is the main method to produce hydrogen. Aiming at the problems of high heat and mass transfer resistance of the existing fixed bed reactors, and the difficulty of replacing the wall-coated catalyst in the microreactors, a composite packed bed was proposed to meet the demand of small-scale hydrogen production. The structure consists of a multi-channel framework with high thermal conductivity, which is filled with Ni/Al2O3 catalyst particles in each channel. A three-dimensional numerical model of the steam methane reforming process in the novel reactor was established using ANSYS FLUENT software. The heat transfer and reaction characteristics in the reactor were studied. Firstly, the advantages of the multi-channel skeleton in enhancing the radial heat transfer performance were verified by comparing it with the traditional randomly packed bed without the channel skeleton. Secondly, the influences of inlet velocity, inlet temperature, and heating wall temperature on the heat transfer and reaction performances in the reactor were studied, and a sensitivity factor was adopted to do the sensitivity analysis. The results show that the methane conversion rate is most sensitive to the wall temperature, while the inlet velocity and inlet temperature have less effect. Finally, the effects of two skeleton materials were studied. The results show that when the wall temperature is higher than 1200 K, there is no significant difference between these two reactors, which indicates that the use of cordierite with a lower price, but also with a lower thermal conductivity can significantly reduce the reactor’s cost. The conclusions can be used as a reference for the design of small-scale hydrogen production reactors.

Interval type-2 T-S fuzzy MPC for CPS under hybrid attacks over a multi-channel framework

In this paper, the problem of observer-based model predictive control (MPC) for a multi-channel cyber-physical system (CPS) with uncertainties and hybrid attacks is investigated via interval type-2 Takagi-Sugeno (IT2 T-S) fuzzy model. Both denial-of-service (DoS) and false data injection (FDI) attacks are studied due to the vulnerability of wireless transmission channels. The objective of the addressed problem is to improve the security performance of the multi-channel CPS under malicious attacks, which has not been adequately investigated in the existing MPC algorithms. Moreover, uncertainties which appear not only in the membership functions but in both state and input matrices are considered. In this paper, different from the method that FDI attacks are handled by the bounded functions, an off-line observer is designed to actively defend against the FDI attacks. Meanwhile, an on-line MPC optimization algorithm, which minimizes the upper bound of objective function respecting input constraints, is presented to obtain the secure controller gains. Finally, an illustrative example is provided to verify the effectiveness and superiority of presented approach.

ECG signal classification based on adaptive multi-channel weighted neural network

The intelligent diagnosis of cardiovascular diseases is a topic of great interest. Many electrocardiogram (ECG) recognition technologies have emerged, but most of them have low recognition accuracy and poor clinical application. To improve the accuracy of ECG classification, this paper proposes a multi-channel neural network framework. Concretely, a multi-channel feature extractor is constructed by using four types of filters, which are weighted according to their importance, as measured by kurtosis. A bidirectional long short-term memory (BLSTM) network structure based on attention mechanism is constructed, and the extracted features are taken as the input of the network, and the algorithm is optimized by attention mechanism. An experiment conducted on the MIT-BIH arrhythmia database shows that the proposed algorithm obtains excellent results, with 99.20 % specificity, 99.87 % sensitivity, and 99.89 % accuracy. Therefore, the algorithm is practical and effective in the clinical diagnosis of cardiovascular diseases.

An automatic framework for evaluating the vascular permeability of bone metastases from prostate cancer

Objectives. Vascular permeability can reflect tumorigenesis and metastasis. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) can assess microvascular permeability by pharmacokinetic parameter estimation. Most estimation methods require manually selected arterial input function (AIF) or reference regions. However, the result will be unstable due to the annotation, which relies on personal experience. In this study, we propose an automatic framework for evaluating vascular permeability of bone metastases from prostate cancer without selecting AIF. Materials and methods. This retrospective study comprised of 15 prostate cancer patients with bone metastases. Based on clinical consensus for three typical DCE-MRI curve patterns, three characteristic curves as regularization constraints were introduced to the extended Tofts model (ETM) using clustering strategy, and the clustering-based blind identification of multichannel (CBM) framework was then proposed for pharmacokinetic parameter estimation. With automatic segmentation of the whole bone area, we obtained the estimation of the pharmacokinetic parameters in the bone area and quantified for bone metastases. Two experienced radiologists compared the CBM estimations with the diagnostic results and we compared the estimations with those of the ETM in bone metastasis regions to evaluate the feasibility of the CBM framework. Results. The higher signal regions of K trans and K ep indicated the metastasis of prostate cancer, which is consistent with the cancer area marked by the radiologists. In addition, the K trans and K ep in bone metastasis regions were significantly higher than in normal bone regions (P < 0.001, P < 0.001). The consistency of estimation by using the CBM framework and conventional ETM method was confirmed by Bland–Altman analysis. Conclusion. The proposed CBM framework can provide a fully automatic and reliable quantitative estimation of vascular permeability for bone metastases in prostate cancer patients.

Multi-Channel Generative Framework and Supervised Learning for Anomaly Detection in Surveillance Videos.

Recently, most state-of-the-art anomaly detection methods are based on apparent motion and appearance reconstruction networks and use error estimation between generated and real information as detection features. These approaches achieve promising results by only using normal samples for training steps. In this paper, our contributions are two-fold. On the one hand, we propose a flexible multi-channel framework to generate multi-type frame-level features. On the other hand, we study how it is possible to improve the detection performance by supervised learning. The multi-channel framework is based on four Conditional GANs (CGANs) taking various type of appearance and motion information as input and producing prediction information as output. These CGANs provide a better feature space to represent the distinction between normal and abnormal events. Then, the difference between those generative and ground-truth information is encoded by Peak Signal-to-Noise Ratio (PSNR). We propose to classify those features in a classical supervised scenario by building a small training set with some abnormal samples of the original test set of the dataset. The binary Support Vector Machine (SVM) is applied for frame-level anomaly detection. Finally, we use Mask R-CNN as detector to perform object-centric anomaly localization. Our solution is largely evaluated on Avenue, Ped1, Ped2, and ShanghaiTech datasets. Our experiment results demonstrate that PSNR features combined with supervised SVM are better than error maps computed by previous methods. We achieve state-of-the-art performance for frame-level AUC on Ped1 and ShanghaiTech. Especially, for the most challenging Shanghaitech dataset, a supervised training model outperforms up to 9% the state-of-the-art an unsupervised strategy.

Automatic lung sound cycle extraction from single and multichannel acoustic recordings

A lung sound signal (LSS) consists of a series of inhalations–exhalations or lung sound cycles (LSCs), which provide valuable information about the status of the lungs. Currently, semiautomatic techniques are used to extract LSCs from an LSS. These face limitations in terms of extra cost and effort due to the need of reference signal and inconvenience caused to the patients. Automatic LSC extraction from lung sound (LS) recording can overcome these limitations. In this work, a novel signal processing based method is proposed for extraction of LSCs automatically. At first, the log variance features are calculated from preprocessed LSS, which represent an approximation of the respiratory flow and envelope, but it exhibits spikes. A novel filter-based approach is implemented to smoothen the envelope for better representation of the LSCs’ onset and offset points. The filtered envelopes representing the LS flow are selected through a majority voting technique employing single and multichannel frameworks. The study is conducted on 32 normal and 90 diseased subjects. The mean accuracy (ACC) and onset–offset error (τ) observed for normal category are 94.61% and 0.22 s for both multichannel and single channel frameworks. The same for diseased categories are within 88.96–94.18% and 0.31–0.34 s in multichannel framework and 86.75–92.73% and 0.28–0.34 s in single channel framework. These results are found to be superior when compared with a recently proposed method. The work addresses an important step towards non-invasive computer-aided LS analysis by automated segmentation of LSS without using any additional sensor.

Detection of coronary artery atherosclerotic disease using novel features from synchrosqueezing transform of phonocardiogram

ObjectiveAtherosclerotic coronary artery disease (CAD) detection through a simple, non-invasive approach will be useful in point-of-care diagnosis. Though numerous studies have addressed CAD detection using phonocardiogram (PCG) signal, none of the studies yet have explored time-varying frequency characteristics of both systolic and diastolic phases of PCG. In this study, we propose a novel method to detect CAD using synchrosqueezing transform (SST) of cardiac cycle. MethodExperiments are performed on 960 PCG collected from four positions/channels on the left anterior chest of 40 CAD and 40 normal subjects. Initially, the temporal variation of subband entropy in SST is analyzed for each channel. Later, their complementary aspect with spectral features is exploited in a fusion framework. Decision from multiple channels are combined to further improve the performance. ResultsThe proposed entropy features from SST resulted in maximum accuracy of 81.92% in multichannel framework. Fusion with spectral features was found to improve the accuracy to 83.48%. Relative improvement of 15.81% and 6.91% in accuracy is obtained over two recently proposed techniques that considered bag-of-features and sub-band based spectral power, respectively. ConclusionSST can capture useful time-frequency information from PCG to facilitate CAD detection. The proposed fusion framework using SST and spectral features in a multichannel PCG acquisition platform performs better than other PCG based approaches. SignificanceThe work shows the potential of developing a non-invasive, inexpensive, point-of-care diagnostic CAD detection system using PCG. Such a system is expected to improve healthcare accessibility in general and reach out to the marginalized in particular.

Bayesian fusion strategy for moving target detection in multichannel SAR framework

Target detection is one of the important subfields in the research of synthetic aperture radar (SAR). It faces many challenges, due to the stationary objects, leading to the presence of a scatter signal. Many researchers have been done on target detection, and most of them prefer filter based techniques. In this work, the moving target detection in SAR using decision fusion method is proposed. The newly developed scheme is named Bayesian fusion for moving target detection (BF-MTD) as the scheme utilizes the Bayesian model for identifying the target location. Initially, the received signals from the SAR are fed through the short-time Fourier transform (STFT) and the matching filters for identifying the target location. Then, the results are fused together by the Bayesian fusion strategy for finding the actual target. For the fusion, the Naive Bayes classifier is used for determining the optimal parameter for the target detection. The simulation of the proposed BF-MTD model is evaluated by varying target, iteration; pulse repetition level and antenna turn velocity of the SAR. Simulation results reveal that the proposed BF-MTD has achieved significant performance for a detection time, missed target rate, and mean square error, respectively.

A Multi-Channel Deep Neural Network for Relation Extraction

The task of relation recognition identifies semantic relationships between two named entities in a sentence. In neural network based models, a convolutional layer is often conducted to extract representative local features of a sentence. The convolution operation is implemented through a whole sentence, without considering the structure of a sentence. Because the task to recognize entity relation is processed in sentence level, many ambiguous phenomena (e.g., polysemy) are influential rather than in a document. Capturing structural information of a sentence is helpful to solve this problem. In this paper, a multi-channel framework is presented, which uses two named entities to divide a sentence into several channels. Each channel is stacked with layered neural networks. These channels do not interact during recurrent propagation, which enables a neural network to learn different representations. In our experiments, it outperforms the widely used position embedding approach. Comparing with the state-of-the-art approaches, its performance shows a meaningful improvement.

Matching Competitor's Price in the Presence of Operational Frictions

Problem Description: A price match guarantee (PMG) is a retail offer to match a rival's retail price on the same product. We ask whether or not a brick-and-mortar retailer should offer a PMG, knowing that a consumer can buy the same product either from her or a competing, manufacturer-owned online store. Academic/Practical Relevance: We provide a new multi-channel framework that combines assortment, stocking, and pricing decisions in the presence of stochastic demand and quantity constraints. Previous structures at most addressed pricing and variety with deterministic demand. Methodology: We theorize that a manufacturer can simultaneously sell his output through an independent brick-and-mortar retailer and his online store. The manufacturer designs the product line and sets a wholesale price. He can develop the same product design for both channels. Or, he can create a separate product for each channel. The independent retailer responds to the product line design and the wholesale price by announcing a stocking quantity. Then, demand realizes, and both firms set retail prices in their stores. The brick-and-mortar retailer cannot sell more than what she ordered. Results: In the literature, PMGs are mostly seen as a profitable way to facilitate cartel pricing. Previous studies, however, did not consider the operational complexities of distributing goods to stores. In our model, which does consider such complexities, PMGs can be either profitable or unprofitable. We also find that the brick-and-mortar retailer might offer a PMG not to engage in cartel pricing but to be given a higher-quality product from the manufacturer that she would be otherwise. Finally, we find that retailers who provide PMGs may go beyond price-matching to adopt a price-beating policy. Managerial Implications: We explain that the optimality PMGs boils down to a parameter that describes order fulfillment efficiency online and a parameter that describes fulfillment efficiency at the brick-and-mortar store.

Multichannel impedance inversion for nonstationary seismic data based on the modified alternating direction method of multipliers

We have upgraded the conventional nonconvex optimization algorithm and addressed a new technique to estimate the acoustic impedance (AI) for attenuated seismic data based on the modified alternating direction method of multipliers (ADMM). To eliminate the discontinuity in the inverted AI profiles with the existing single-trace processing strategy, we construct a multichannel framework and a dimensional reduction operation is used with a brief matrix manipulation in the designed forward-simulation model. The [Formula: see text] norm constraint in the inversion function can assist to search a global optimal AI solution. In the promoted ADMM solving procedure, we invoke the limited-memory Broyden-Fletcher-Goldfarb-Shanno algorithm and the generalized iterated-shrinkage algorithm to solve the relative suboptimization problems. Our new technique could recover an absolute AI from the nonstationary seismic data directly, even though this inversion problem is seriously ill-posed and nonlinear. The inverted AI profile will be satisfied with the assumption that the calculated reflectivity is comparatively sparse corresponding to its minimum [Formula: see text]-norm. Appending this adequate constraint term to the objective function is crucially significant in reducing the number of estimated AI. Due to the consideration of seismic attenuation in the equation, the inversion approach is deployed into the nonstationary reflection data avoiding the drawbacks brought by the energy-compensation processing (e.g., inverse [Formula: see text]-filtering). Using synthetic and field data, we determine the performance of our method.

New set of multi-channel orthogonal moments for color image representation and recognition

Orthogonal moments and their invariants to similarity transformations for monochrome and gray-scale images are widely used in many pattern recognition and image processing applications. Quaternion orthogonal moments are used with color images. Recently, the multi-channel framework is proposed as a successful alternative of the quaternion orthogonal moments in representation and recognition of the color images. In this paper, a new set of multi-channel orthogonal moments and their invariants to rotation, scaling and translation (RST) is proposed for color image representation and recognition. The proposed multi-channel moments are based on the orthogonal radial substituted Chebyshev functions. The multi-channel orthogonal radial substituted Chebyshev moments (MORSCMs) are defined in polar coordinates over a unit circle. An accurate kernel-based method is utilized for accurate computation of the MORSCMs. A series of experiments is performed to validate this new set of multi-channel moments and compare its performance with the existing quaternion and multi-channel orthogonal moments. The obtained results ensure the superiority of the proposed MORSCMs over all existing moments in representation and recognition of the color images.

Resilient strategy design for cyber-physical system under DoS attack over a multi-channel framework

This paper is concerned with the design of resilient strategy for a class of cyber-physical system (CPS) under denial-of-service (DoS) attack. The DoS attack occurring in wireless network between sensor and remote estimation is considered. For the addressed CPS, a minimax control strategy is designed in the presence of DoS attack in delta domain under imperfect state measurements (IPSM). Then, a novel multi-channel transmission framework is constructed to reduce the probability of being attacked, where transmitter chooses a single channel to send data packet and meanwhile the DoS attacker decides a target channel to attack. Taking account of the control system performance, a two-player Markov stochastic game is built to model the interactive decision-making of both sides under energy budget constraints. Dynamic programming and value iteration methods are applied for the control strategy and power transmission strategy design, respectively. Numerical simulations are presented to illustrate the validity of the scheme.

Wearable Multichannel Photoplethysmography Framework for Heart Rate Monitoring During Intensive Exercise

We propose a multichannel-PPG sensor that uses truncated singular value decomposition (SVD) for accurate estimation of heart rate (HR) during high-intensity exercise. We implemented a sensor to acquire nine channel signals simultaneously, and to enable real-time monitoring of PPG signals and HR. For denoising, we used SVD programmed in the microcontroller unit and showed that the pure pulse is concentrated mostly within the first largest singular value with the related two eigenvectors by extracting the significant feature components of the pulse signals. Our proposed sensor was tested in 32 subjects during exercise of various intensities: walking and medium- and high-intensity jogging. We compared our proposed method with the single-channel method, the multichannel weighting method, and the multichannel best signal selection. Furthermore, we incorporated acceleration signals and a HR tracking algorithm, into our proposed multichannel framework and evaluated its performance using the 12 data sets from the 2015 IEEE Signal Processing Cup Database. Our method showed the lowest absolute errors in HR estimation when compared to the single-channel, multichannel weighting, and multichannel best signal selection methods, irrespective of exercise intensity. We also found that our proposed multichannel framework incorporating acceleration signals and a HR tracking algorithm provided the lowest average absolute error of 0.94 beats/m. The results show that our proposed multichannel framework can be used for accurate HR estimation during high-intensity exercise.

QoS Improvement in MANET Using Multi-Channel MAC Framework

Provision of quality of service (QoS) in computer networks has been primary design criteria for network scientists and engineers. QoS in network communication deals with throughput, bandwidth utilization, delay, link reliability, availability, security etc. Mobile ad hoc Networks (MANETs) are dynamic networks where topology changes rapidly and high adaptability is required staying connected, which is a complicated and challenging task. In the present work the notion of multi-channel communication has been extended to Mobile ad hoc Networks. Prevailing mobile ad hoc network (MANET) uses single channel approach for communication which limits their capabilities and hampers QoS. Simple single channel based MANETs struggles on performance end and can be empowered with multi-channel capabilities for enhanced QoS. This paper discusses QoS provisioning using multi-channel support in mobile ad hoc networks and associated techniques and algorithms. A multi-channel framework has been developed, tested and compared with existing multi-channels for evaluation and comparison.

Research on Quality Control of Digital Production Lines in Aviation Enterprises

Focusing on the requirements of efficient quality control of digital production lines in aviation enterprises, the framework quality control system of digital production lines were studied in this paper. First, the quality date categories of digital production lines were sorted out, a multi-source quality data acquisition scheme was proposed,. Second, the importance of key quality characteristics in quality control was discussed as well as a solution to find key quality features was established. Again, a multi-channel statistical quality control framework was constructed and a detailed implementation process was proposed. Finally, a digital production line quality control prototype system was designed and developed and validated in an aviation manufacturing enterprise.

Adaptive Multiresolution Energy Consumption Prediction for Electric Vehicles

This paper introduces an adaptive multiresolution framework for electric vehicle (EV) energy consumption estimation with real-time capability. Three key parameters, namely powertrain efficiency, wind speed, and rolling resistance, are adaptively estimated using a two-step nonlinear iterative algorithm. Based on this algorithm, a multichannel framework for high-resolution powertrain efficiency estimation is introduced. Employing the “connected vehicles” concept, more reliable trip level energy estimates are achieved by sharing sensed environmental information. In addition, state-of-charge aware energy cost prediction methods of different accuracy and complexity are introduced to combat range anxiety and reduce computational complexity during times of high energy reserves. A variety of detailed simulations illustrate the introduced concept and its benefits for future EV systems.