Search

In this paper, a high-precision rotary encoder based on a sinusoidal transmissive pattern (STP) is proposed for detecting shaft rotation. The STP disk can be realized by coating bell-shaped transmissive area on a transparent glass. We optimize the STP design shape and parameters to achieve sinusoidal transmissive characteristics. The STP-based encoder consists of a light-emitting diode, a STP disk, and a photodetector array. Experimentally, the produced STP disk had a diameter of 60 mm. 256 columns of STPs were arranged on the circumference of this disk to provide a detection pitch of 1.40625°. Each column consisted of 50 bell-shaped STPs. In principle, an incident beam from a blue-ray light emitting diode illuminates the STP of the disk, and the angle of rotation is decoded from the sinusoidal distribution of transmissive spots on the PD. Compared with a conventional incremental rotary encoder, the proposed STP-based encoder achieves a superior angular resolution.

Requirements of compressed sensing techniques targeted at miniaturized hyperspectral satellite applications include lightweight onboard hardware, high-speed sensing, low sampling rate for compressing the massive volume of typical hyperspectral data, and noise robustness for reliable data transmission to the ground station. We achieve all these aims via deep learning, and neural networks resulted from which can be implemented on-chip, thereby allowing light hardware implementation. Our neural networks were trained from small-scaled data, but, even so, the resulting encoder achieves a very low sampling rate and very high speed. Unlike typical network training, the input–output pairs are not square but stripe-like images, partly because compressed acquisition does not allow performing compression after obtaining complete data cube and partly because stripe-like acquisition well matches the popular pushbroom hyperspectral sensing schemes. Even with such hard restriction caused by nontraditional training, the resulting decoder still reconstructs the image with high accuracy. To match the requirement of pushbroom sensing, a lightweight encoder is proposed to compress the stripe-like images immediately. Meanwhile, multiscale feature fusion block (MFB) and aggregation (MFA) modules are proposed to form our decoder for enhancing the feature representation of the compressed acquisitions. Furthermore, we achieve joint spatial/spectral super-resolution (SR) progressively, ensuring accurate hyperspectral reconstruction via a low-rank-driven decoder. The encoder and decoder are trained in an end-to-end manner, where noise robustness is forced during the training stage. Comprehensive experiments demonstrate the superiority of the proposed hyperspectral compressed sensing method, as well as its one-shot transfer learning (OTL)-based extension, both quantitatively and qualitatively.

Discovering traffic anomaly propagation enables a thorough understanding of traffic anomalies and dynamics. Existing methods, such as Outlier-Tree, are not accurate to find out the trend of abnormal traffic for two reasons. First, they discover the propagation pattern based on the detected traffic anomalies. The imperfection of the detection method itself may introduce false anomalies and miss the real anomaly. Second, they develop a propagation tree of anomalies by searching continuous spatial and temporal outlier neighborhoods rather than considering from a global perspective, and thus cannot form a complete propagation tree if a spatial or temporal gap exists. In this paper, we propose a novel discovering traffic anomaly propagation method using the mesh data and enhanced traffic change peaks (en-TCP) to visualize the change of traffic anomalies (e.g., an area where vehicles are gathering or evacuating) and thus accurately capture traffic anomaly propagation. Inspired by image processing techniques, the GPS trajectory dataset in each time bin can be converted to one grid traffic image and be stored in the grid density matrix, in which the grid cell corresponds to the pixel and the density of grid cells corresponds to the Gray level ([Formula: see text]) of pixels. An enhanced adaptive filter is developed to generate traffic change graph sequences from grid traffic images in consecutive periods, and clustering en-TCP in a continuous period is to discover the propagation of traffic anomalies. The accuracy and effectiveness of the proposed method have been demonstrated using a real-world GPS trajectory dataset.

Driving cycle is an important indicator to evaluate vehicle performance and to measure fuel consumption. Its impact is not only on pollution emissions but also on asset management. Based on the real life detailed data from the city tour buses in Beijing, this paper proposes to couple a genetic algorithm based GA-K-means clustering with the Hidden Markov Model (HMM) to construct a city tour bus urban road driving cycle in Beijing. Compared to the standard driving cycle C-WTVC used for heavy commercial vehicle fuel consumption certification in China, our driving cycle model can reflect the real life situation more accurately. With our model, the influence of driving state and driving behavior on fuel consumption is also analyzed, and the fuel consumption estimation model under the corresponding driving cycle is constructed. To gain deeper insight and better fuel consumption, we further segment and construct the driving sub-cycles of peak and off-peak periods according to the difference in fuel consumption at different time periods in Beijing in September 2019. Through this fine-grain driving cycle, we can further improve the accuracy of the fuel consumption estimation model.

Dynamic Time Warping (DTW) is a widely used distance measurement in time series clustering. DTW distance is invariant to time series phase perturbations but has a quadratic complexity. An effective acceleration method must reduce the DTW utilization ratio during time series clustering; for example, TADPole uses both upper and lower bounds to prune off a large ratio of expensive DTW calculations. To further reduce the DTW utilization ratio, we find that the linear-complexity L1-norm distance (Manhattan distance) is effective enough when the time series only comprise small phase perturbations. Therefore, we propose a novel time series clustering by Minimizing Dynamic Time Warping Utilization (MiniDTW) algorithm to accelerate time series clustering. In MiniDTW, the dataset is first $greedily$ summarized into seed clusters, which comprise time series of small phase perturbations, by L1-norm distance. Then, we develop a new Sparse Symmetric Non-negative Matrix Factorization (SSNMF) algorithm, which factorizes the DTW distance matrix of seed cluster centers, to merge the seed clusters into the final clusters. The experiments on UCR time series datasets demonstrate that MiniDTW, pruning 98.52% of the DTW utilization, is better than the counterpart method, TADPole, which only prunes 75.56% of the DTW utilization; and thus MiniDTW is 10 times faster than TADPole.

BACKGROUND After an operation, the shoulder and wrist might not be able to lift and swing freely, and must be assisted with rehabilitation training. OBJECTIVE In this paper, Kinect combined with multiple sensors of a Bluetooth ball is proposed to improve the measurement function of the arm's micro-motion trajectory, rotation amount, and acceleration, which cannot be detected by Kinect alone. METHODS We designed two virtual scene rehabilitation games for clinical trials. We performed validity analysis with a paired sample t-test. RESULTS A significance value of P*< 1 was obtained, and the arm lift angle shows an improvement from 30∘ to 60∘, indicating that the range of motion of the hand and shoulder is gradually improving. CONCLUSION Experiments show that virtual games combined with multiple sensors can better understand the patient's rehabilitation situation.

The opto-coupler is usually used for isolated resonant converter to feedback the secondary-side output information. However, the characteristics of the opto-coupler are easily affected by the operating conditions, which will cause higher standby power loss. In this article, a novel primary-side controlled isolated half-bridge resonant converter with output voltage and current estimation for LLC and series resonant converter (SRC) operations is presented. The output voltage is estimated by sampling the auxiliary winding voltage when the output diode current flows to zero, the output current is estimated by integrating the resonant current until the transformer decoupled. The operating principles of the half-bridge resonant converter and the design criteria of the key component parameters are addressed and analyzed. Also, the methods for estimating output voltage and output current are discussed in detail. Finally, an experimental prototype with rated power 100 W half-bridge resonant converter for input voltage 280-342 V and output voltage 24 V is built with microcontroller to validate the proposed primary-side regulation technique. Experimental results show the output voltage and current can be controlled with the error less than 1% and 8%, respectively. The highest efficiency of the system is 95.2%.

Shapelets are subsequences of time-series that represent local patterns and can improve the accuracy and the interpretability of time-series classification. The major task of time-series classification using shapelets is to discover high quality shapelets. However, this is challenging since local patterns may have various scales/lengths rather than a unified scale. In this paper, we resolve this problem by discovering shapelets with multiple scales. We propose a novel Multi-Scale Shapelet Discovery (MSSD) algorithm to discover expressive multi-scale shapelets by extending initial single-scale shapelets (i.e., shapelets with a unified scale). MSSD adopts a bi-directional extension process and is robust to extend single-shapelets obtained by different methods. A supervised shapelet quality measurement is further developed to qualify the extension of shapelets. Comprehensive experiments conducted on 25 UCR time-series datasets show that multi-scale shapelets discovered by MSSD improve classification accuracy by around 10% (in average), compared with single-scale shapelets discovered by counterpart methods.

Implicit authentication mechanisms are expected to prevent security and privacy threats for mobile devices using behavior modeling. However, recently, researchers have demonstrated that the performance of behavioral biometrics is insufficiently accurate. Furthermore, the unique characteristics of mobile devices, such as limited storage and energy, make it subject to constrained capacity of data collection and processing. In this paper, we propose an implicit authentication architecture based on edge computing, coined Edge computing-based mobile Device Implicit Authentication (EDIA), which exploits edge-based gait biometric identification using a deep learning model to authenticate users. The gait data captured by a device’s accelerometer and gyroscope sensors is utilized as the input of our optimized model, which consists of a CNN and a LSTM in tandem. Especially, we deal with extracting the features of gait signal in a two-dimensional domain through converting the original signal into an image, and then input it into our network. In addition, to reduce computation overhead of mobile devices, the model for implicit authentication is generated on the cloud server, and the user authentication process also takes place on the edge devices. We evaluate the performance of EDIA under different scenarios where the results show that i) we achieve a true positive rate of 97.77% and also a 2% false positive rate, and ii) EDIA still reaches high accuracy with limited dataset size.

In smart grid, the “smartness” usually comes from the two-way communication between the utility and its users as well as the sensing along the transmission lines. As a result, securing grid data that are transmitted over public network has always been one big challenge. Different authentication schemes have been proposed to address this security problem. However, most of them still suffer from privacy and/or performance issues. In this paper, we propose a lightweight anonymous authentication and key agreement scheme for smart grid, which allows the smart meter and the service provider to authenticate each other and establish a shared session key between them. Compared with existing authentication solutions for smart grid, it achieves fast mutual authentication between the smart meter and the service provider while ensuring the smart meter anonymity and untraceability. We evaluate the proposed scheme using Real-or-Random Oracle Model as well as the actual hardware prototype implementation. The results show that our proposed scheme can outperform the state-of-the art schemes in terms of the performance overhead while ensuring privacy protection.

Soft, lightweight, underactuated assistive gloves (exogloves) can be useful for enhancing the capabilities of a healthy individual or to assist the rehabilitation of patients who suffer from conditions that limit the mobility of their fingers. However, most solutions found in the literature do not offer individual control of the fingers, hindering the execution of different types of grasps. In this letter, we focus on the development of a soft, underactuated, tendon-driven exo-glove that is equipped with a muscle-machine interface combining Electromyography and Forcemyography sensors to decode the user intent and allow the execution of specific grasp types. The device is experimentally tested and evaluated using different types of experiments: first, grasp experiments to assess the capability of the proposed muscle machine interface to discriminate between different grasp types and second, force exertion capability experiments, which evaluate the maximum forces that can be applied for different grasp types. The proposed device weighs 1150 g and costs $\sim$ 1000 USD (in parts). The exoglove is capable of considerably improving the grasping capabilities of the user, facilitating the execution of different types of grasps and exerting forces up to 20 N.

The most important components of the internal nut thread inspection process are measuring the pitch and pitch diameter value and comparing these with their respective specifications, as a test nut is considered to be abnormal when its pitch or pitch diameter does not meet specifications. This paper focuses on defects arising when the pitch diameter does not meet its specification. Because conventional contact strategies are not suited to the real-time inspection of internal thread defects, there have been a number of recent attempts to develop noncontact methods for real-time measurement. Some approaches have applied laser triangulation techniques in which reflected light is used to measure the z-axis depth on an object’s surface. Although point lasers have been shown to be highly precise in measuring pitch diameter, the optical architecture involved in these approaches is complicated and, as the laser must be scanned in a pointwise manner to derive cross-sectional measurements, the measurement process is too long to be compatible with real-time defect inspection. In this papere, an image-based method employing a line laser projector was used to develop a real-time image-based nut pitch diameter defect inspection system. The proposed method, which improves upon the conventional, widely used template-matching method for rapid defect detection, was shown to be effective through experimental validation.

Wireless Sensor Networks (WSNs) have been widely used as the communication system in the Internet of Things (IoT). In addition to the services provided by WSNs, many IoT-based applications require reliable data delivery over unstable wireless links. To guarantee reliable data delivery, existing works exploit geographic opportunistic routing with multiple candidate forwarders in WSNs. However, these approaches suffer from serious Denial of Service (DoS) attacks, where a large number of invalid data are deliberately delivered to receivers to disrupt the normal operations of WSNs. In this paper, we propose a selective authentication-based geographic opportunistic routing (SelGOR) to defend against the DoS attacks, meeting the requirements of authenticity and reliability in WSNs. By analyzing statistic state information (SSI) of wireless links, SelGOR leverages an SSI-based trust model to improve the efficiency of data delivery. Unlike previous opportunistic routing protocols, SelGOR ensures data integrity by developing an entropy-based selective authentication algorithm, and is able to isolate DoS attackers and reduce the computational cost. Specifically, we design a distributed cooperative verification scheme to accelerate the isolation of attackers. This scheme also makes SelGOR avoid duplicate data transmission and redundant signature verification resulting from opportunistic routing. The extensive simulations show that SelGOR provides reliable and authentic data delivery, while it only consumes 50% of the computational cost compared to other related solutions.

Vehicular ad hoc networks (VANETs) in which vehicles act as the mobile nodes provide a wide variety of services, such as audio and video surveillance. However, such networks suffer from an important problem of service delivery reliability as the network performance degrades significantly in the presence of misbehaving vehicles. Most of the existing works either assume that vehicles' misbehaviors are constant or ignore heterogeneous traffic for multiservice (a mix of video, voice, and data traffic). In this paper, we investigate the problem of trust-based multiservice delivery via integrated misbehavior detection and tolerance for fault-aware VANETs. To model the effects of time-varying misbehaviors, modern fog computing could help to analyze and to store related data in VANETs while evaluating the dynamic trust weights based on attribute parameters of each vehicle. Then, we incorporate these weights into our service delivery framework that integrates trustworthy vehicle selection for misbehavior detection and uses differential resource allocation to achieve misbehavior tolerance. We present a multi-path selection criterion based on the trust evaluation and design a trust-aware heterogeneous traffic allocation algorithm over multiple routing paths. Finally, our scheme is evaluated using extensive simulations where we show that (i) approximately 12% - 40% higher successful service delivery can be achieved by our routing algorithm at the expense of acceptable delay loss compared to other three routing protocols; (ii) our trust-aware traffic allocation algorithm can gain 10% - 20% higher effective network utility and better fairness than that of the baseline solution; and (iii) the integrated scheme significantly improves both effective utility ratio and the path-usage proportion by comparing it with only routing scheme and only traffic allocation scheme.

In this study, we make use of both the specific method of Monte Carlo simulation and the spot-futures parity with the cost of carry to establish a dynamic price model of Bitcoin futures and to conduct the appraisals and numerical analyses. More specifically, the electricity fees and equipment costs are taken into account and the proposed model is thereby built. Numerical results show that various cost factors have significant effects on the Bitcoin futures price. We employ Monte Carlo simulation to approximate the Bitcoin futures price and we use Python to program the computations.

This work proposes a comprehensive framework of biocatalytic continuous manufacturing of sitagliptin, the active pharmaceutical ingredient of the leading dipeptidyl peptidase-4 inhibitor antidiabetic drug. Continuous manufacturing has the advantages of quality consistency, reduced waste generation, and cost-effectiveness in comparison to batch processes. Furthermore, compared to traditional catalysts, biocatalysts have lighter environmental footprints. An end-to-end continuous manufacturing process is designed and the reaction kinetics of the biocatalytic reaction is determined according to the published data. Based on the steady-state model of a plug-flow microreactor, the optimal productivity is determined to be 2.6 × 10–2mol h–1 using surrogate-based optimization. In addition, an assessment of the process’ environmental impacts demonstrates its sustainability with a lower E-factor of 53 compared to 200 of traditional processes. A comprehensive techno-economic analysis has also been performed, validatin...

Response inhibition has been widely explored using the stop signal paradigm in the laboratory setting. However, the mechanism that demarcates attentional capture from the motor inhibition process is still unclear. Error monitoring is also involved in the stop signal task. Error responses that do not complete, i.e., partial errors, may require different error monitoring mechanisms relative to an overt error. Thus, in this study, we included a “continue go” (Cont_Go) condition to the stop signal task to investigate the inhibitory control process. To establish the finer difference in error processing (partial vs. full unsuccessful stop (USST)), a grip-force device was used in tandem with electroencephalographic (EEG), and the time-frequency characteristics were computed with Hilbert–Huang transform (HHT). Relative to Cont_Go, HHT results reveal (1) an increased beta and low gamma power for successful stop trials, indicating an electrophysiological index of inhibitory control, (2) an enhanced theta and alpha power for full USST trials that may mirror error processing. Additionally, the higher theta and alpha power observed in partial over full USST trials around 100 ms before the response onset, indicating the early detection of error and the corresponding correction process. Together, this study extends our understanding of the finer motor inhibition control and its dynamic electrophysiological mechanisms.

Human spinal balance assessment relies considerably on sagittal radiographic parameter measurement. Deep learning could be applied for automatic landmark detection and alignment analysis, with mild to moderate standard errors and favourable correlations with manual measurement. In this study, based on 2210 annotated images of various spinal disease aetiologies, we developed deep learning models capable of automatically locating 45 anatomic landmarks and subsequently generating 18 radiographic parameters on a whole-spine lateral radiograph. In the assessment of model performance, the localisation accuracy and learning speed were the highest for landmarks in the cervical area, followed by those in the lumbosacral, thoracic, and femoral areas. All the predicted radiographic parameters were significantly correlated with ground truth values (all p

Mobile Social Networks have become one of the most convenient services for users to share information everywhere. This crowdsourced information is often meaningful and recommended to users, e.g., reviews on Yelp or high marks on Dianping, which poses the threat of Sybil attacks. To address the problem of Sybil attacks, previous solutions mostly use indirect/direct graph model or clickstream model to detect fake accounts. However, they are either dependent on strong connections or solely preserved by servers of social networks. In this paper, we propose a novel predictable approach by exploiting users’ custom patterns to distinguish Sybil attackers from normal users for the application of recommendation in mobile social networks. First, we introduce the entropy of spatial-temporal features to profile the mobility traces of normal users, which is quite different from Sybil attackers. Second, we develop discriminative entropy-based features, i.e., users’ preference features, to measure the uncertainty of users’ behaviors. Third, we design a smart Sybil detection model based on a binary classification approach by combining our entropy-based features with traditional behavior-based features. Finally, we examine our model and carry out extensive experiments on a real-world dataset from Dianping. Our results have demonstrated that the model can significantly improve the detection accuracy of Sybil attacks.

With the rapid development of information science today, multifunctional and intelligent applications have gradually become the focus of attention. In the data management system, the first consideration is the reliability of the data source, followed by the intelligent processing after the data are collected. Due to the upgrade of the Internet to the Internet of Things, the way of network information transmission has also become a problem that people need to think about. The transmission mode of network information services will be converted from the passive transmission of information by traditional servers to the form of actively pushing information. The application of intelligent push technology in the field of the Internet of Things is a prominent and important direction in the development of the Internet of Things. This article mainly introduces the research on the intelligent music push and data analysis system based on the 5G Internet of Things, with the intention of providing some ideas and directions for the research of the music intelligent push and play and data analysis system. This paper proposes a research method for music intelligent push playback and data analysis system based on 5G Internet of Things, including current intelligent push related technologies, music evaluation matrix, user dissimilarity matrix, and music feature similarity calculation. The experimental results in this paper show that with the increase in the number of users, the accuracy of the recommended results of the system under the Hadoop framework gradually stabilizes, eventually reaching 91.2%.

The Prime Focus Spectrograph (PFS) is a new optical/near-infrared multi-fiber spectrograph designed for the prime focus of the 8.2m Subaru telescope. PFS will cover a 1.3 degree diameter field with 2394 fibers to complement the imaging capabilities of Hyper SuprimeCam. To retain high throughput, the final positioning accuracy between the fibers and observing targets of PFS is required to be less than 10 µ m. The metrology camera system (MCS) serves as the optical encoder of the fiber positioners for configuring of fibers. The MCS locates at the Cassegrain focus of the Subaru telescope to cover the whole focal plane with one 50M pixel CMOS sensor. The information from MCS will be fed into the fiber positioner control system for closed loop control. The MCS was delivered to Subaru Observatory in Apr. 2018 and it had two engineering runs in Oct. 2018 and Aug. 2019. The 1st engineering run concluded that the original mirror supports need to be improved to provide better image quality. The newly designed mirror supports were installed before the 2nd engineering run. The 2nd engineering run result shows that the MCS overall position accuracy is better than 4μm and the image processing time is less than 4 seconds. The MCS is ready for the system integration with other PFS components.

The Prime Focus Spectrograph (PFS) is a new optical/near-infrared multi-fiber spectrograph design for the prime focus of the 8.2m Subaru telescope. PFS will cover 1.3 degrees diameter field with 2394 fibers to complement the imaging capability of Hyper SuprimeCam (HSC). The prime focus unit of PFS called Prime Focus Instrument (PFI) provides the interface with the top structure of Subaru telescope and also accommodates the optical bench in which Cobra fiber positioners and fiducial fibers are located. In addition, the acquisition and guiding cameras (AGCs), the cable wrapper, the fiducial fiber illuminator, and viewer, the field element, and the telemetry system are located inside the PFI. The mechanical structure of the PFI was designed with special care such that its deflections sufficiently match those of the HSC’s Wide Field Corrector (WFC) so the fibers will stay on targets over the course of the observations within the required accuracy. The delivery of PFI components started in 2017. After the verification of these components, the mechanical structure of the PFI is fully assembled in early 2019 and all Cobra positioners are integrated in summer 2020. A temperature controlled chamber with precise x-y scanner was setup for the verification of the fiber positioners. The testing of the target convergence performance of Cobra positioners is now in progress.

This paper describes the implementation of a low cost service robot to help the elderly to improve their quality of life. It includes different technologies such as Simultaneous Localization and Mapping (SLAM), Open Source Computer Vision Library (OpenCV) and ChatterBot (a machine learning, conversational dialog engine for creating chat bots) to develop intelligence for autonomous control of the robot itself and easy-to-use interfaces for the elderly. The robot equipped with a 2D low cost laser sensor and SLAM technology would be able to build a map in an unknown area and navigate the mapped area after mapping. The ChatterBot provides the learning and conversational dialog engine which provides user-friendly interaction and feedbacks as a robotic companion to the elderly. The OpenCV is mainly used to perform the face detection, face tracking and motion detection tasks. A mobile application software (App) with streaming service is also developed to allow elderly's close relatives to monitor the health status and condition of the elderly.

Popularity prediction of social posts is one of the most critical issues for social media analysis and understanding. In this paper, we discover a more dominant feature representation of text information, as well as propose a singe ensemble learning model to obtain the popularity scores, for social media prediction challenge. However, most social media prediction techniques focus on predicting the popularity score of social posts based on a single model, such as deep learning-based or ensemble learning-based approaches. However, it is well-known that the model stacking strategy is a more effective way to boost the performance on various regression tasks. In this paper, we also show that the model stacking can be modeled as a simple recurrent neural network problem with comparable performance on predicting popularity scores. Firstly, a single strong baseline is proposed based on the deep neural network with a prediction branch. Then, the partial feature maps of the last layer of our strong baseline are used to establish a new branch with an isolated predictor. It is easy to obtain multi-prediction by repeating the above two steps. These preliminary predicted scores are then formed as the input of the recurrent unit to learn the final predicted scores, called Recurrent Stacking Model (RSM). Our experiments show that the proposed ensemble learning approach outperforms other state-of-the-art methods. Furthermore, the proposed RSM also shows the superiority over our ensemble learning approach, having verified that the model stacking problem can be transformed into the training problem of a recurrent neural network.

Visual working memory (VWM) relies on sustained neural activities that code information via various oscillatory frequencies. Previous studies, however, have emphasized time-frequency power changes, while overlooking the possibility that rhythmic amplitude variations can also code frequency-specific VWM information in a completely different dimension. Here, we employed the recently-developed Holo-Hilbert spectral analysis to characterize such nonlinear amplitude modulation(s) (AM) underlying VWM in the frontoparietal systems. We found that the strength of AM in mid-frontal beta and gamma oscillations during late VWM maintenance and VWM retrieval correlated with people's VWM performance. When behavioral performance was altered with transcranial electric stimulation, AM power changes during late VWM maintenance in beta, but not gamma, tracked participants' VWM variations. This beta AM likely codes information by varying its amplitude in theta period for long-range propagation, as our connectivity analysis revealed that interareal theta-beta couplings-bidirectional between mid-frontal and right-parietal during VWM maintenance and unidirectional from right-parietal to left-middle-occipital during late VWM maintenance and retrieval-underpins VWM performance and individual differences.

The coronavirus disease 2019 (COVID-19) pandemic has killed over 0.3 million people, disrupted people’s normal lives, and severely restricted economic activities globally. In this work, a model for the next-day COVID-19 prediction in China was built based on the ensemble back-propagation neural network machine learning technique, Baidu migration index, internal travel flow index, and confirmed cases from the previous days. The 10-fold cross-validation results showed that the model performs well in estimating the next-day confirmed cases with a correlation coefficient of 0.97. To investigate the impacts of government interventions on the spread of this new coronavirus infection, the Baidu migration index and internal travel flow index multiplied by a factor of two were input into the trained machine learning model, and the results showed that the confirmed cases in the analyzed cities would increase dramatically. The correlation between the daily new confirmed cases and some meteorological factors were also analyzed, and the results revealed that these factors are not dominant in influencing the spread of this disease. Overall, the results of this work suggest that besides early diagnosis and medical treatment, a city lockdown policy is one of the most effective methods in suppressing the rapid spread of COVID-19.

Currently, valuable tickets are scanned and verified by embedding quick response (QR) codes, but few studies have embedded encrypted information into QR codes to prevent counterfeiting. In the existing literature on color image-based QR codes, there is room for improvement in image quality and anti-counterfeiting functions. We propose an optically decrypted microstructure overlapping image technology, hiding the information points of two-dimensional barcodes on the image reservation area in the ticket, and using the optical principle to use the lenticular lens as a decryption component, to make the image produce continuous dynamic effects and increase the difficulty of forgery. In addition, the distribution of embedded parameters and differences will also affect the distortion of hidden information, which may cause the QR code to fail to read, so the median edge detection predictor is proposed for distortion control comparison. The experimental results prove that our method provides low distortion of information-hiding technology.

Directly transmitting the huge amount of typical hyperspectral data acquired on satellite to the ground station is inefficient. This paper proposes a new compressed sensing strategy for hyperspectral imagery on spaceborne sensors systems. As the onboard computing/storage resources are limited, e.g., on CubeSat, the measurement strategy should be computationally very light. Furthermore, considering the limited communication bandwidth, a very low sampling rate is desired. Our encoder accounts for these requirements by separately recording the spatial details and the spectral information, both of which essentially require only simple averaging operators. Our measurement strategy naturally induces a reconstruction criterion that can be elegantly interpreted as a well-known fusion problem in satellite remote sensing, allowing the adoption of a convex optimization method for simple and fast decoding. Our method, termed spatial/spectral compressed encoder (SPACE), is experimentally evaluated on real hyperspectral data, showing superior efficacy in terms of both sampling rate and reconstruction accuracy.

The production process of many active pharmaceutical ingredients such as sitagliptin could cause severe environmental problems due to the use of toxic chemical materials and production infrastructure, energy consumption and wastes treatment. The environmental impacts of sitagliptin production process were estimated with life cycle assessment (LCA) method, which suggested that the use of chemical materials provided the major environmental impacts. Both methods of Eco-indicator 99 and ReCiPe endpoints confirmed that chemical feedstock accounted 83% and 70% of life-cycle impact, respectively. Among all the chemical materials used in the sitagliptin production process, trifluoroacetic anhydride was identified as the largest influential factor in most impact categories according to the results of ReCiPe midpoints method. Therefore, high-throughput screening was performed to seek for green chemical substitutes to replace the target chemical (i.e. trifluoroacetic anhydride) by the following three steps. Firstly, thirty most similar chemicals were obtained from two million candidate alternatives in PubChem database based on their molecular descriptors. Thereafter, deep learning neural network models were developed to predict life-cycle impact according to the chemicals in Ecoinvent v3.5 database with known LCA values and corresponding molecular descriptors. Finally, 1,2-ethanediyl ester was proved to be one of the potential greener substitutes after the LCA data of these similar chemicals were predicted using the well-trained machine learning models. The case study demonstrated the applicability of the novel framework to screen green chemical substitutes and optimize the pharmaceutical manufacturing process.

In recent years, high-throughput experimental techniques have significantly enhanced the accuracy and coverage of protein–protein interaction identification, including human–pathogen protein–protein interactions (HP-PPIs). Despite this progress, experimental methods are, in general, expensive in terms of both time and labour costs, especially considering that there are enormous amounts of potential protein-interacting partners. Developing computational methods to predict interactions between human and bacteria pathogen has thus become critical and meaningful, in both facilitating the detection of interactions and mining incomplete interaction maps. In this paper, we present a systematic evaluation of machine learning-based computational methods for human–bacterium protein–protein interactions (HB-PPIs). We first reviewed a vast number of publicly available databases of HP-PPIs and then critically evaluate the availability of these databases. Benefitting from its well-structured nature, we subsequently preprocess the data and identified six bacterium pathogens that could be used to study bacterium subjects in which a human was the host. Additionally, we thoroughly reviewed the literature on ‘host–pathogen interactions’ whereby existing models were summarized that we used to jointly study the impact of different feature representation algorithms and evaluate the performance of existing machine learning computational models. Owing to the abundance of sequence information and the limited scale of other protein-related information, we adopted the primary protocol from the literature and dedicated our analysis to a comprehensive assessment of sequence information and machine learning models. A systematic evaluation of machine learning models and a wide range of feature representation algorithms based on sequence information are presented as a comparison survey towards the prediction performance evaluation of HB-PPIs.

In the 14nm FinFET(Fin-shaped Field-Effect Transistor) node, SADP(Self-Aligned Double Patterning) technology has been introduced to produce Fin because of the exposure limit of 193nm DUV immersion lithography. As is known to all, pitch walking issue appears when the technology comes to SADP, so how to accurately measure pitch walking is particularly important. In this paper, we use CD-SEM(Critical Dimension Scanning Electron Microscope) to measure the CD(Critical Dimension) of Fin pitch inline, and evaluate different parameter settings or machine type to improve the accuracy of the measurement results. For sub-nanometer accuracy of line width measurement, TEM(Transmission Electron Microscope) image is used to calibrate the line width measurements as a kind of reference metrology.

Research on protein-protein interactions (PPIs) data paves the way towards understanding the mechanisms of infectious diseases, however improving the prediction performance of PPIs of inter-species remains a challenge. Since one single type of sequence data such as amino acid composition may be deficient for high-quality prediction of protein interactions, we have investigated a broader range of heterogeneous information of sequences data. This paper proposes a novel framework for PPIs prediction based on Heterogeneous Information Mining and Ensembling (HIME) process to effectively learn from the interaction data. In particular, the proposed approach introduces an ensemble process together with substantial features that generate better performance of PPIs prediction task. The performance of the proposed framework is validated on real protein interaction datasets. The extensive experiments show that HIME achieves higher performance over all existing methods reported in literature so far.

Twitter hashtags provide a high-level summary of tweets, while cluster hashtags have many applications. Existing text-based methods (relying on explicit words in tweets) are greatly affected by the sparsity of the short tweet texts and the low co-occurrence rates of hashtags in tweets. Meanwhile, semantically related hashtags but using different text-expressions may show similar temporal patterns (i.e., the frequencies of hashtag usages changing with the time), which can help capture events, opinions and synonyms. In this paper, we propose a novel clustering hashtags by their temporal patterns (CHTP) method as a complement to text-based methods. In CHTP, hashtags are represented as hashtag time series that show their temporal patterns, so, hashtag clusters can be discovered by clustering hashtag time series. Density-based clustering algorithms are suitable to discover naturally shaped hashtag clusters but they are not fine enough (use one distance threshold to define density) to differentiate clusters of various density levels. Therefore, we develop a new parameter-free Density-Sensitive Clustering (DSC) algorithm to discover clusters of different density levels and use it in CHTP to group hashtags by temporal patterns. DSC recursively partitions the dataset from coarse-grained to fine-grained (using adaptive distance thresholds) to discover hashtag clusters of different density levels. Experiments conducted on Twitter datasets show that the DSC algorithm finds hashtag clusters of different densities more effectively than counterpart methods, and CHTP (using DSC) can discover meaningful hashtag clusters, 36% of which cannot be found by the text-based approaches.

Based on the authentication scheme of Wu et al. (published in the Journal of Ambient Intelligence and Humanized Computing, Volume 8, Issue 1, pp 101–116, 2017), we propose a robust privacy protection authentication scheme to prevent malicious attacks in general wireless sensor networks. In addition, we implement the proposed scheme in Rasperry Pi to simulate a home monitoring system and validate the feasibility of the proposed method.

Time series classification is important due to its pervasive applications, especially for the emerging Smart City applications that are driven by intelligent sensors. Shapelets are sub-sequences of time series that have highly predictive abilities, and time series represented by shapelets can better reveal the patterns thus have better classification accuracy. Finding shapelets is challenging as its computational in-feasibility, most existing methods only finds shapelets with a same length or a few fixed length shapelets because the searching space of shapelets with arbitrary length is too large. In this paper, we improve the time series classification accuracy by discovering shapelets with arbitrary lengths. We borrow the idea of Apriori algorithm in association rule learning, that is, the superset shapelet candidates of a poor predictive shapelet candidate also have poor predictive abilities. Therefore, we propose a Flexible Shapelets Discovery (FSD) algorithm to discover shapelets with varying lengths. In FSD, shapelet candidates having the lower bound of length are discovered, and then we extend them into arbitrary lengths shapelets as long as their predictive abilities increases. Experiments conducted on 6 UCR time series datasets demonstrate that the arbitrary length shapelets discovered by FSD achieves better classification accuracy than those using fixed length shapelets.

In the modern era, people spend approximately 90% of their time in indoor settings, such as offices and residential buildings. As prolonged exposure to indoor environments can significantly impact health outcomes, it is important to ensure that good indoor air quality (IAQ) is achieved. The main obstacles to achieving effective control of IAQ are twofold. First, it is very difficult to monitor the values of IAQ parameters, especially within a person's breathing zone. Second, current heating ventilation and air conditioning systems are unable to rapidly predict and optimize IAQ. This study aims to obtain accurate indoor environmental parameters for rapidly predicting and optimizing IAQ. To achieve this, a computational fluid dynamics (CFD)-based back propagation neural network (BPNN) combined with a particle swarm optimizer (PSO) algorithm is proposed. Notably, the BPNN-PSO algorithm can rapidly predict and optimize IAQ while using a limited number of CFD runs. In comparison with other state-of-the-art methods for controlling IAQ, such as the artificial neural network (ANN)-genetic algorithm (GA), the BPNN-PSO algorithm further reduces the concentration of indoor air pollutants by up to 6.44% and computational costs by as much as 23.53%. The use of CFD ensures that the information obtained is accurate and allows for rapid prediction of indoor environmental conditions. The BPNN-PSO algorithm has the potential to contribute very effective and intelligent control strategies for ventilating indoor environments.

© 2019 IEEE. For classification tasks, such as protein-protein interactions (PPI), support vector machines (SVMs) have been continually utilised as a standard machine learning model. However, most practices in PPIs classifications are limited to common circumstances with small datasets and low feature dimensions, due to the big computation burden of kernel functions and quadratic optimization of SVM. Alternatively, these practical experiences might tend to employ a linear model once the dataset becomes larger, which may have exclusively lost the kernel function's potential. Since there are different defined kernels and various groups of hyperparameter, the time costs in estimating a best set of hyperparameter by traditional grid search are subsequently tremendous for PPI classification. To address this challenge, in this paper, we present a more efficient solution of hyperparameter estimation by gaining acceleration with GPU, which trains SVM efficiently and accurately with kernel functions calculation accelerated on various PPI datasets. The experiments are firstly conducted on PPI classification task, and we have exclusively evaluated the effectiveness on five public classification datasets. Our solution demonstrates a faster and more accurate performance comparing with the state-of-the-art.

One type of frequently occurring construction accident is the failure of falsework systems during demolition and in new construction work. This paper describes a falsework monitoring system utilizing IoT to detect and report abnormalities prior to the occurrence of an accident. The contents presented include the research and design of special new sensor devices, communication means, temporary support structure and the adopted IoT means to improve safety.

Clustering ensembles combine multiple partitions of data into a single clustering solution. It is an effective technique for improving the quality of clustering results. Current clustering ensemble algorithms are usually built on the pairwise agreements between clusterings that focus on the similarity via consensus functions, between data objects that induce similarity measures from partitions and re-cluster objects, and between clusters that collapse groups of clusters into meta-clusters. In most of those models, there is a strong assumption on IIDness (i.e., independent and identical distribution), which states that base clusterings perform independently of one another and all objects are also independent. In the real world, however, objects are generally likely related to each other through features that are either explicit or even implicit. There is also latent but definite relationship among intermediate base clusterings because they are derived from the same set of data. All these demand a further investigation of clustering ensembles that explores the interdependence characteristics of data. To solve this problem, a new coupled clustering ensemble ( CCE ) framework that works on the interdependence nature of objects and intermediate base clusterings is proposed in this article. The main idea is to model the coupling relationship between objects by aggregating the similarity of base clusterings, and the interactive relationship among objects by addressing their neighborhood domains. Once these interdependence relationships are discovered, they will act as critical supplements to clustering ensembles. We verified our proposed framework by using three types of consensus function: clustering-based, object-based, and cluster-based. Substantial experiments on multiple synthetic and real-life benchmark datasets indicate that CCE can effectively capture the implicit interdependence relationships among base clusterings and among objects with higher clustering accuracy, stability, and robustness compared to 14 state-of-the-art techniques, supported by statistical analysis. In addition, we show that the final clustering quality is dependent on the data characteristics (e.g., quality and consistency) of base clusterings in terms of sensitivity analysis. Finally, the applications in document clustering, as well as on the datasets with much larger size and dimensionality, further demonstrate the effectiveness, efficiency, and scalability of our proposed models.

Diabetes alone does not affect the health, but its complications always bring danger to people’s health. To avoid the development of diabetes and posterior to discover diabetes, checking blood glucose level is an effective way. The development of glucose screening measurements is introduced. Non-invasive glucose monitor methods, as accurate, safe, convenient and comfortable methods, are also illustrated and compared in this paper. Also, an Intelligent dynamic health inspection is designed with a proposed non-invasion glucose monitor measurement.

This paper deals with the tire dynamic characteristics, which enable to forecast the trend of vibration signals, obtained from the tire-road system. Theory of the long-range dependence and the finite-element method were employed in this paper. To forecast the resonance effects, we use different frequency modes to avoid the appearance of undesirable inherent frequencies. The vibration intensity of the tire is related to the attrition rate with long-range dependence, which is distinguished by the Hurst effect. The model data can be obtained by the tire vibration sequence from a common pavement. Consequently, we used the fractional auto-regressive integrated moving average approach to forecast the vibration trend and the tire spectrum characteristics simultaneously. The proposed method can be useful for the monitoring of driving performance, tire design, and driving comfort improvement.

Heating ventilation and air conditioning (HVAC) systems provide a comfortable indoor thermal environment, but in the process of attaining appropriate indoor thermal comfort levels, they usually entail high energy consumptions. It is therefore imperative to balance thermal comfort value with energy consumption. However, such research currently faces two problems: one, it is difficult to obtain accurate parameters pertaining to the indoor environment of buildings, particularly near heat source areas; two, it is the diametrical nature of having to simultaneously maintain thermal comfort and keep energy consumption low. Therefore, this study aims to propose a rapid thermal comfort level prediction and optimization algorithm, as well as a method to minimize the energy consumption using only a computational fluid dynamic (CFD) database that is compact in size. Firstly, CFD is used to implement the database that stores data on indoor airflow and temperature distributions of different building structures and indoor conditions. Next, using the database as a basis, a back-propagation neural network (BPNN) is developed to predict the thermal comfort level. The adaptive grey wolf optimizer (GWO) algorithm is then applied to optimize the thermal comfort value, and the latest control methods: the artificial neural network (ANN)-genetic algorithm (GA) and ANN-particle swarm optimizer (PSO) algorithm are compared against the BPNN-based adaptive GWO method. The results show that the BPNN-based adaptive GWO algorithm can rapidly predict the thermal comfort level and have strong optimization ability. Meanwhile, 1.01% of energy savings are achieved.

The industrial internet of things (IIoT) is an intelligent network of industrial components that can communicate with each other to enhance the performance of manufacturing and industrial processes. For the many advantages, such as less no contracts, free to develop and high interoperability, in this work we propose using open-source technologies to develop IIoT platforms for information integration and data analysis. To realize the development and deployment of open source IIoTs, we present our framework OPIIoT (Open Communication Protocol Platform for Industrial Internet of Things) with the system construction cycle to solve various integration problems of operational technology and information technology. Moreover, we describe the underlying design principles and the implementation details of the proposed framework are given to make it more understandable and applicable to the IIoT platform developers. To verify the proposed framework, we have conducted both qualitative and quantitative evaluations . We have compared the important features of our OPIIoT and several commercial platforms for industry, performed trials on data communication and analytics for information supports, and demonstrated the web interface for system implementation. The results prove the efficiency and effectiveness of the proposed framework.

Service recommendation systems have been trying to utilize context-aware information to recommend services that better meet the needs of the service consumers. However, current context-aware service recommendation techniques are mainly based on individual intelligence or the local knowledge of users, and do not take into consideration the common knowledge among different users. To address this, recent research has attempted to use role-based approaches to recommend services to other members within the same context group. However, these proposed algorithms are inefficient and may not scale to cope with the large amount of mobile traffic in the real-world. This paper proposes novel algorithms with better runtime complexity, and further extends them to a MapReduce style to take advantage of popular distributed computing platforms. Experiments running on a medium-sized high performance computing cluster demonstrate that our proposed algorithms outperform previous work in runtime complexity and scalability.

The disaster brought by heavy rain has become more and more serious due to global warming. The intention of this paper is to automatically produce a route which does not pass through flooded areas to ensure the safety of travelers. We explore the techniques of using cloud services, which are very popular recently due to their efficiency and availability. The basic idea of the proposed approach is to first get the usual best path from the online route API provided by cloud services. If any portion of this route is flooded and unpassable, we identify their nearby unflooded intersections. From all the candidate sets, we then pick those which can form the shortest path between the source and the target locations, and invoke the route API again with the chosen intersections as waypoints to re-plan the route. Spatial indices and corresponding algorithms are designed to expedite the above process. We have implemented the proposed approach and conducted extensive experiments with real and synthetic data-s...

Devices embedded with position tracking facilities are now widely available, such as smartphones, smartwatches, vehicle location trackers, etc. However, data mining and advanced analytics are rarely bundled with these devices that limits their utility. In this paper, we present the design of a generic, programmable position tracking platform, namely CQtracker. In particular, this platform is incorporated with a cloud-based engine of advanced analytics. CQtracker is constructed based on a concept of system-of-systems service architecture to deliver data-system-as-a-service. It is designed for the consumption by a variety of spatio-temporal applications. Spatio-temporal data exhibit strong heterogeneous patterns, data sparseness and distribution skewness. Hence, they are difficult to analyze. CQtracker reveals relationships and structures from these data by self-regularized time-varying dynamic Bayesian networks. In addition, a Bayesian parameter estimation approach is applied to an epidemic model for outbreak predictions. Sample applications are presented in this paper, in which CQtracker successfully reveals the evolution of time-varying structures from traffic trajectories.

In this study, we explore paper characteristics that facilitate the knowledge flow from science to technology by using the patent-to-paper citation data. The linear growth trajectory of the number of patent citations to a scientific paper over time is used to measure the dynamism of its utilization for technology applications. The citation data used were obtained from the USPTO database based on two 5-year citation windows, 2001–2005 and 2009–2013. The former included patent citations to the publications in the Thomson Reuters Web of Science in 1998, and the latter included those in 2006. Only the publications in the top ten most frequently cited subject categories in the Web of Science were selected. By using growth modeling, we have found that the mean slope of the trajectory is significant. Moreover, the paper citation count, the ranking factor of the journal in which the paper was published, whether the paper is an industrial publication, and whether it is a review article have been identified to exert significant effects on the growth of the citation of scientific literature by patented inventions. Some policy implications are discussed.