Search

Indoor human action recognition, essential across various applications, faces significant challenges such as orientation constraints and identification limitations, particularly in systems reliant on non-contact devices. Self-occlusions and non-line of sight (NLOS) situations are important representatives among them. To address these challenges, this paper presents a novel system utilizing dual Kinect V2, enhanced by an advanced Transmission Control Protocol (TCP) and sophisticated ensemble learning techniques, tailor-made to handle self-occlusions and NLOS situations. Our main works are as follows: (1) a data-adaptive adjustment mechanism, anchored on localization outcomes, to mitigate self-occlusion in dynamic orientations; (2) the adoption of sophisticated ensemble learning techniques, including a Chirp acoustic signal identification method, based on an optimized fuzzy c-means-AdaBoost algorithm, for improving positioning accuracy in NLOS contexts; and (3) an amalgamation of the Random Forest model and bat algorithm, providing innovative action identification strategies for intricate scenarios. We conduct extensive experiments, and our results show that the proposed system augments human action recognition precision by a substantial 30.25%, surpassing the benchmarks set by current state-of-the-art works. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) approaches in cancer analysis typically utilize a 'one-size-fits-all' methodology characterizing average patient responses. This manner neglects the diverse conditions in the pancancer and cancer subtypes of individual patients, resulting in suboptimal outcomes in diagnosis and treatment. To overcome this limitation, we shift from a blanket application of statistics to a focus on the explicit recognition of patient-specific abnormalities. Our objective is to use multiomics data to empower clinicians with personalized molecular descriptions that allow for customized diagnosis and interventions. Here, we propose a highly trustworthy multiomics learning (HTML) framework that employs multiomics self-adaptive dynamic learning to process each sample with data-dependent architectures and computational flows, ensuring personalized and trustworthy patient-centering of cancer diagnosis and prognosis. Extensive testing on a 33-type pancancer dataset and 12 cancer subtype datasets underscored the superior performance of HTML compared with static-architecture-based methods. Our findings also highlighting the potential of HTML in elucidating complex biological pathogenesis and paving the way for improved patient-specific care in cancer treatment. [ABSTRACT FROM AUTHOR]

The relevant problem of analyzing the features of measurement systems and their metrological support is formulated. Promising areas for the development of measurement systems, including virtual systems, are considered. The article analyzes such development areas as remote and synchronized vector measurements, cloud technologies, the Internet of Things, big data, and artificial intelligence. Several innovative solutions employing both biosensors and hybrid metrology are described (introduction of structural redundancy into measurement systems, as well as the introduction of temporal and algorithmic redundancies into the software of measurement systems). It is shown that the integration of component functions objectively complicates the identification of measurement systems in complex technical systems, yet providing new possibilities for technical systems with measurement functions. The following areas for improving the metrological support of measurement systems are analyzed: self-diagnostics and self-monitoring; remote and automated calibration and verification; use of digital twins, big data, and artificial intelligence; establishment of adaptive calibration and verification intervals. The study reveals the need to minimize bureaucratic procedures and automate metrological procedures while showing practicability in switching from periodic to adaptive procedures requiring no human involvement. [ABSTRACT FROM AUTHOR]

To meet the high dynamic voltage or current range measuring in real-time in the modern electrical industry, ranging from the surface science to non-destructive testing, this paper reports two broad dynamic ranging, universal, and compact digital lock-in amplifier methods for the huge dynamic range signal, termed as the hybrid and all-digital amplifiers. Both have reduced the complex components required in the traditional amplifiers to only two or three components without sacrificing the measuring accuracy, even by less than 0.05% in some situations, which has been evaluated via simulations and experiments with the FPGA circuit. Additionally, benefiting from the single-power supply strategy, the proposed methods are suitable for portable devices, including the pocket spectrometer, mechanical resonator monitor, and powered on battery. Such results in this paper illustrate the phase coherent technology with a compact, universal, and integrated circuit with a promising future. [ABSTRACT FROM AUTHOR]

In this paper, we propose two inertial algorithms with a new self-adaptive step size for approximating a solution of the split common null-point problem in the framework of Banach spaces. The step sizes are adaptively updated over each iteration by a simple process without the prior knowledge of the operator norm of the bounded linear operator. Under suitable conditions, we prove the weak-convergence results for the proposed algorithms in p-uniformly convex and uniformly smooth Banach spaces. Finally, we give several numerical results in both finite- and infinite-dimensional spaces to illustrate the efficiency and advantage of the proposed methods over some existing methods. Also, data classifications of heart diseases and diabetes mellitus are presented as the applications of our methods. [ABSTRACT FROM AUTHOR]

In the field of smart manufacturing, the data-driven scheduling approach has become an effective way to solve the smart shop floor scheduling problem with high complexity and dynamics. However, most existing approaches rely too heavily on manual supervision in implementation, resulting in poor adaptability and effectiveness in dynamic production environments. Therefore, this paper proposes a self-adaptive scheduling approach based on double deep Q-network (DDQN), which can reduce manual supervision and realize the autonomy of the whole scheduling process. In the presented approach, first, a self-adaptive scheduling framework, which forms a closed-loop optimization structure for scheduling model evaluation, generation/updating, and application, is designed. Second, the interactive learning mechanism of reinforcement learning is introduced, and the scheduling model is generated through the DDQN algorithm without manual supervision. In addition, dynamic reward function based on simulation is proposed to promote the rationality and accuracy of the reward in reinforcement learning. The effectiveness of the proposed approach is validated on a semiconductor production shop floor, and the experimental results illustrate that the proposed approach can improve the effectiveness of self-adaptive scheduling and significantly reduce the time and labour costs in the dynamic production environments. [ABSTRACT FROM AUTHOR]

The development of the future's smart cities will increasingly depend on autonomous vehicles. For example, object detection, tracking, path planning, and sentiment or intent recognition, among other things, have all been the subject of several suggestions in recent years in an effort to address specific components of the working pipeline and develop a useful end-to-end system. This article presents a simple benchmark to evaluate the performance of object identification algorithms under deteriorating image quality. In this article, the input image is taken from the car object detection dataset, then, it gets preprocessed. The preprocessing involved Histogram equalization-based contrast enhancement, geometric transformations, and image augmentation. After getting preprocessed the region of interest is segmented with the help of a mask convolutional recurrent neural network (MCRNN). Then, the segmented region is classified using the long short-term memory (LSTM) with the self-improved chimp optimization algorithm. Object detection is performed through the proposed model called MCRNN-SICLSTM. The implementation is performed using MATLAB software. The performance of the proposed model is compared with the existing techniques using the performance metrics accuracy, precision, recall, F-Measure, mean absolute error, mean square error, false positive rate, false negative rate. The accuracy achieved is 99.23%. [ABSTRACT FROM AUTHOR]

Large-sized rotating machines usually contain weak feature information of rub-impact fault, which is hard to extract. General scale transformation stochastic resonance (GSTSR) can match input signals with different frequencies by using the optimal barrier height and boost the weak fault feature in signals. The performance of GSTSR is determined by systemic parameters. When a rub-impact fault occurs between rotor and stator, vibration signals are often accompanied by an impact. Therefore, the paper takes advantage of sensibility of waveform factor to rub-impact fault information and margin factor to impact properties of signal and reconstructs a new signal evaluation index based on waveform and margin factors. The signal evaluation index is treated as the fitness function of grey wolf optimization (GWO) algorithm and combined with GSTSR to perform a comprehensive evaluation to rub-impact fault feature information. The result of comparison with conventional method (with signal to noise ratio (SNR) as fitness function) indicates that in case of extracting rub-impact fault features, the proposed method identifies a rotor-stator rub-impact fault more precisely than the classical method. [ABSTRACT FROM AUTHOR]

At present, the research on retaining walls with relief shelves mainly focuses on translational displacement mode, and the research on RB mode is less. Using adaptive software, it is possible to obtain the grid distribution diagram and shear dissipation diagram of the filling soil by simulating the retaining wall's rotation around its heel. The shear dissipation diagram was examined to determine the distribution of sliding surfaces in the active limit condition. The plastic region was meshed using the slip line technique. The slip line mesh distribution and the finite-element mesh distribution were compared to ensure that the two mesh distributions were consistent. Simultaneously, with the effect of retaining wall geometry, infill soil quality, and other elements on the grid distribution and earth pressure being investigated, by contrasting the computed normal earth pressure (σn) distribution with the finite-element findings, the correctness of the approach was demonstrated. It was discovered that increasing the relief shelf position and wall heel length will reduce the retaining wall's earth pressure, while increasing the relief shelf length will have the reverse effect. In the active limit condition, the retaining wall earth pressure decreased with increasing filling strength, but changes in filling quality had little impact on grid distribution. [ABSTRACT FROM AUTHOR]

The Industrial Revolution drives the digitization of society and industry, entailing Cyber-Physical Systems (CPSs) that form ecosystems where system owners and third parties share responsibilities within and across industry domains. Such ecosystems demand smart CPSs that continuously align their architecture and governance to the concerns of various stakeholders, including developers, operators, and users. In order to satisfy short- and long-term stakeholder concerns in a continuously evolving operational context, this work proposes self-adaptive software models that promote DevOps for smart CPS. Our architectural approach extends to the embedded system layer and utilizes embedded and interconnected Digital Twins to manage change effectively. Experiments conducted on industrial embedded control units demonstrate the approach's effectiveness in achieving sub-millisecond real-time closed-loop control of CPS assets and the simultaneous high-fidelity twinning (i.e., monitoring) of asset states. In addition, the experiments show practical support for the adaptation and evolution of CPS through the dynamic reconfiguring and updating of real-time control services and communication links without downtime. The evaluation results conclude that, in particular, the embedded Digital Twins can enhance CPS smartness by providing service-oriented access to CPS data, monitoring, adaptation, and control capabilities. Furthermore, the embedded Digital Twins can facilitate the seamless integration of these capabilities into current and future industrial service ecosystems. At the same time, these capabilities contribute to implementing emerging industrial services such as remote asset monitoring, commissioning, and maintenance. [ABSTRACT FROM AUTHOR]

Efficient recognition of emotions has attracted extensive research interest, which makes new applications in many fields possible, such as human-computer interaction, disease diagnosis, service robots, and so forth. Although existing work on sentiment analysisrelying on sensors or unimodal methods performs well forsimple contexts like business recommendation and facial expression recognition, it does far below expectations for complex scenes, such as sarcasm, disdain, and metaphors. In this article, we propose a novel two-stage multimodal learning framework, called AMSA, to adaptively learn correlation and complementarity between modalities for dynamic fusion, achieving more stable and precise sentiment analysis results. Specifically, a multiscale attention model with a slice positioning scheme is proposed to get stable quintuplets of sentiment in images, texts, and speeches in the first stage. Then a Transformer-based self-adaptive network is proposed to assign weights flexibly for multimodal fusion in the second stage and update the parameters of the loss function through compensation iteration. To quickly locate key areas for efficient affective computing, a patch-based selection scheme is proposed to iteratively remove redundant information through a novel loss function before fusion. Extensive experiments have been conducted on both machine weakly labeled and manually annotated datasets of self-made Video-SA, CMU-MOSEI, and CMU-MOSI. The results demonstrate the superiority of our approach through comparison with baselines. [ABSTRACT FROM AUTHOR]

This paper deals with self-adaptive real-time embedded systems (RTES). A self-adaptive system can operate in different modes. Each mode encodes a set of real-time tasks. To be executed, each task is allocated to a processor (placement) and assigned a priority (scheduling), while respecting timing constraints. An adaptation scenario allows switching between modes by adding, removing, and updating task parameters that must meet related deadlines after adaptation. For such systems, anticipating all operational modes at design time is usually impossible. Online reinforcement learning is increasingly used in the presence of design-time uncertainty. To tackle this problem, we formalize the placement and scheduling problems in self-adaptive RTES as a Markov decision process and propose related algorithms based on Q-learning. Then, we introduce an approach that integrates the proposed algorithms to assist designers in the development of self-adaptive RTES. At the design level, the RL Placement and the RL Scheduler are proposed to process predictable adaptation scenarios. These modules are designed to generate placement and scheduling models for an application while maximizing system extensibility and ensuring real-time feasibility. At the execution level, the RL Adapter is defined to process online adaptations. Indeed, the goal of the RL Adapter agent is to reject the adaptation scenario when feasibility concerns are raised; otherwise, it generates a new feasible placement and scheduling. We apply and simulate the proposed approach to a healthcare robot case study to show its applicability. Performance evaluations are conducted to prove the effectiveness of the proposed approach compared to related works. [ABSTRACT FROM AUTHOR]

Background: Walking impairments are a common consequence of neurological disorders and are assessed with clinical scores that suffer from several limitations. Robot-assisted locomotor training is becoming an established clinical practice. Besides training, these devices could be used for assessing walking ability in a controlled environment. Here, we propose an adaptive assist-as-needed (AAN) control for a treadmill-based robotic exoskeleton, the Lokomat, that reduces the support of the device (body weight support and impedance of the robotic joints) based on the ability of the patient to follow a gait pattern displayed on screen. We hypothesize that the converged values of robotic support provide valid and reliable information about individuals' walking ability. Methods: Fifteen participants with spinal cord injury and twelve controls used the AAN software in the Lokomat twice within a week and were assessed using clinical scores (10MWT, TUG). We used a regression method to identify the robotic measure that could provide the most relevant information about walking ability and determined the test–retest reliability. We also checked whether this result could be extrapolated to non-ambulatory and to unimpaired subjects. Results: The AAN controller could be used in patients with different injury severity levels. A linear model based on one variable (robotic knee stiffness at terminal swing) could explain 74% of the variance in the 10MWT and 61% in the TUG in ambulatory patients and showed good relative reliability but poor absolute reliability. Adding the variable 'maximum hip flexor torque' to the model increased the explained variance above 85%. This did not extend to non-ambulatory nor to able-bodied individuals, where variables related to stance phase and to push-off phase seem more relevant. Conclusions: The novel AAN software for the Lokomat can be used to quantify the support required by a patient while performing robotic gait training. The adaptive software might enable more challenging training conditions tuned to the ability of the individuals. While the current implementation is not ready for assessment in clinical practice, we could demonstrate that this approach is safe, and it could be integrated as assist-as-needed training, rather than as assessment. Trial registration: ClinicalTrials.gov Identifier: NCT02425332. [ABSTRACT FROM AUTHOR]

As a newly developed metaheuristic algorithm, the artificial bee colony (ABC) has garnered a lot of interest because of its strong exploration ability and easy implementation. However, its exploitation ability is poor and dramatically deteriorates for high-dimension and/or non-separable functions. To fix this defect, a self-adaptive ABC with a candidate strategy pool (SAABC-CS) is proposed. First, several search strategies with different features are assembled in the strategy pool. The top 10% of the bees make up the elite bee group. Then, we choose an appropriate strategy and implement this strategy for the present population according to the success rate learning information. Finally, we simultaneously implement some improved neighborhood search strategies in the scout bee phase. A total of 22 basic benchmark functions and the CEC2013 set of tests were employed to prove the usefulness of SAABC-CS. The impact of combining the five methods and the self-adaptive mechanism inside the SAABC-CS framework was examined in an experiment with 22 fundamental benchmark problems. In the CEC2013 set of tests, the comparison of SAABC-CS with a number of state-of-the-art algorithms showed that SAABC-CS outperformed these widely-used algorithms. Moreover, despite the increasing dimensions of CEC2013, SAABC-CS was robust and offered a higher solution quality. [ABSTRACT FROM AUTHOR]

To meet increasingly restrictive requirements and improve quality of service (QoS), Internet of Things (IoT) systems have embraced multi-layered architectures leveraging edge and fog computing. However, the dynamic and changing IoT environment can impact QoS due to unexpected events. Therefore, proactive evolution and adaptation of the IoT system becomes a necessity and concern. In this paper, we present a model-based approach for the specification and execution of self-adaptive multi-layered IoT systems. Our proposal comprises the design of a domain-specific language (DSL) for the specification of such architectures, and a runtime framework to support the system behaviuor and its self-adaptation at runtime. The code for the deployment of the IoT system and the execution of the runtime framework is automatically produced by our prototype code generator. Moreover, we also show and validate the extensibility of such DSL by applying it to the domain of underground mining. The complete infrastructure (modeling tool, generator and runtime components) is available in a online open source repository. [ABSTRACT FROM AUTHOR]

Filtering from airborne LiDAR datasets in urban area is one important process during the building of digital and smart cities. However, the existing filters encounter poor filtering performance and heavy computational burden when processing large-scale and complicated urban environments. To tackle this issue, a self-adaptive filtering method based on object primitive global energy minimization is proposed in this paper. In this paper, mode points were first acquired for generating the mode graph. The mode points were the cluster centers of the LiDAR data obtained in a mean shift algorithm. The graph constructed with mode points was named "mode graph" in this paper. By defining the energy function based on the mode graph, the filtering process is transformed to iterative global energy minimization. In each iteration, the graph cuts technique was adopted to achieve global energy minimization. Meanwhile, the probability of each point belonging to the ground was updated, which would lead to a new refined ground surface using the points whose probabilities were greater than 0.5. This process was iterated until two successive fitted ground surfaces were determined to be close enough. Four urban samples with different urban environments were adopted for verifying the effectiveness of the filter developed in this paper. Experimental results indicate that the developed filter obtained the best filtering performance. Both the total error and the Kappa coefficient are superior to those of the other three classical filtering methods. [ABSTRACT FROM AUTHOR]

Graph-based clustering methods offer competitive performance in dealing with complex and nonlinear data patterns. The outstanding characteristic of such methods is the capability to mine the internal topological structure of a dataset. However, most graph-based clustering algorithms are vulnerable to parameters. In this paper, we propose a self-adaptive graph-based clustering method (SAGC) with noise identification based on directed natural neighbor graph to auto identify the desired number of clusters and simultaneously obtain reliable clustering results without prior knowledge and parameter setting. This method adopts parameter adaptive process to deal with specific data patterns and can identify clusters with diverse shapes and detect noises. We use synthetic and UCI real-world datasets to prove the validity of the innovatory method by comparing it to k-means, DBSCAN, OPTICS, AP, SC, CutPC, and WC algorithms in terms of clustering Accuracy, Adjusted Rand index, Normalized Mutual Information and Fowlkes–Mallows index. The experimental results confirm that the proposed method contributes to the progress of graph-based clustering algorithms. [ABSTRACT FROM AUTHOR]

Earthquake forecasting aims to determine the likelihood of a damaging earthquake occurring in a particular area within a period of days to months. This provides ample preparation time for potential seismic hazards, resulting in significant socioeconomic benefits. Surface and atmospheric parameters derived from satellite thermal infrared observations have been utilized to identify pre-earthquake anomalies that may serve as potential precursors for earthquake forecasting. However, the correlation between these anomalies and impending earthquakes remains a significant challenge due to high false alarm and missed detection rates. To address this issue, we propose a spatially self-adaptive multiparametric anomaly identification scheme based on global strong earthquakes to establish the optimal recognition criteria. Each optimal parameter exhibits significant spatial variability within the seismically active region and indicates transient and subtle anomaly signals with a limited frequency of occurrences (<10 for most regions). In comparison to the fixed criterion for identifying anomalies, this new scheme significantly improves the positive Matthew's correlation coefficient (MCC) values from ~0.03 to 0.122–0.152. Additionally, we have developed a multi-parameter anomaly synthesis method based on the best MCC value of each parameter anomaly. On average, the MCC increased from 0.143 to 0.186, and there are now more earthquake-prone regions with MCC values > 0.5. Our research emphasizes the critical importance of a multiparametric system in earthquake forecasting, where each geophysical parameter can be assigned a distinct weight, and the findings specifically identify OLR, including all-sky and clear-sky ones, as the most influential parameter on a global scale, highlighting the potential significance of OLR anomalies for seismic forecasting. Encouraging results imply the effectiveness of utilizing multiparametric anomalies and provide some confidence in advancing our knowledge of operational earthquake forecasting with a more quantitative approach. [ABSTRACT FROM AUTHOR]

One of the most common symptoms among diverse patients (e.g. Parkinson's disease, essential tremor) is hand tremor. Deep brain stimulation (DBS) has been a popular method of controlling hand tremor in recent years. Although recent advancements in the DBS system have enabled DBS applications in the context of various neurodegenerative disorders, the deployment of such interface tactics in DBS applications is seriously threatened by un-adaptability concerns and a lack of an accurate model. In response to these issues, the current letter discusses the construction of an intelligent technique for minimizing hand tremor and self-adaptive DBS settings without requiring a proper hand tremor system modeling setup. For this purpose, the hand tremor is suppressed using an intelligent proportional–integral (iPI) as the main controller and the sliding mode (SM) as an observer. Furthermore, as an auxiliary controller, a deep reinforcement learning (DRL) method with the Actor-Critic structure is adaptively incorporated. The feedback parameters of the ultra-local model (ULM) controller will be adaptively modified utilizing the DRL method by online neural network (NN) updating in the proposed strategy. The advantages of the ULM based on DRL structure over state-of-the-art approaches are demonstrated by a complete assessment. [ABSTRACT FROM AUTHOR]

To improve the accuracy of material identification under low contrast conditions, this paper proposes an improved YOLOv4-tiny target detection method based on an adaptive self-order piecewise enhancement and multiscale feature optimization. The model first constructs an adaptive self-rank piecewise enhancement algorithm to enhance low-contrast images and then considers the fast detection ability of the YOLOv4-tiny network. To make the detection network have a higher accuracy, this paper adds an SE channel attention mechanism and an SPP module to this lightweight backbone network to increase the receptive field of the model and enrich the expression ability of the feature map. The network can pay more attention to salient information, suppress edge information, and effectively improve the training accuracy of the model. At the same time, to better fuse the features of different scales, the FPN multiscale feature fusion structure is redesigned to strengthen the fusion of semantic information at all levels of the network, enhance the ability of network feature extraction, and improve the overall detection accuracy of the model. The experimental results show that compared with the mainstream network framework, the improved YOLOv4-tiny network in this paper effectively improves the running speed and target detection accuracy of the model, and its mAP index reaches 98.85%, achieving better detection results. [ABSTRACT FROM AUTHOR]

The landing gear structure suffers from large loads during aircraft takeoff and landing, and an accurate prediction of landing gear performance is beneficial to ensure flight safety. Nevertheless, the landing gear performance prediction method based on machine learning has a strong reliance on the dataset, in which the feature dimension and data distribution will have a great impact on the prediction accuracy. To address these issues, a novel MCA-MLPSA is developed. First, an MCA (multiple correlation analysis) method is proposed to select key features. Second, a heterogeneous multilearner integration framework is proposed, which makes use of different base learners. Third, an MLPSA (multilayer perceptron with self-attention) model is proposed to adaptively capture the data distribution and adjust the weights of each base learner. Finally, the excellent prediction performance of the proposed MCA-MLPSA is validated by a series of experiments on the landing gear data. [ABSTRACT FROM AUTHOR]

The UASNs are widely used in underwater communications and monitoring, and many applications require accurate information regarding the position of nodes. However, intentional attacks against devices or information transmission may exist in the network, and the localization process is periodic, so it is necessary to quickly address attacks and optimize the network structure. This paper proposed an anchor node self-adaptive adjustment localization scheme (ASAL), in which the anchor node can adjust the state and depth of its participation. Two filters were used to adjust the states of referable nodes. The first filter was based on the distance difference of reverse information transmission after direct localization based on anchor nodes. The second was based on the error of the anchor node's reverse localization after network localization was completed. In addition, a depth-adjustment mechanism of anchor nodes was proposed to optimize the network structure, the virtual force vector was introduced to describe the cost of depth adjustment, and the whale optimization algorithm was used to converge to the depth with the minimum total cost. The simulation results showed that the scheme can ensure localization accuracy and coverage in attack scenarios and reduce localization energy consumption. [ABSTRACT FROM AUTHOR]

The explosive growth of videos on the Internet makes it a great challenge to use texts to retrieve the videos we need. The general method of text-video retrieval is to project them into a common semantic space to calculate the similarity score. The key technologies of a retrieval model are how to get strong feature representations of text and video and bridge the semantic gap between the two modalities. Moreover, most existing methods do not consider the strong consistency of text-video positive sample pairs. Considering the above problems, we proposed a text-video retrieval method based on enhanced self-attention and multi-task learning in this paper. Firstly, while encoding, the extracted text feature vectors and the extracted video feature vectors are input into Transformer based on enhanced self-attention mechanism for encoding and fusion. Then the text representations and video representations are projected into a common semantic space. Finally, by introducing multi-task learning in the common semantic space, our proposed approach combines the semantic similarity measurement task and the semantic consistency judgement task to optimize the common space through semantic consistency constraints. Our method obtains better retrieval performance on the MSR-Video to Text (MSRVTT), Large Scale Movie Description Challenge (LSMDC), and ActivityNet datasets than some existing approaches, which proves the effectiveness of our proposed strategies. [ABSTRACT FROM AUTHOR]

Copyright of Telecommunication Engineering is the property of Telecommunication Engineering and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Digital low-dropout (DLDO) is widely used for power management in the system-on-chip (SoC) because of its low-voltage operation and process scalability. However, conventional DLDOs suffer from the trade-off between transient response and power consumption of the DLDO and the clock generator. This paper proposes an event-driven self-clocked DLDO regulator. The proposed low quiescent current (IQ) event-driven adaptive frequency clock generator (EACG) adapts its frequency in different load conditions without a current sensor or complex compensation circuit for stable operation in the entire load range. The proposed DLDO does not need any external clocking signal and can maintain low output ripple and low power consumption in the steady-state. The clock-less transient detector (CLTD), consisting of two clock-independent transient detection paths, uses power more efficiently and improves the transient response significantly without sacrificing the power consumption. This work was fabricated in a 40 nm CMOS process with an 0.3 nF on-chip capacitor. The measurement results show that with the step load current between 1 mA and 60 mA, the proposed DLDO achieves a transient recovery time of 220 ns. The total IQ of the proposed DLDO is only 26 μA in steady-state, and it achieves stable operation in the entire load range. [ABSTRACT FROM AUTHOR]

Self-adaptive software can assess and modify its behavior when the assessment indicates that the program is not performing as intended or when improved functionality or performance is available. Since the mid-1960s, system adaptivity has been extensively researched, and during the last decade, many application areas and technologies involving self-adaptation have gained prominence. All of these efforts have in common the introduction of self-adaptability through software. Thus, it is essential to investigate systematic software engineering methods to create self-adaptive systems that may be used across different domains. The primary objective of this research is to summarize current advances in awareness requirements for adaptive strategies and their performance management based on an examination of state-of-the-art methods described in the literature. This paper reviews self-adaptive systems in the context of requirement awareness and summarizes the most common methodologies applied. At first glance, it examines the previous surveys and works about self-adaptive systems. Afterward, it classifies the current self-adaptive systems based on six criteria. Then, it presents performance management in the current adaptive systems and then evaluates the most common self-adaptive approaches. Lastly, the self-adaptive models are evaluated based on four concepts (requirements description, monitoring, relationship, dependency/impact, and tools). [ABSTRACT FROM AUTHOR]

Currently, the superconducting diode effect (SDE) is being actively discussed, due to its large application potential in superconducting electronics. In particular, superconducting hybrid structures, based on three-dimensional (3D) topological insulators, are among the best candidates, due to their having the strongest spin–orbit coupling (SOC). Most theoretical studies on the SDE focus either on a full numerical calculation, which is often rather complicated, or on the phenomenological approach. In the present paper, we compare the linearized and nonlinear microscopic approaches in the superconductor/ferromagnet/3D topological insulator (S/F/TI) hybrid structure. Employing the quasiclassical Green's function formalism we solve the problem self-consistently. We show that the results obtained by the linearized approximation are not qualitatively different from the nonlinear solution. The main distinction in the results between the two methods was quantitative, i.e., they yielded different supercurrent amplitudes. However, when calculating the so-called diode quality factor the quantitative difference is eliminated and both approaches result in good agreement. [ABSTRACT FROM AUTHOR]

Digital transformation is both an opportunity and a challenge. To take advantage of this opportunity for humans and the environment, the transformation process must be understood as a design process that affects almost all areas of life. In this paper, we investigate AI-Based Self-Adaptive Cyber-Physical Process Systems (AI-CPPS) as an extension of the traditional CPS view. As contribution, we present a framework that addresses challenges that arise from recent literature. The aim of the AI-CPPS framework is to enable an adaptive integration of IoT environments with higher-level process-oriented systems. In addition, the framework integrates humans as actors into the system, which is often neglected by recent related approaches. The framework consists of three layers, i.e., processes, semantic modeling, and systems and actors, and we describe for each layer challenges and solution outlines for application. We also address the requirement to enable the integration of new networked devices under the premise of a targeted process that is optimally designed for humans, while profitably integrating AI and IoT. It is expected that AI-CPPS can contribute significantly to increasing sustainability and quality of life and offer solutions to pressing problems such as environmental protection, mobility, or demographic change. Thus, it is all the more important that the systems themselves do not become a driver of resource consumption. [ABSTRACT FROM AUTHOR]

For sparse signal reconstruction (SSR) problem in compressive sensing (CS), by the splitting technique, we first transform it into a continuously differentiable convex optimization problem, and then a new self-adaptive gradient projection algorithm is proposed to solve the SSR problem, which has fast solving speed and pinpoint accuracy when the dimension increases. Global convergence of the proposed algorithm is established in detail. Without any assumptions, we establish global R − linear convergence rate of the proposed algorithm, which is a new result for constrained convex (rather than strictly convex) quadratic programming problem. Furthermore, we can also obtain an approximate optimal solution in a finite number of iterations. Some numerical experiments are made on the sparse signal recovery and image restoration to exhibit the efficiency of the proposed algorithm. Compared with the state-of-the-art algorithms in SSR problem, the proposed algorithm is more accurate and efficient. [ABSTRACT FROM AUTHOR]

Several real-world parallel applications are becoming more dynamic and long-running, demanding online (at run-time) adaptations. Stream processing is a representative scenario that computes data items arriving in real-time and where parallel executions are necessary. However, it is challenging for humans to monitor and manually self-optimize complex and long-running parallel executions continuously. Moreover, although high-level and structured parallel programming aims to facilitate parallelism, several issues still need to be addressed for improving the existing abstractions. In this paper, we extend self-adaptiveness for supporting autonomous and online changes of the parallel pattern compositions. Online self-adaptation is achieved with an online profiler that characterizes the applications, which is combined with a new self-adaptive strategy and a model for smooth transitions on reconfigurations. The solution provides a new abstraction layer that enables application programmers to define non-functional requirements instead of hand-tuning complex configurations. Hence, we contribute with additional abstractions and flexible self-adaptation for responsiveness at run-time. The proposed solution is evaluated with applications having different processing characteristics, workloads, and configurations. The results show that it is possible to provide additional abstractions, flexibility, and responsiveness while achieving performance comparable to the best static configuration executions. [ABSTRACT FROM AUTHOR]

Represented by reactive security defense mechanisms, cyber defense possesses a static, reactive, and deterministic nature, with overwhelmingly high costs to defend against ever-changing attackers. To change this situation, researchers have proposed moving target defense (MTD), which introduces the concept of an attack surface to define cyber defense in a brand-new manner, aiming to provide a dynamic, continuous, and proactive defense mechanism. With the increasing use of machine learning in networking, researchers have discovered that MTD techniques based on machine learning can provide omni-bearing defense capabilities and reduce defense costs at multiple levels. However, research in this area remains incomplete and fragmented, and significant progress is yet to be made in constructing a defense mechanism that is both robust and available. Therefore, we conducted a comprehensive survey on MTD research, summarizing the background, design mechanisms, and shortcomings of MTD, as well as relevant features of intelligent MTD that are designed to overcome these limitations. We aim to provide researchers seeking the future development of MTD with insight into building an intelligently affordable, optimized, and self-adaptive defense mechanism. [ABSTRACT FROM AUTHOR]

Copyright of Nonferrous Metals (Mineral Processing Section) is the property of Beijing Research Institute of Mining & Metallurgy Technology Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Attributed networks, such as social networks, citation networks, and traffic networks, are ubiquitous nowadays. Node classification is an essential analysis task on attributed networks that has attracted increasing research attention. However, the conventional node classification methods usually require a large number of labeled training instances, which is labor-consuming and cannot be applied to real-world attributed networks. Thus, various meta-learning methods have been proposed to tackle the problem of insufficient labeled instances. However, few-shot learning for graph-structured data is still under-explored due to two reasons: (1) In the few-shot scenario, equally treating each node makes the model vulnerable to noisy instances; (2) The adjacency matrix and attributed features suffer from the data sparsity and not all dimensions in the feature space are discriminative to support final classification. To solve the above problems, we propose a novel graph meta-learning framework: Multi-Level Self-Adaptive Prototypical Networks (MSPN). The proposed MSPN leverages multi-level adaptive learning to highlight the most expressive nodes and features and rectify the prototype to obtain a more accurate one. The episodic training strategy mimics the real testing scenario and transfers knowledge from previous tasks to unseen target tasks. Extensive experimental results demonstrate that the MSPN is superior to the other methods on few-shot node classification. [ABSTRACT FROM AUTHOR]

This paper proposes an energy-efficient hybrid Networks-on-Chip architecture called SaHNoC, which is more self adaptive with respect to selection of Wi-Fi positions in a Mesh topology. In comparison with wired on-chip interconnect NoC, wireless interconnect NoC offers a promising performance with respect to communication speed. However, the wireless switch occupies a larger area and has higher power-energy consumption than the wired switch. In this paper, we develop an optimal energy-efficient wire-wireless NoC structure called Hybrid NoC. The proposed hybrid NoC approach explores the self-adaptive technique for the selection of wireless switching positions in a suitable location against topology with respect to network diameter and average hop count. The effectiveness of the proposed SaHNoC is validated with a modified version of Booksim 2.0 in mutual with Orion 3.0. [ABSTRACT FROM AUTHOR]

Change detection (CD) methods using synthetic aperture radar (SAR) data have received significant attention in the field of remote sensing Earth observation, which mainly involves knowledge-driven and data-driven approaches. Knowledge-driven CD methods are based on the physical theoretical models with strong interpretability, but they lack the robust features of being deeply mined. In contrast, data-driven CD methods can extract deep features, but require abundant training samples, which are difficult to obtain for SAR data. To address these limitations, an end-to-end unsupervised CD network based on self-adaptive superpixel segmentation is proposed. Firstly, reliable training samples were selected using an unsupervised pre-task. Then, the superpixel generation and Siamese CD network were integrated into the unified framework to train them end-to-end until the global optimal parameters were obtained. Moreover, the backpropagation of the joint loss function promoted the adaptive adjustment of the superpixel. Finally, the binary change map was obtained. Several public SAR CD datasets were used to verify the effectiveness of the proposed method. The transfer learning experiment was implemented to further explore the ability to detect the changes and generalization performance of our network. The experimental results demonstrate that our proposed method achieved the most competitive results, outperforming seven other advanced deep-learning-based CD methods. Specifically, our method achieved the highest accuracy in OA, F1-score, and Kappa, and also showed superiority in suppressing speckle noise, refining change boundaries, and improving detection accuracy in a small area change. [ABSTRACT FROM AUTHOR]

Neurobiological systems continually interact with the surrounding environment to refine their behaviour toward the best possible reward. Achieving such learning by experience is one of the main challenges of artificial intelligence, but currently it is hindered by the lack of hardware capable of plastic adaptation. Here, we propose a bio-inspired recurrent neural network, mastered by a digital system on chip with resistive-switching synaptic arrays of memory devices, which exploits homeostatic Hebbian learning for improved efficiency. All the results are discussed experimentally and theoretically, proposing a conceptual framework for benchmarking the main outcomes in terms of accuracy and resilience. To test the proposed architecture for reinforcement learning tasks, we study the autonomous exploration of continually evolving environments and verify the results for the Mars rover navigation. We also show that, compared to conventional deep learning techniques, our in-memory hardware has the potential to achieve a significant boost in speed and power-saving. A big challenge for artificial intelligence is to gain the ability of learning by experience like biological systems. Here Bianchi et al. propose a hardware neural network based on resistive-switching synaptic arrays which dynamically adapt to the environment for autonomous exploration. [ABSTRACT FROM AUTHOR]

Physics-informed neural networks (PINNs) have been widely adopted to solve partial differential equations (PDEs), which could be used to simulate physical systems. However, the accuracy of PINNs does not meet the needs of the industry, and severely degrades, especially when the PDE solution has sharp transitions. In this paper, we propose a ResNet block-enhanced network architecture to better capture the transition. Meanwhile, a constrained self-adaptive PINN (cSPINN) scheme is developed to move PINN's objective to the areas of the physical domain, which are difficult to learn. To demonstrate the performance of our method, we present the results of numerical experiments on the Allen–Cahn equation, the Burgers equation, and the Helmholtz equation. We also show the results of solving the Poisson equation using cSPINNs on different geometries to show the strong geometric adaptivity of cSPINNs. Finally, we provide the performance of cSPINNs on a high-dimensional Poisson equation to further demonstrate the ability of our method. [ABSTRACT FROM AUTHOR]

The differential protection of transformers sometimes does not have reliable performance facing some faults and events such as saturation of current transformers (CTs) during internal and external faults, inrush currents, and cross‐country faults (CCFs). In this paper, a new transformer differential protection scheme based on the self‐adaptive biased characteristic curve and considering CCFs and various disturbances are used. First, the maximum overlap discrete wavelet transform (MODWT) was used to extract the superior characteristics of sampled differential current of the transformer at different levels. Then, by calculating the duty cycle (D) values of energy coefficients extracted in each periodicity by MATLAB software, a new self‐adaptive biased characteristic curve will be obtained in different operating conditions. Finally, the percentage differential protection operation is adjusted based on the new adaptive characteristic curve zones. Differential current sampling under different operating conditions was performed on a real three‐phase transformer in EMTP software. The simulation results show that the proposed adaptive differential protection scheme can accurately detect various faults and events occurring in the transformer and increase system reliability. [ABSTRACT FROM AUTHOR]

Copyright of Experimental Technology & Management is the property of Experimental Technology & Management Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Background. The article is devoted to the problem of predicting interactions between individual components in self-adaptive software systems. Two types of interactions are considered as possible unaccounted for interactions: the “inclusion” relation, in which the presence of a certain component in the configuration of the software product also requires the automatic inclusion of another in it; the “exclusion” relation, in which the presence of one component in the configuration excludes the presence of another. Since selfadaptation implies minimal participation of a professional developer in restructuring and building up new system functions, additional tools are needed to allow the system to also independently verify the changed structure. Since the mathematical apparatus of selfadaptation developed earlier by the authors of the article is based on the application of graph theory, the new proposed solution is an improved version of this apparatus using artificial intelligence methods. Materials and methods. As the main methods, artificial intelligence and neural network technologies are used, in particular, methods for predicting relationships in graph structures, machine learning methods. Results. The main results include a method for predicting possible interactions in a self-adaptive software system based on predicting relationships in graph structures. Conclusions. The proposed method is an improvement of the theoretical apparatus of self-adaptation of software systems developed earlier by the authors, based on graph formalization of the behavioral structure of the program. Since self-adaptation in this approach implies an automatic change in the structure, for its regulation, the use of artificial intelligence algorithms is envisaged, which makes it possible to evaluate new potential connections in terms of their usefulness or destructiveness for the system as a whole.

The synchronization of videos is an essential pre-processing step for multi-view reconstruction such as the image mosaic by UAV remote sensing; it is often solved with hardware solutions in motion capture studios. However, traditional synchronization setups rely on manual interventions or software solutions and only fit for a particular domain of motions. In this paper, we propose a self-supervised video synchronization algorithm that attains high accuracy in diverse scenarios without cumbersome manual intervention. At the core is a motion-based video synchronization algorithm that infers temporal offsets from the trajectories of moving objects in the videos. It is complemented by a self-supervised scene decomposition algorithm that detects common parts and their motion tracks in two or more videos, without requiring any manual positional supervision. We evaluate our approach on three different datasets, including the motion of humans, animals, and simulated objects, and use it to build the view panorama of the remote sensing field. All experiments demonstrate that the proposed location-based synchronization is more effective compared to the state-of-the-art methods, and our self-supervised inference approaches the accuracy of supervised solutions, while being much easier to adapt to a new target domain. [ABSTRACT FROM AUTHOR]

With the development of the smart ocean which contains a large number of wireless sensor nodes, it is a great demand to develop high‐performance marine energy harvesters for powering those sensors. In this work, a highly adaptive hybrid nanogenerator based on triboelectric nanogenerator and electromagnetic generator is proposed. The hybrid nanogenerator can be used for scavenging both wind energy and ocean current energy. The peak power of the hybrid nanogenerator can reach 449.74 mW, which can recharge a 50 mAh‐3.7 V Lithium battery. In addition, it is found that there is a linear relationship between the voltage frequency of the triboelectric nanogenerator and the rotation speed, indicating the hybrid nanogenerator can serve as a flow velocity sensor. A fully self‐powered marine wireless sensor node is fabricated based on the hybrid nanogenerator and management circuit. The demonstrations show that the present hybrid nanogenerator has great potential applications for marine wireless sensing in the scenarios of nearshore, offshore, and underwater. [ABSTRACT FROM AUTHOR]

Machine learning is a new discipline in computer science that has given learning a new dimension. It emulates human intelligence in the system by learning from previous data. Due to its self-adaptive capabilities and mathematical foundation, machine learning has several applications in healthcare, sentiment analysis, recommendation systems, natural language processing, information retrieval, gameplay, market research, text recognition, and computer vision. Machine learning may be used to enhance healthcare services. Machine learning decision support systems may be used to diagnose various illnesses. This article's study focuses on detecting heart disease using a machine learning technique. CAD (coronary artery disease) is a kind of heart disease that, if not treated promptly, may result in cardiac arrest. It is the leading cause of death globally. CAD is caused by a blockage in the arteries that provide blood to the heart muscles. Plaque deposition, composed of cholesterol and calcium, causes artery blockages. Atherosclerosis is the medical term for this condition. The endothelium is a thin strip of cells that causes smooth blood flow through arteries. Plaque accumulates in the arterial wall when the endothelium deteriorates. This plaque formation affects the blood flow to the heart, causing blood cells to weaken. CAD worsens with time, and if not treated promptly, symptoms might deteriorate. Plaque accumulates over time, hardening plaque. Platelets may cling together to create blood clots in arteries if the plaque ruptures. It has the potential to entirely cut off the blood flow, resulting in a heart attack. Angiography is a common approach for diagnosing CAD. This procedure is invasive and has certain adverse effects. As a result, non-invasive approaches for diagnosing CAD based on clinical data are required. This article provides an ET-SVMRBF (extra tree support vector machine radial basis function) technique for diagnosing CAD using clinical deed. This approach was validated using the Z-Alizadeh Sani CAD dataset from UCI University of California, Irvine). The primary goal for introducing this technology is to minimize mortality by detecting CAD early. [ABSTRACT FROM AUTHOR]

Featured Application: The Timed-SAS approach can be used to design and quantitatively analyze the complex self-adaptive software systems, such as cloud computing systems. Self-adaptive software (SAS) is gaining in popularity as it can handle dynamic changes in the operational context or in itself. Time behaviors are of vital importance for SAS systems, as the self-adaptation loops bring in additional overhead time. However, early modeling and quantitative analysis of time behaviors for the SAS systems is challenging, especially under uncertainty environments. To tackle this problem, this paper proposed an approach called Timed-SAS to define, describe, analyze, and optimize the time behaviors within the SAS systems. Concretely, Timed-SAS: (1) provides a systematic definition on the deterministic time constraints, the uncertainty delay time constraints, and the time-based evaluation metrics for the SAS systems; (2) creates a set of formal modeling templates for the self-adaptation processes, the time behaviors and the uncertainty environment to consolidate design knowledge for reuse; and (3) provides a set of statistical model checking-based quantitative analysis templates to analyze and verify the self-adaptation properties and the time properties under uncertainty. To validate its effectiveness, we presented an example application and a subject-based experiment. The results demonstrated that the Timed-SAS approach can effectively reduce modeling and verification difficulties of the time behaviors, and can help to optimize the self-adaptation logic. [ABSTRACT FROM AUTHOR]

This paper studies a task execution and routing problem in the autonomous transportation system that maps to the famous orienteering problem (OP) in operations research. By extending the classical OP to stochastic and dynamic OP with stochastic travel time and service time, an investigation is conducted that aims to optimize the selection of travel routes with the objective of maximizing the total collected rewards in accordance with the time availability of self-adaptive agent (e.g., unmanned aircraft vehicles and unmanned ground vehicle). Three self-adaptive multi-stage heuristics-based adjustment strategies (i.e., Insertion-and-Removal Heuristics, Hybrid Simulated Annealing-Tabu Search-based Re-Optimization, and Threshold-based Re-Optimization) are proposed to equip the autonomous agent with the ability of real-time routing adjustment in the context of the dynamic and stochastic orienteering problem. Finally, benchmark simulation experiments are used to test the effectiveness of all the three proposed adjustment strategies. The experimental results show that, compared with a priori solutions acquired by a static meta-heuristic algorithm (Hybrid Simulated Annealing-Tabu Search algorithm), the three strategies can improve the solution qualities at different levels. [ABSTRACT FROM AUTHOR]