Search

In this article, the l₁-induced performance of the stochastic switched Boolean control network (BCN) is investigated. The switched signal is considered to follow a time-varying probability distribution, the switching of which is considered to have a random dwell time. The asynchronous state feedback control (SFC) is studied to achieve the control objective. This kind of control can avoid the failure of the control due to the inconsistency between the system mode and the control mode, so the results obtained are more general. Using the semitensor product of matrices, the algebraic form of the considered BCN is represented. Under this framework, sufficient conditions are obtained to ensure that the closed-loop system is stochastic stabilized with a prescribed l₁-induced performance level ɣ. Parameters can be solved by inequalities. In addition, when the dwell time converges to infinity, the probability distribution of the switched signal becomes fixed. Necessary and sufficient conditions are presented to ensure the stabilization of the closed system under asynchronous SFC as well as the design of the asynchronous SFC. Then, sufficient condition is obtained for the prescribed l₁-induced performance level. Examples are presented to show the effectiveness of the obtained results.

Recent studies have demonstrated the success of using the channel state information (CSI) from the WiFi signal to analyze human activities in a fixed and well-controlled environment. Those systems usually degrade when being deployed in new environments. A straightforward solution to solve this limitation is to collect and annotate data samples from different environments with advanced learning strategies. Although workable as reported, those methods are often privacy sensitive because the training algorithms need to access the data from different environments, which may be owned by different organizations. We present a practical method for the WiFi-based privacy-preserving cross-environment human activity recognition (HAR). It collects and shares information from different environments, while maintaining the privacy of individual person being involved. At the core of our approach is the utilization of the Johnson-Lindenstrauss transform, which is theoretically shown to be differentially private. Based on that, we further design an adversarial learning strategy to generate environment-invariant representations for HAR. →doWe demonstrate the effectiveness of the proposed method with different data modalities from two real-life environments. More specifically, on the raw CSI dataset, it shows 2.18% and 1.24% improvements over challenging baselines for two environments, respectively. Moreover, with the discrete wavelet transform features, it further yields 5.71% and 1.55% improvements, respectively.

A principal has to take a binary decision. She relies on information privately held by a completely biased agent. The principal cannot incentivize with transfers but can learn the agent's information at a cost. Additionally, the principal privately observes a signal correlated with the agent's type. Transparent mechanisms are optimal: unlike in standard results with correlation, the principal's payoff is the same as if her signal was public. They take a simple cut-off form: favorable signals ensure the agent's preferred action. Signals below this cut-off lead to the nonpreferred action unless the agent appeals. An appeal always triggers type verification.

This article considers the security-based passivity problem for a class of discrete-time Markov jump systems in the presence of deception attacks, where the deception attacks aim to change the transmitted signal. Considering the impact of deception attacks on network disruption, it causes the existence of time-varying delays in signal transmission inevitably, which makes the controlled system and the controller work asynchronously. The asynchronous control method is employed to overcome the nonsynchronous phenomenon between the system mode and controller mode. On the other hand, to reduce the frequency of data transmission, a resilient asynchronous event-triggered control scheme taking deception attacks into account is designed to save communication resources, and the proposed controller can cover some existing ones as special examples. Moreover, different triggering conditions corresponding to different jumping modes are developed to decide whether state signals should be transferred. A new stability criterion is derived to ensure the passivity of the resultant system although there exist deception attacks. Finally, a simulation example is given to verify the theoretical analysis.

A major challenge for online and data-driven model learning in robotics is the high sample-complexity. This hinders its efficiency and practical feasibility for lifelong learning, in particular for developmental robots that autonomously bootstrap their sensorimotor skills in an open-ended environment. In this work, we propose new methods to mediate this problem in order to permit the learning of robot models online, from scratch, and in a learning while behaving fashion. Exploration is utilized and autonomously driven by a novel intrinsic motivation signal which combines knowledge-based and competence-based elements and surpasses other state-of-art methods. In addition, we propose an Episodic Online Mental Replay to accelerate the online learning, to ensure the sample-efficiency, and to update the model online rapidly. The efficiency as well as the applicability of our methods are demonstrated with a physical 7-DoF Baxter manipulator. We show that our learning schemes are able to drastically reduce the sample complexity and learn the data-driven model online, even within a limited time frame.

The elimination of ocular artifacts is critical in analyzing electroencephalography (EEG) data for various brain-computer interface (BCI) applications. Despite numerous promising solutions, electrooculography (EOG) recording or an eye-blink detection algorithm is required for the majority of artifact removal algorithms. This reliance can hinder the model's implementation in real-world applications. This paper proposes EEGANet, a framework based on generative adversarial networks (GANs), to address this issue as a data-driven assistive tool for ocular artifacts removal (source code is available at https://github.com/IoBT-VISTEC/EEGANet). \textcolor{red}{After the model was trained, the removal of ocular artifacts could be applied calibration-free without relying on the EOG channels or the eye blink detection algorithms.} First, we tested EEGANet's ability to generate multi-channel EEG signals, artifacts removal performance, and robustness using the EEG eye artifact dataset, which contains a significant degree of data fluctuation. According to the results, EEGANet is comparable to state-of-the-art approaches that utilize EOG channels for artifact removal. Moreover, we demonstrated the effectiveness of EEGANet in BCI applications utilizing two distinct datasets under inter-day and subject-independent schemes. Despite the absence of EOG signals, the classification performance of the signals processed by EEGANet is equivalent to that of traditional baseline methods. This study demonstrates the potential for further use of GANs as a data-driven artifact removal technique for any multivariate time-series bio-signal, which might be a valuable step towards building next-generation healthcare technology.

Language-independent spoken term detection (LI-STD) refers to the process of locating the occurrences of spoken queries from speech databases of any language. This paper alization of a multilingual broad phoneme classifier (BPC) and its application for the development of an LI-STD system. This work proposes a multi-stage architecture to address the task of LI-STD for low-resourced languages, where there is limited amount of labelled training data. The proposed LI-STD system contains three stages; one label sequence matching stage and two template matching stages. A deep neural network (DNN) based BPC trained using 16 handcrafted, signal-based features is the backbone of the proposed LI-STD system. In LI-STD system, stage 1 performs a broad phoneme sequence matching, while stage 2 and 3 perform template matching on posteriorgram and feature sequence, respectively. Concatenation of multiple stages results in search space reduction for the later computationally intensive template matching stages. In order to adapt to a new/unseen language, the BPC gets retrained using selected broad phoneme labelled data of the language generated by itself. The effectiveness of the proposed system is demonstrated on a set of low-resourced Indian languages.

Despite having plethora of works, wind speed time-series forecasting capabilities are prone to errors due to its intermittent and non-stationary nature as well as the limited generalization capabilities of forecasting methods for non-Gaussian distributed data. In this paper, a hybrid method that consists of elastic variational mode decomposition (eVMD) and forecasting random convolution nodes (fRCN) is proposed to forecast the Gaussian heteroscedastic wind speed time-series. The proposed eVMD algorithm gauges the non-stationary characteristics (complexity) of the wind speed signal and thereafter decomposes the signal into its intrinsic components (ICs) accordingly. The fRCN method rely on local receptive fields (LRF) to extract features that contribute to the local variations and the global trend in each IC. These features are subsequently learned using extreme learning machines (ELM) theories. An ensemble unit is employed to learn appropriate weightages for each forecasted IC before yielding the final forecasting values. Suitability of the proposed hybrid method for wind speed forecasting is evaluated via an actual wind speed dataset and comparing against various existing hybrid methods.

As a typical big data representative, the information of global Cellular Signal Strength (CSS), defined as the signal power received by mobile phones, plays an important role in geographic flow analysis, because the density of such information can reflect the urbanization variables such as population, gross domestic product, built-up area, electric power consumption, and etc. Despite the importance, the real-time analysis of global CSS distribution remains a challenging problem due to the large data scale. In this paper, a Display-driven Computing (DisDC) technique is designed and applied to provide efficient large scale interactive CSS visualization, generating results by calculating the value of each pixel that directly for display that can dramatically improve system capacity in big data handling. Specifically, we present an efficient CSS measurement algorithm, which introduces spatial indexes and a corresponding query strategy; besides, an optimized parallel computing architecture is proposed to ensure the ability of real-time visualization. Experiments show that our approach obviously outperforms traditional methods and is capable of handling more than 40 million base stations in real-time. Moreover, an online demonstration is provided at https://github.com/MemoryMmy/CSSMap.

Intelligent bearing diagnostic methods are developing rapidly, but they are difficult to implement due to the lack of real industrial data. A feasible way to deal with this problem is to train a network through laboratory data to mine the causality of bearing faults. This means that the constructed network can handle domain deviations caused by the change of machines, working conditions, noise, and so on which is, however, not a simple task. In response to this problem, a new domain generalization framework--Whitening-Net--was proposed in this article. This framework first defined the homologous compound domain signal as the data basis. Subsequently, the causal loss was proposed to impose regularization constraints on the network, which enhances the network's ability to mine causality. To avoid domain-specific information from interfering with causal mining, a whitening structure was proposed to whiten the domain, prompting the network to pay more attention to the causality of the signal rather than the domain noise. The results of diagnosis and interpretation proved the ability of Whitening-Net in mining causal mechanisms, which shows that the proposed network can generalize to different machines, even if the tested working conditions and bearing types are completely different from the training domains.

Seizure prediction from intracranial electroencephalogram (iEEG) has great potentials to improve the life quality of epileptic patients, but faces big challenges. One major difficulty lies in that brain signal changes occasionally during long-term monitoring, due to electrode movements or the nonstationary brain dynamics. This leads to a serious situation that a predictor learned from historical data usually only works well in a short time period as long as the data do not change much. While in a long time span, the performance of the learned features decreases or even become totally invalid. To deal with the problem, we propose a domain adaptation convolutional neural network to learn robust preictal features that is invariant across different time periods. Specifically, the preictal feature is learned and enhanced by multi-scale temporal convolutions in the neural network. Based on this, a domain adaptation method is adopted to constrain that the learned features should be invariant across different time periods. Experimental results demonstrate that our approach can effectively improve seizure prediction performance against signal changes.

The performance of traffic signal phasing and timing (SPaT) plans are directly related to temporal fluctuations in traffic volumes and their distribution across competing movements at the intersection. A well-timed signal plan generally allocates green times in proportion to the observed volumes. With the advance of data collection techniques and technology, it is possible to obtain and use high-resolution traffic data using detectors at intersections to measure the real-time performance of the SPaT plan to better achieve this goal. This study introduces novel green time metrics and methods to evaluate the efficiency of green time allocation along arterials and at individual intersections using automated, real-time, and high-resolution traffic data. An empirical application is presented for four intersections along a major radial arterial in Salt Lake City, Utah and a detailed analysis of the phases and movements for a single intersection located on that arterial. The proposed methods and results can help identify opportunities to improve the signal timing plan for the entire corridor or movements within secondary approaches by identifying locations and/or time periods where signal timing does not align well with observed traffic volumes. These metrics can be automated to provide traffic engineers with an alert as to when a particular intersection should be examined as a candidate for signal retiming, and they can also help identify what changes need to be made with the current SPaT plans.

Bus priority signal control can reduce bus delay at intersections and improve bus operation efficiency. The existing research focus mainly on bus priority control methods at isolated intersections, as well as urban arterial corridors. Rarely existing works consider a network-level bus priority control. Treating regional-wide intersections can be challenging, given that more traffic directions and large multi-modal traffic volume should be considered. The conventional transit signal priority (TSP) research focus on the delay of buses and social vehicles and neglect the delay of pedestrians. To fill these gaps, this paper proposes a regional coordinated bus priority signal control (RCBPSC) method, which is a network-level bus priority control method considering pedestrian and passenger delays. This method is divided into two stages. The first stage is the regional coordinated signal control to obtain the basic signal timing schemes. The second stage is the bus priority signal control. At this stage, the timing schemes at each intersection will be adjusted according to the bus arrival time and the delays. In order to verify the effectiveness of this method, we choose four intersections in Chengdu to study. The results show that the total delay of the proposed method at the first stage (control case 1) can be reduced by 602s (2.3%) and 606s (3.7%) comparing with the conventional timing method in the peak and non-peak period. At the second stage, the proposed method can reduce more pedestrian delay than the conventional TSP method in both scenario 1 and scenario 2.

For distributed sensor systems using neural networks, each sub-network has a different electromagnetic environment, and these recognition accuracy is also different. In this paper, we propose a distributed sensor system using incremental learning to solve the problem of radio frequency fingerprint identification. First, the intelligent representation of the received signal is linearly fused into a four-channel image. Then, convolutional neural network is trained by using the existing data to obtain the preliminary model of the network, and decision fusion is used to solve the problem in the distributed system. Finally, using new data, instead of retraining the model, we employ incremental learning by fine-tuning the preliminary model. The proposed method can significantly reduce the training time and is adaptive to streaming data. Extensive experiments show that the proposed method is computationally efficient, and also has satisfactory recognition accuracy, especially at low signal-to-noise ratio (SNR) regime.\ud

Fault diagnosis has a great significance in preventing serious failures of rotating machinery and avoiding huge economic losses. The performance of the existing fault diagnosis approaches might be affected by two factors, i.e., the quality of fault features extracted from monitoring signals and the capability of fault diagnosis model. This paper proposes a new fault diagnosis method combined mel-frequency cepstral coefficients (MFCC) with a designed parallel multi-fusion convolutional neural network (MFCNN) Specifically, a MFCC-based feature extraction methos is defined to reduce the noise components in monitoring signal of rotating machinery and extract more useful low-frequency fault information for downstream task. Furthermore, a novel MFCNN is designed to enrich the high-level features after each convolution operation by using multiple activation functions, so as to improve the quality of the obtained fault features. Meanwhile, a new parallel MFCNN is constructed by using a defined structural ensemble operation to improve its diagnostic performance in different noise environments. Two typical bearing and gearbox failure datasets are applied to evaluate the performance of the proposed fault diagnosis method. The experimental results indicate that the proposed parallel MFCNN has the better diagnostic performance than other methods.

Learning to drive in urban areas is an open challenge for autonomous vehicles (AVs), as complex decision making requirements are needed in multi-task co-ordinations environments. In this paper, we propose a hybrid framework with a new perception model involving affordance learning to simplify the surrounding urban scenes for training an AV agent, along with a planned trajectory and the associated driving measurements. Our proposed solution encompasses two main aspects. Firstly, a supervised learning network is used to map the input sensory data into affordance predictions. The predicted affordances provide a low-dimensional representation of surrounding scenes of the AV in the form of key perception indicators, e.g., true or false with respect to a traffic light signal. Secondly, a deep deterministic policy gradient model that maps the perception information into a series of actions is devised. We evaluate the proposed solution using the CARLA driving simulator in an urban town and evaluate the performance in a new, unseen town under different weather conditions. The quantitative and qualitative results indicate that our proposed solution can generalize well to cope with different traffic solutions and environmental conditions. Our proposed solution also outperforms other baseline methods in a comparative study in handling various AV driving tasks with different levels of difficulty. In addition, the model trained with simulated scenes yields promising prediction results when testing on recorded video streams on real-world highway and suburb environments with varying traffic and weather conditions.

Brain-computer interface (BCI) speller is a system that provides an alternative communication for the disable people. The brain wave is translated into machine command through a BCI speller which can be used as a communication medium for the patients to express their thought without any motor movement. A BCI speller aims to spell characters by using the electroencephalogram (EEG) signal. Several types of BCI spellers are available based on the EEG signal. A standard BCI speller system consists of the following elements: BCI speller paradigm, data acquisition system and signal processing algorithms. In this work, a systematic review is provided on the BCI speller system and it includes speller paradigms, feature extraction, feature optimization and classification techniques for BCI speller. The advantages and limitations of different speller paradigm and machine learning algorithms are discussed in this article. Also, the future research directions are discussed which can overcome the limitations of present state-of-the-art techniques for BCI speller.

Gesture recognition has become increasingly important in human-computer interaction and can support different applications such as smart home, VR, and gaming. Traditional approaches usually rely on dedicated sensors that are worn by the user or cameras that require line of sight. In this paper, we present fine-grained finger gesture recognition by using commodity WiFi without requiring user to wear any sensors. Our system takes advantages of the fine-grained Channel State Information available from commodity WiFi devices and the prevalence of WiFi network infrastructures. It senses and identifies subtle movements of finger gestures by examining the unique patterns exhibited in the detailed CSI. We devise environmental noise removal mechanism to mitigate the effect of signal dynamic due to the environment changes. Moreover, we propose to capture the intrinsic gesture behavior to deal with individual diversity and gesture inconsistency. Lastly, we utilize multiple WiFi links and larger bandwidth at 5GHz to achieve finger gesture recognition under multi-user scenario. Our experimental evaluation in different environments demonstrates that our system can achieve over 90% recognition accuracy and is robust to both environment changes and individual diversity. Results also show that our system can provide accurate gesture recognition under different scenarios.

A central capability of a long-lived reinforcement learning (RL) agent is to incrementally adapt its behavior as its environment changes, and to incrementally build upon previous experiences to facilitate future learning in real-world scenarios. In this paper, we propose LifeLong Incremental Reinforcement Learning (LLIRL), a new incremental algorithm for efficient lifelong adaptation to dynamic environments. We develop and maintain a library that contains an infinite mixture of parameterized environment models, which is equivalent to clustering environment parameters in a latent space. The prior distribution over the mixture is formulated as a Chinese restaurant process (CRP), which incrementally instantiates new environment models without any external information to signal environmental changes in advance. During lifelong learning, we employ the expectation maximization (EM) algorithm with online Bayesian inference to update the mixture in a fully incremental manner. In EM, the E-step involves estimating the posterior expectation of environment-to-cluster assignments, while the M-step updates the environment parameters for future learning. This method allows for all environment models to be adapted as necessary, with new models instantiated for environmental changes and old models retrieved when previously seen environments are encountered again. Experiments demonstrate that LLIRL outperforms relevant existing methods, and enables effective incremental adaptation to various dynamic environments for lifelong learning., Comment: Accepted by IEEE Transactions on Neural Networks and Learning Systems, 2021

The robust iterative learning control (RILC) can deal with the systems with unknown time-varying uncertainty to track a repeated reference signal. However, the existing robust designs consider all the possibilities of uncertainty, which makes the design conservative and causes the controlled process converging to the reference trajectory slowly. To eliminate this weakness, a data-driven method is proposed. The new design intends to employ more information from the past input-output data to compensate for the robust control law and then to improve performance. The proposed control law is proved to guarantee convergence and accelerate the convergence rate. Ultimately, the experiments on a robot manipulator have been conducted to verify the good convergence of the trajectory errors under the control of the proposed method.

The benefit of real-time joint dynamic route guidance and signal control (DRG-SC) is usually compromised by a centralized framework since it naturally leads to an un-timely solution with the growing data-processing needs and problem-solving complexity. Mobile edge computing (MEC) pushes the data storage and computation from the remote cloud to local infrastructures and hence reduces response time and improves network bandwidth when further combined with 5G. As such, our study first develops a novel distributed framework to facilitate DRG-SC in connected vehicle (CV) environment with clarifying the MEC's vital role. The method captures the interaction of vehicles' routing and signal control, wherein we use a more realistic and accurate way to define the relationship between travel time and traffic volume. Vehicles make route decisions and cooperate to reach user optimal (UO) or system optimal (SO) traffic state. In tandem, the developed adaptive signal control (ASC) adjusts the signal timing plan with considering both the adjacent intersections' traffic volume and the vehicles' waiting time. Our method achieves significant reductions in vehicles' average departure delay, waiting time and travel time when justified by a comprehensive case study implemented in SUMO. Moreover, the effectiveness of adopting such a distributed framework in saving computation time is verified. Overall, our study provides valuable and practical insights into the intelligent operation and control.

The aim of steganography detection is to identify whether the multimedia data contain hidden information. Although many detection algorithms have been presented to solve tasks with inconsistent distributions between the source and target domains, effectively exploiting transferable correlation information across domains remains challenging. As a solution, we present a novel multiperspective progressive structure adaptation (MPSA) scheme based on active progressive learning (APL) for JPEG steganography detection across domains. First, the source and target data originating from unprocessed steganalysis features are clustered together to explore the structures in different domains, where the intradomain and interdomain structures can be captured to provide adequate information for cross-domain steganography detection. Second, the structure vectors containing the global and local modalities are exploited to reduce nonlinear distribution discrepancy based on APL in the latent representation space. In this way, the signal-to-noise ratio (SNR) of a weak stego signal can be improved by selecting suitable objects and adjusting the learning sequence. Third, the structure adaptation across multiple domains is achieved by the constraints for iterative optimization to promote the discrimination and transferability of structure knowledge. In addition, a unified framework for single-source domain adaptation (SSDA) and multiple-source domain adaptation (MSDA) in mismatched steganalysis can enhance the model's capability to avoid a potential negative transfer. Extensive experiments on various benchmark cross-domain steganography detection tasks show the superiority of the proposed approach over the state-of-the-art methods.

Although the deep learning diagnosis model has been widely used in the fault diagnosis of rotating machinery. However, these methods lack the interpretability of the diagnostic process. In other words, it is still a difficult problem to understand that the structural function and the diagnosis process in the model correspond to each other. Therefore, this paper discusses how to add multi-scale and multi attention mechanism to lightweight network. From different scales, different dimensions, combined with the fault signal characteristics of centrifugal fan, the attention structure of cross layer fusion is designed. How to integrate different functions continuously and effectively to achieve better diagnostic performance is answered. The proposed lightweight multi-scale multi-attention feature fusion network adaptively recalibrates feature weights, which effectively enhances the fault feature learning ability and anti-noise ability. Experimental results show that this network is stronger than other advanced diagnostic models.

Broad learning systems (BLSs) have attracted considerable attention due to their powerful ability in efficient discriminative learning. In this article, a modified BLS with reinforcement learning signal feedback (BLRLF) is proposed as an efficient method for improving the performance of standard BLS. The main differences between our research and BLS are as follows. First, we add weight optimization after adding additional nodes or new training samples. Motivated by the weight iterative optimization in the convolution neural network (CNN), we use the output of the network as feedback while employing value iteration (VI)-based adaptive dynamic programming (ADP) to facilitate calculation of near-optimal increments of connection weights. Second, different from the homogeneous incremental algorithms in standard BLS, we integrate those broad expansion methods, and the heuristic search method is used to enable the proposed BLRLF to optimize the network structure autonomously. Although the training time is affected to a certain extent compared with BLS, the newly proposed BLRLF still retains a fast computational nature. Finally, the proposed BLRLF is evaluated using popular benchmarks from the UC Irvine Machine Learning Repository and many other challenging data sets. These results show that BLRLF outperforms many state-of-the-art deep learning algorithms and shallow networks proposed in recent years.

In this survey, various concepts and methodologies developed over the past two decades for varying and learning the impedance or admittance of robotic systems that physically interact with humans are explored. For this purpose, the assumptions and mathematical formulations for the online adjustment of impedance models and controllers for physical human-robot interaction (HRI) are categorized and compared. In this systematic review, studies on: 1) variation and 2) learning of appropriate impedance elements are taken into account. These strategies are classified and described in terms of their objectives, points of view (approaches), and signal requirements (including position, HRI force, and electromyography activity). Different methods involving linear/nonlinear analyses (e.g., optimal control design and nonlinear Lyapunov-based stability guarantee) and the Gaussian approximation algorithms (e.g., Gaussian mixture model-based and dynamic movement primitives-based strategies) are reviewed. Current challenges and research trends in physical HRI are finally discussed.

Emotion recognition based on electroencephalography (EEG) signal features is now one of the booming big data research areas. As the number of commercial EEG devices in the current market increases, there is a need to understand current trends and provide researchers and young practitioners with insights into future investigations of emotion recognition systems. This paper aims to evaluate popular consumer-grade EEG devices’ status and review relevant studies that examined the reliability of these low-cost devices for emotion recognition over the last five years. Additionally, a comparison with research-grade devices is conducted. This paper also highlights EEG-based emotion recognition research’s key areas, including different feature extraction capabilities, characteristics, and machine learning algorithms. Finally, the main challenges for building an EEG-based emotion recognition system, focusing on the data collection process with commercial EEG devices and machine learning algorithms’ performance, are presented.

The security of computers is at risk because of information leaking through their power consumption. Attackers can use advanced signal measurement and analysis to recover sensitive data from this side channel. To address this problem, this paper presents Maya, a simple and effective defense against power side channels. The idea is to use formal control to re-shape the power dissipated by a computer in an application-transparent manner---preventing attackers from learning any information about the applications that are running. With formal control, a controller can reliably keep power close to a desired target function even when runtime conditions change unpredictably. By selecting the target function intelligently, the controller can make power to follow any desired shape, appearing to carry activity information which, in reality, is unrelated to the application. Maya can be implemented in privileged software, firmware, or simple hardware. In this paper, we implement Maya on three machines using privileged threads only, and show its effectiveness and ease of deployment. Maya has already thwarted a newly-developed remote power attack.

There is an increasing trend in the use of wireless communication along with new traffic signal control (TSC) algorithms to leverage and accommodate connected and autonomous vehicles. However, this development has increased the potential for cyber-attacks on TSC that can undermine the benefits of these new algorithms. An advanced persistent adversary can learn the behavior of TSC algorithms and launch attacks to preferentially get green time and/or to create traffic congestion in one intersection which can spread to the entire network. In this paper, we consider backpressure-based (BP-based) TSC algorithms and compare their performance under two misinformation attacks - 1) time spoofing attack in which vehicles alter their arrival times at the intersection and 2) ghost vehicle attack in which vehicles disconnect the wireless communication and thereby hide from the TSC. We show that these misinformation can influence the signal phases determined by BP-based TSC algorithms. We consider an adversary that determines a set of arriving vehicles to be attack vehicles from many candidate sets (attack strategies) in order to maximize the number of disrupted signal phases. We show that by formulating the problem as a 0/1 Knapsack problem, the adversary can explore the space of attack strategies and determine the optimal strategy that maximally compromises the performance in terms of average delay and fairness. We propose two protection algorithms, namely, auction-based (APA) and hybrid-based (HPA) algorithms and show that they are able to mitigate the impacts of the misinformation attacks.

In daily pipeline inspection, it is significant to ensure good network communication and security. With the development of drone technology, it is possible to apply drones as air routers to collect information from pipeline networks and transmit it to pipeline inspectors. It is also crucial to achieve optimal drone deployment in pipeline networks. This article proposes a two-phase evolution optimal 3-D drone layout algorithm to deploy drones in pipeline networks. First, a 3-D pipeline graph model is designed to represent the possible projection position of drones, and the objective function is proposed for optimal drone deployment. Then, in the first phase, based on the features of the 3-D pipeline graph, the drone flight rules and constraint conditions are presented to calculate the number of drones and the initial layout sequence. In the second phase, according to the objective function and the above results, every drone is continuously moved in a small area to achieve a tradeoff between signal coverage and interference. Moreover, the key parameters of the objective function can be discussed to further optimize drone deployment. Simulation results are presented to illustrate the effectiveness and advantages of the proposed algorithm.

Emotion recognition using Electroencephalography (EEG) is a convenient and reliable technique. EEG based emotion detection study can find its application in various fields such as defense, aerospace, medical, and many more. This analysis helps to understand the emotional state of mind. There are two approaches to study EEG analysis known as subject dependent and independent. In this paper, Modified Differential Entropy (MD-DE) feature extractor is proposed to detect nonlinearity and non-Gaussianity of the EEG signal. The paper adopts both approaches by conducting an EEG analysis on own generated database named as ‘IDEA- Intellect Database for Emotion Analysis’ on 14 subjects. In this work, bidirectional long short-term memory (BiLSTM) network and multilayer perceptron (MLP) network is used to classify emotional state of mind of the subjects. On the ‘IDEA’ database, subject dependent average accuracy achieved is in the order of 98.5% and for subject independent, 88.57%. To reaffirm the improvement in accuracy level, a new approach of Modified Differential Entropy and BiLSTM network is applied on the openly available SEED and DEAP database as well. This experiment established that the average accuracy of emotion detection using MD-DE and BiLSTM network is better than the established methods.

Data-driven methods, especially deep neural network, received increasing attention in machinery fault diagnosis field. Many works focus on how to design effective model while ignoring a fundamental problem, i.e., directly using raw machinery signal as the input of model. In this work, we analyze from two aspects: model mechanism and mechanical monitoring signal, it shows the limitation of learning raw data directly, which led to the research idea of improving the generalization ability of model by multi-frequency information augmentation. In order to make machinery intelligent model capture multi-frequency information more directly and actively, Multi-Frequency Augmentation framework is proposed in this paper. Firstly, we proposed a data augmentation method to split the raw sample into sample pair. And we could choose to further augment the dataset by Frequency Components Recombination, especially under few-shot scenes. Then, Multi-Frequency Capture Network is built to achieve feature augmentation by learning the sample pair. Finally, fault diagnosis is performed on testing set. The effectiveness and compatibility of Multi-Frequency Augmentation framework is verified with two experiments, which also verifies the feasibility of the proposed research idea. In addition, it could also achieve competitive performance with latest literature methods. Further discussion indicate that the proposed framework provides a new perspective to analyze the model and dataset, which has good application potential.

Abnormal event detection with the lowest latency is an indispensable function for safety-critical systems, such as cyber defense systems. However, as systems become increasingly complicated, conventional sequential event detection methods become less effective, especially when we need to define indicator metrics from complicated data manually. Although Deep Neural Networks (DNNs) have been used to handle heterogeneous data, the theoretic assurability and explainability are still insufficient. This paper provides a holistic framework for the quickest and sequential detection of abnormalities and time-dependent abnormal events. We explore the latent space characteristics of zero-bias neural networks considering the classification boundaries and abnormalities. We then provide a novel method to convert zero-bias DNN classifiers into performance-assured binary abnormality detectors. Finally, we provide a sequential Quickest Detection (QD) scheme that provides the theoretically assured lowest abnormal event detection delay under false alarm constraints using the converted abnormality detector. We verify the effectiveness of the framework using real massive signal records in aviation communication systems and simulation. Codes and data are available at.

Deep Learning Classification is often used to analyze biomedical data. One of them is to analyze the Fetal Heart Rate (FHR) signal data used to check and monitor maternal and fetal health and prevent mobility and mortality in fetuses at risk of developing hypoxia. The problem that often occurs in the data is data unbalance. Time Series Generative Adversarial Networks (TSGAN) solves data imbalance in the FHR signal and generate more data and better classification performance. Augmentation using the GAN model in this study obtained an increase in the Quality Index of 3%–44% from other models.

Deep learning methods achieve great success in many areas due to their powerful feature extraction capabilities and end-to-end training mechanism, and recently they are also introduced for radio signal modulation classification. In this paper, we propose a novel deep learning framework called SigNet, where a signal-to-matrix (S2M) operator is adopted to convert the original signal into a square matrix first and is co-trained with a follow-up CNN architecture for classification. This model is further accelerated by integrating 1D convolution operators, leading to the upgraded model SigNet2.0. The simulations on two signal datasets show that both SigNet and SigNet2.0 outperform a number of well-known baselines. More interestingly, our proposed models behave extremely well in small-sample learning when only a small training dataset is provided. They can achieve a relatively high accuracy even when 1\% training data are kept, while other baseline models may lose their effectiveness much more quickly as the datasets get smaller. Such result suggests that SigNet/SigNet2.0 could be extremely useful in the situations where labeled signal data are difficult to obtain. The visualization of the output features of our models demonstrates that our model can well divide different modulation types of signals in the feature hyper-space., Comment: 13 pages, 8 figures

With the rapid development of information technology, the number of devices in the Internet of Things (IoT) is increasing explosively, which makes device identification a great challenge. Deep neural networks (DNN) have been used for device identification in IoT due to their superior learning ability. However, DNNs are susceptible to adversarial attacks, which can greatly degrade the accuracy of deep learning (DL) models for device identification. Adversarial attack is one of the fundamental security concerns for DNNs, and it is of great importance to study the generation of adversarial examples and to examine the attack effects for the design of robust DNN-based device identification schemes. In this paper, we examine the effects of non-targeted and targeted adversarial attacks on CNN-based device identification and propose combined evaluation indicators of logits to enrich the evaluation criteria. Our experimental results demonstrate that the identification accuracy degrades with the increase of perturbation level and iteration step size, and the proposed combined evaluation indicators are effective to show the individual device signal differences. The insights from this study will be useful for the design of robust DL-based IoT systems.

This paper considers the robust optimal control problem for a class of nonlinear systems in the presence of unmodeled dynamics. The adaptive optimal controller design is achieved by the online actor-critic learning, which is robust against the unmodeled dynamics. To deal with the unmodeled dynamics, a dynamical signal with the system state as the input signal is designed to capture the input-to-state stability. In addition to the critic network for value function approximation, a novel robustifying term is developed and introduced into the actor network and applied to the closed-loop system in the presence of the dynamical uncertainty due to the unmodeled dynamics. It is shown that both the actor and critic weights learning converge towards the optimum while guaranteeing the boundedness of all the signals in the closed-loop. Simulation examples are conducted to verify the efficacy of the presented scheme in the end.

This paper considers wireless communication system consisted of one transmitter, one legitimate receiver, and one eavesdropper. The transmitter transmits perturbation-added signal (i.e. adversarial example) with a certain modulation type, while the legitimate receiver and the eavesdropper adopt deep neural networks (DNN)-based classifier to recognize the modulation type of the received signal. Compared to the fact that the general goal of adversarial examples being a misclassification of all available classifiers, our objective is to design an adversarial example that lets the legitimate receiver classify accurately while the eavesdropper misclassifies. To this end, we propose two design approaches of the adversarial examples: ( i ) A priori; ( i i ) A posteriori, i.e. before and after learning steps of the receiver, respectively. Numerical results show that both approaches are effective for securing the communication link.

Data driven method has been investigated in fault diagnosis. This kind of methods usually need of a large number of labeled data in order to obtain a good model. However, the unlabeled data dominates the data availability in the real industries, which impedes the successfully application of these methods. To remedy this situation, this research investigates a new semi-supervised fault diagnosis method, and it aims at training an adaptive model by using the small labeled data and large unlabeled data. Firstly, the fault signal is transformed to the RGB format. Then, Convolutional Neural Network (CNN) is trained to generate the fault features combining with the Correlation Alignment (CORAL) loss. Finally, the Transfer Component Analysis (TCA) is adopted to classify the fault features to improve the prediction accuracy. The proposed semi-supervised method is conducted on three bearing datasets. The experiments with different proportions of labeled data are done to validate the proposed CNN with CORAL and TCA. The comparisons between the proposed CNN model with other popular deep learning methods are also presented. The results show that CNN with CORAL and TCA is potential for fault diagnosis, and the prediction accuracies of proposed method outperform other comparisons methods.

Secure state estimation (SSE) is a problem to defense false-data injection attacks. This study designs an online distributed SSE of heterogeneous multiagent systems under homologous attack. A triple-loop observer is proposed to estimate the state and attack signal simultaneously. The inner loop keeps the estimations of the attack signal the same using average consensus. The middle loop adjusts the estimations via residual information. The outer loop runs when system measurements change. A sufficient condition that estimations asymptotically converge to the real value is obtained and proved. Finally, the proposed observer has been tested on the global positioning system.

This study investigates a parameterization of all output-rectifying retrofit controllers for distributed design of a structured controller. It has been discovered that all retrofit controllers can be characterized as a constrained Youla parameterization, which is difficult to solve analytically. For synthesis, a tractable and insightful class of retrofit controllers, referred to as output-rectifying retrofit controllers, has been introduced. An unconstrained parameterization of all output-rectifying retrofit controllers can be derived under a technical assumption on measurability of particular signals. The aim of this note is to reveal the structure of all output-rectifying retrofit controllers in the general output-feedback case. It is found out that the existing developments can be generalized based on the notions of state projection and an inverse system. The result leads to the conclusion that output-rectifying retrofit controllers can readily be designed even in the general case., to be published at IEEE Transactions on Automatic Control

Energy-efficient communication technologies are a key enabler for many IoT applications. Many existing communication protocols are based on duty-cycling techniques, that have an inherent tradeoff between delay and energy consumption. In the field of sensor networks, wake-up receivers have been investigated to overcome these problems and to further reduce energy consumption. We now go one step further and investigate the use of wake-up receivers in combination with IEEE 802.11 WLAN. We extend the protocol used to communicate between the access point and the client to introduce a wake-up signal. This can be implemented in a way that is fully compatible with existing WLAN standards, thus, it can be deployed gradually with little effort and no need to change existing systems. As a proof of concept and to perform first lab experiments, we developed a hardware prototype using a selective wake-up receiver and off-the-shelf USB-WLAN dongles. All experimental results are verified using an analytical model and a detailed simulation study. We show that our wake-up WLAN can provide connectivity for low-power devices with low delays and low energy consumption at the same time.

We study the inspection scheduling decisions for a production process that goes through a hidden defective state before its failure. The production process is equipped with a predictive model, generating alert and no-alert signals. An alert signal indicates that production process is in the defective state, while a no-alert signal indicates it is in the healthy state. The signals are imperfect, meaning that an alert signal can be generated for a healthy process and a no-alert signal can be generated for a defective process. Only a costly inspection can detect the true condition. We introduce a new inspection policy, which generalizes the age-based inspection policy that performs planned inspections at predetermined intervals, by considering that an inspection can also be triggered by a certain number of alerts from the predictive model. To optimize the proposed inspection policy, a stochastic dynamic programming model is formulated with the objective of minimizing the long-run expected cost rate. The performance improvement achieved by the optimal policy is quantified by comparing it to practically relevant benchmark policies. Numerical experiments with a set of realistic problem instances show that adding alert-triggered inspections to traditional age-based inspection scheduling brings up to 44% reduction in the expected cost rate when the predictive model is sufficiently accurate. Characterizing the performance of the optimal policy at a given level of imperfectness is especially useful in practice as it allows making an assessment on how much can be invested to justify a certain level of improvement in the predictive model.

Online continuous monitoring of dissolved gasses in transformer oil has become an emerging trend in the recent years. This way, in addition to sending the warning signal in the case of gas concentration, a fault diagnosis procedure can be conducted using standard Dissolved Gas Analysis (DGA) methods. However, since the fault diagnosis outcome and result of these methods are sometimes different, an appropriate solution should be suggested to accurately describe the transformer internal condition. To do so, a robust multi-layer framework has been proposed in this paper. Handling the measurement uncertainties and fusing the results of independent DGA methods without losing their diagnosis resolutions are the interesting features of this framework. To improve the overall accuracy of fused diagnosis result, an artificial neural network-based approach is proposed. It intelligently assigns the weight of each independent method in the fusion procedure according to its fault type detection accuracy for the range of input gasses amounts. To evaluate performance of the proposed framework, comprehensive test studies have been conducted and the promising obtained results prove its superior performance for use in online applications.

Online matching platforms that lack common informational mechanisms such as ratings and reviews often resort to mandatory verification of their users to enhance credibility. However, such mandatory verification may suppress the transmission of useful information for individual users. This study focuses on an alternate role of verification - its ability to serve as a credible signal for a user, when such verification is made optional and visible to other users. In collaboration with a leading online dating platform with no reputation mechanisms and where most of the information is self-disclosed, we design and conduct a randomized field experiment to examine not only who chooses to verify, but also the effectiveness of such optional verification for different types of users. Interestingly, we find users on the two sides use the same signal very differently. Males act consistent with the conventional prediction of signaling with High type males being more likely to opt-in to verification. As for females, we find Medium type females are the most likely to opt-in to verification as compared to High type females. We also find that such differential opt-in decisions are related to the differences in the credibility of the existing key attribute of each side, viz. income for males and beauty for females. In examining the outcomes of verification, we find that verified users receive more contacts from higher-type users, with the High type males and Medium type females benefiting the most. More interestingly, we find that verified users become more proactive and reach out to more and better potential partners. Further, the introduction of this voluntary verification facilitates desirable matching outcomes and benefits the platform as a whole. These findings have useful implications for research as well as practice.

The brain must extract behaviorally relevant latent variables from the signals streamed by the sensory organs. Such latent variables are often encoded in the dynamics that generated the signal rather than in the specific realization of the waveform. Therefore, one problem faced by the brain is to segment time series based on underlying dynamics. We present two algorithms for performing this segmentation task that are biologically plausible, which we define as acting in a streaming setting and all learning rules being local. One algorithm is model-based and can be derived from an optimization problem involving a mixture of autoregressive processes. This algorithm relies on feedback in the form of a prediction error, and can also be used for forecasting future samples. In some brain regions, such as the retina, the feedback connections necessary to use the prediction error for learning are absent. For this case, we propose a second, model-free algorithm that uses a running estimate of the autocorrelation structure of the signal to perform the segmentation. We show that both algorithms do well when tasked with segmenting signals drawn from autoregressive models with piecewise-constant parameters. In particular, the segmentation accuracy is similar to that obtained from oracle-like methods in which the ground-truth parameters of the autoregressive models are known. We also test our methods on datasets generated by alternating snippets of voice recordings. We provide implementations of our algorithms at https://github.com/ttesileanu/bio-time-series., Comment: v2.1; 34 pages, 14 figures

The finite-time output regulation (FTOR) problem for linear heterogeneous multi-agent systems (MAS) is investigated in this paper. The objective of this paper is to design controllers such that each agent can accurately track the reference signal and reject the disturbance. The disturbance and reference signal are generated by an exosystem. It is assumed that only a part of agents can obtain the state and system matrix of the exosystem. Then, distributed fixed-time observers are firstly proposed to estimate them, which can ensure the fixed-time convergence of the estimation errors. Next, we design both state feedback controller and output feedback controller to solve the FTOR problem using feedforward method. By virtue of Lyapunov theory and output regulation technique, it is shown that the proposed control strategy can solve the FTOR problem. Finally, a numerical example is included for illustration.

Error-related potential (ErrP) are a particular type of Event-Related Potential (ERP) elicited by a person attending a recognizable error. These Electroencephalographic (EEG) signals can be used to train a gaming agent by a Reinforcement Learning (RL) algorithm to learn an optimal policy. The experimental process consists of an observational human critic (OHC) observing a simple game scenario while their brain signals are captured. The game consists of a grid, where a green spot has to reach a desired target in the fewest amount of steps. Results show that there is an effective transfer of information and that the agent successfully learns to solve the game efficiently, from the initial 97 steps on average required to reach the target to the optimal number of 8 steps. Our results are expressed in threefold: (i) the mechanics of a simple grid-based game that can elicit the ErrP signal component, (ii) the verification that the reward function only penalizes wrong steps, which means that type II error (not properly identifying a wrong movement) does not affect significantly the agent learning process; (iii) collaborative rewards from multiple observational human critics can be used to train the algorithm effectively and can compensate low classification accuracies and a limited scope of transfer learning schemes.

Prevailing theory argues that more certifications increase performance. However, emerging empirical evidence implies that obtaining more certifications may actually decrease performance. How do we reconcile this tension? Practically speaking, why would ventures seek additional certifications in light of these recently identified risks? To address this gap between existing theory and recent empirics, we look more closely at ventures’ activities and performance outcomes after they receive their first certification. We posit that different patterns of certification reflect different forms of experimentation. In particular, ventures may be willing to experiment in ways that incur an inappropriateness penalty for the chance to gain a subsequent desirability premium if their experiments succeed. Inappropriateness means that certifications signal divergence from accepted market norms and standards. Desirability means that certifications signal activities that are in the perceived self-interest of the potential audience. We hypothesize that certifications reflecting broad experimentation incur initial inappropriateness penalties, yet when successful, they are more likely to lead to breakthroughs that generate desirability premia. We find support for this idea through an empirical analysis drawing from a sample of 7,440 U.S. ventures that receive one or more Small Business Innovation Research (SBIR) or Small Business Technology Transfer (STTR) grants to commercialize new technologies. This study advances institutional theory of certification to better account not only for its benefits but also for its costs.

The chemical production process is a special dynamic and complex system. It has the characteristics of instability and danger, thus making safety management in the production process very difficult. To support the long-term and stable operation of the chemical process, timely and accurate fault diagnosis is very necessary. Aiming at the high-dimensional nonlinearity of chemical process data, this paper proposes a fault diagnosis method in chemical process that is based on multi-scale convolutional neural network (MsCNN) combined with matrix diagram. This model uses software to convert the pre-processed time series signal data of the chemical process into sets of matrix diagrams, and then uses the MsCNN model to accurately diagnose various types of faults in the chemical production process. Taking Tennessee Eastman (TE) process fault data set as an example, verification shows good simulation results in detection accuracy and loss. This method can create high-precision judgments on fault types, which is beneficial to timely elimination of faults and avoidance of safety accidents. The results show that the new model proposed in this paper has good potential in the field of fault diagnosis in chemical processes, and is also reliable, practical and scientific.

This paper studies the cooperative output regulation problem for heterogeneous multi-agent systems with actuator faults. A new two-layer distributed control strategy is proposed: (i) Two types of distributed observers are designed for agents to estimate the exogenous signal, and an adaptive event-triggering mechanism is introduced to reduce unnecessary information transmission between agents; (ii) A decentralized adaptive fault-tolerant control scheme is proposed to achieve the cooperative output regulation of heterogeneous multi-agent systems and compensate for the actuator faults automatically. Finally, a numerical example is given to verify the feasibility of the proposed algorithm.