Real-time Information Processing Research Articles

A Bayesian network method using transfer learning for solving small data problems in abnormal condition diagnosis of fused magnesia smelting process

Intelligent abnormal condition diagnosis is key to the safe and stable operation of the fused magnesia smelting process (FMSP). Existing studies have applied the Bayesian network (BN) to the FMSP with sufficient data. However, learning an accurate BN model for a new FMSP using small data isn't easy. Furthermore, the unreasonable diagnosis may seriously affect product quality and security threats. To this end, we present a novel BN transfer learning (BNTL) method to resolve this abnormal condition diagnosis problem using small data. The abnormal condition diagnosis model is established offline based on the proposed BNTL method, and the occurrence or extent of abnormal conditions is identified online based on the diagnosis model and real-time process information. The BNTL involves structure and parameter transfer learning. Among them, structure transfer learning is based on score and search. A new scoring function is proposed to evaluate the local structure of the candidate target domain by weighted integration of source and target domain data based on similarity. The search process is based on the genetic algorithm with other constraints that remove loops and control the number of parent nodes. After learning the structure, parameter transfer learning is done according to a new similarity evaluation and fusion algorithm. The public network is first used to evaluate the BNTL method. Then, it is utilized to build the abnormal condition diagnosis model of the FMSP using small data. Experimental results show that the recognition accuracy is significantly improved compared with other advanced methods in a limited data environment.

Analysis and Recommendation of Outdoor Activities for Smart City Users Based on Real-Time Contextual Data

Smart cities leverage advanced technologies to enhance urban living through the real-time collection, processing, and analysis of contextual information. The potential to improve residents’ outdoor experiences in these cities increases dramatically as smart technologies are integrated into urban environments, making them more interconnected. It helps to explore the pivotal role of real-time data in optimizing various aspects of city management, focusing on key domains such as traffic, public transportation, emergency response, waste management, and environmental monitoring. A variety of datasets, such as those on the weather, air quality, traffic patterns, event schedules, and user activity patterns, are gathered and analyzed as part of the methodology. This data are processed and interpreted using machine learning algorithms, which find correlations, trends, and patterns that affect outdoor activities. Suggestions for appropriate outdoor activities can be generated in real time based on contextual information, past behavior, and user preferences. This model addresses the dynamic and context-aware nature of Smart Cities by proposing a novel framework for real-time contextual information prediction and personalized outdoor activity suggestions for users. Leveraging the vast amount of data generated by Smart City infrastructure, this study integrates advanced data analysis techniques with deep learning models to enhance the urban living experience. A variety of datasets, such as those on the weather, air quality, traffic patterns, event schedules, and user activity patterns, are gathered and analyzed as part of the methodology. This data are processed and interpreted using machine learning algorithms, which find correlations, trends, and patterns that affect outdoor activities. Suggestions for appropriate outdoor activities can be generated in real time based on contextual information, past behavior, and user preferences. The framework begins by collecting and processing diverse datasets from sensors, Internet of Things (IoT) devices, and other urban sources to create a comprehensive understanding of the current city context. Deep learning models, such as recurrent neural networks (RNNs) and convolutional neural networks (CNNs), are employed to analyze this data and predict real-time contextual information, including weather conditions, traffic patterns, and social events. It contributes to the growing field of Smart Cities by introducing a scalable and adaptable framework that harnesses the power of deep learning to improve urban living. The result shows that the proposed air pollution model predicted 96.06700 PM2.5 concentration levels, subsequently the temperature model predicted 14.06800∘C. The integration of real-time contextual information prediction and personalized outdoor activity suggestions showcases the potential for creating more engaging and user-centric Smart City ecosystems. This research attempts to provide personalized recommendations that are in line with users’ preferences, the state of the environment at the time, and other pertinent contextual factors by utilizing data from multiple sources, including IoT devices, mobile applications, and environmental sensors.

Artificial Tactile Perception System Based on Spiking Tactile Neurons and Spiking Neural Networks.

The artificial tactile perception system of this work utilizes a fully connected spiking neural network (SNN) comprising two layers. Its architecture is streamlined and energy-efficient as it directly integrates spiking tactile neurons with piezoresistive sensors and Pt/NbOx/TiN memristors as input neurons. These spiking tactile neurons possess the ability to perceive and integrate pressure stimuli from multiple sensors and encode the information into rate-coded electrical spikes, closely resembling the behavior of a biological tactile neuron. The system's real-time information processing capability is demonstrated through an artificial perceptual learning system that successfully encodes and decodes the Morse code; the artificial perceptual learning system accurately recognizes and displays 26 English letters. Furthermore, the artificial tactile perception system is evaluated for the recognition of the MNIST data set, achieving a classification accuracy of 85.7% with the supervised spiking-rate-dependent plasticity learning rule. The key advantages of this artificial tactile perception system are its simple structure and high efficiency, which contributes to its practicality for various real-world applications.

Cloud material handling systems: a cyber-physical system to enable dynamic resource allocation and digital interoperability

AbstractThe existing logistics practices frequently lack the ability to effectively handle disruptions. Recent research called for dynamic, digital-driven approaches that can help prioritise allocation of logistics resources to design more adaptive and sustainable logistics networks. The purpose of this study is to explore inter-dependencies between physical and digital assets to examine how cyber-physical systems could enable interoperability in logistics networks. The paper provides an overview of the existing literature on cyber-physical applications in logistics and proposes a conceptual model of a Cloud Material Handling System. The model allows leveraging the use of digital technologies to capture and process real-time information about a logistics network with the aim to dynamically allocate material handling resources and promote asset and infrastructure sharing. The model describes how cloud computing, machine learning and real-time information can be utilised to dynamically allocate material handling resources to product flows. The adoption of the proposed model can increase efficiency, resilience and sustainability of logistics practices. Finally, the paper offers several promising research avenues for extending this work.

Digital enabled agility: Industry 4.0 unlocking real-time information processing, traceability, and visibility to unleash the next extent of agility

ABSTRACT Digital enabled agility reflects the transformative outset of an organisation to first upgrade its business processes digitally towards facilitating real-time information processing/capabilities. Second, tapping from real-time capabilities, digital enabled agility maximises an agile response's efficacy in enhancing the sophistication of rapidity, flexibility, and sensing involved. In this backdrop, driven by newer IT infrastructure and disruptive technology, this research accentuates on Industry 4.0 digitalisation to understand how a manufacturer can gain digital enabled agility. An empirical survey of manufacturers within the premise of Industry 4.0 is conducted. Drawn from the findings, the capabilities of traceability and visibility can be facilitated by Industry 4.0 that would improve information sharing and information quality in real-time. These capabilities are key propellants to unlock the digital extent of agility. This research however details further how this scope of agility is an extension to that of the scope of typical agility in terms of its ability to handle expansive information processing needs and dynamic coordination/resource orchestration. Insights are weaved under the purviews of (a) organisational information processing view and (b) the manufacturer who is coordinating contemporary dynamic production and allied value chain requirements.

Alternative to mean and least squares methods used in processing the results of scientific and technical experiments

Increasing the complexity and size of systems of various nature requires constant improvement of modeling and verification of the obtained results by experiment. It is possible to clearly conduct each experiment, objectively evaluate the summaries of the researched process, and spread the material obtained in one study to a series of other studies only if they are correctly set up and processed. On the basis of experimental data, algebraic expressions are selected, which are called empirical formulas, which are used if the analytical expression of some function is complex or does not exist at this stage of the description of the object, system or phenomenon. When selecting empirical formulas, polynomials of the form: у = А0 + А1х+ А2х2+ А3х3+…+ Аnхn are widely used, which can be used to approximate any measurement results if they are expressed as continuous functions. It is especially valuable that even if the exact expression of the solution (polynomial) is unknown, it is possible to determine the value of the coefficients An using the methods of mean and least squares. But in the method of least squares, there is a shift in estimates when the noise in the data is increased, as it is affected by the noise of the previous stages of information processing. Therefore, for real-time information processing procedures, a pseudo-reverse operation is proposed, which is performed using recurrent formulas. This procedure is a procedure of successive updating (with a shift) along the columns of the matrix of given sizes and pseudo-reversal at each step of information change. This approach is straightforward and takes advantage of the bounding method. With pseudo-inversion, it is possible to control the correctness of calculations at each step, using Penrose conditions. The need for pseudo-inversion may arise during optimization, forecasting of certain parameters and characteristics of systems of various purposes, in various problems of linear algebra, statistics, presentation of the structure of the obtained solutions, to understand the content of the incorrectness of the resulting solution, in the sense of Adomar-Tikhonov, and to see the ways of regularization of such solutions.

Task Scheduling of Real-Time Traffic Information Processing Based on Digital Twins

The Intelligent Transportation System under Digital Twins can provide accurate data sources for traffic control. The present work focuses on the real-time information processing and task scheduling problems of the Internet of Vehicles (IoV) system based on Virtual Reality. They are the Quality/Distance Algorithm (QDA), Task Density Algorithm, Distance Balance Algorithm (DBA), and Bionic-DBA (B-DBA). The simulation experiment analysis suggests that the DBA algorithm takes the balance of travel distance into account and effectively improves task quality. The Utility Function in B-DBA and the Biological Heuristic Search Algorithm in Pareto Ant Colony Optimization play a critically important role in enhancing the overall task quality. In addition, a Transmission based on Privacy Protection (TPP) algorithm is designed to protect the attribute-based privacy information in the traffic information transmission system. This algorithm ensures that the real-time traffic information processing system resists various attacks from malicious nodes. It has been verified that when the number of selfish nodes accounts for 30%, the transmission efficiency of the TPP algorithm reaches 0.77. The research content has a practical reference value for providing users with continuous and high-quality IoV network services.

A Large-Scale Photonic CNN Based on Spike Coding and Temporal Integration

The real-time massive-data intelligent information processing tasks highlight a vital requirement for the novel, intelligent optimization hardware. Convolutional neural networks are highly capable of extracting the hierarchical feature map and enhancing recognition accuracy, with photonics-enabled implementations drawing considerable interest. Here, we propose a large-scale and reconfigurable photonic convolutional neural network (PCNN), based on a hardware-friendly distributed feedback laser diode (DFB-LD). Our approach applies biological time-to-first-spike coding to a DFB-LD neuron to perform a temporal convolutional operation (TCO) for image processing tasks. In PCNN, experimental results demonstrate that we successfully implement the TCO to extract the image features with convolutional kernels of size 11×11. Furthermore, we investigate the temporal pulse shaping of a DFB-LD neuron to build a densely-connected fully connected layer, which synaptic weights can be rapidly adjusted at a rate of 5 GHz, achieving full MNIST and Fashion-MNIST benchmark image classification tasks, with classification accuracies of 98.56% and 87.48%, respectively. This work highlights the potential of neuron-like optical analog computing platforms for real-time and more complex intelligent processing networks.

Design of an ALU in QCA Technology Dedicated to Intelligent Edge Computing Systems

Following the considerable development of big data and the need for real-time information processing, the recent computing paradigm changed from centralized intelligence on cloud computing to the distributed intelligence on Edge Computing. On the algorithm layer, Artificial Intelligence (AI) has been applied in different areas related to intelligent-edge applications. In this type of application, consumption and speed present essential challenges when designing hardware-level architecture on the embedded systems. In particular, an Edge Computing System (ECS) must be an energy efficient, compact, and fast component integrated into edge devices. Miniaturization has been happening at a steady pace for the last years. However, the MOS transistor size has been slowly reaching the physical limit. In this paper, we propose a Quantum Cellular Automata clock-phase-based technique for the design of the proposed ALU module. The obtained results using the QCA architecture designer exhibit better performances over the existing designs in terms of size, cell number, cost, latency, and power consumption. We have also used QCAPro software for power consumption estimation. The results show the effectiveness of the design in terms of cost and power.

Are medical oncologists ready for the artificial intelligence revolution? Evaluation of the opinions, knowledge, and experiences of medical oncologists about artificial intelligence technologies.

The use of artificial intelligence technologies (AIT) in medicine is increasing worldwide. In this study, it was aimed to evaluate the experiences, opinions, and future expectations of medical oncologists on artificial intelligence (AI). After the reliability and validity analyses were carried out by a pilot study, the main online questionnaire was sent to the members of the "Turkish Society of Medical Oncology" mail group byan invitation e-mail. The anonymized responses of the participants were analyzed. The median age of the 156 participants was 36 (34-43) years and half (51%) were male. Most (45%) were fellows. Forty-six percent were working in university hospitals, 56% were visiting 20-40 patients a day. Medical oncologists' view of AIT was mostly positive (78%). However, some (especially women) had doubts about the reliability of AI (44%) and the establishment of its ethical/legal basis (49%). Sixty-five percent of the participants had no/superficial knowledge about AI. More than half (55%) had never used AI-based applications in their academic or clinical work. However, unlike now, 80% of the participants believed that they would use AIT frequently in their practice in thefuture and it would be beneficial. The most anticipated (81%) benefit was real-time information processing and real-time access to big data. Sixty-two percent believed that information about AI should be in the education curriculum. The vast majority of respondents (79%) thought that AI would not completely replace medical oncologists in thefuture. Some differences were found in the perception and experience of oncologists according to age, gender, title, and the number of patients examined per day. About AI, the general opinion of medical oncologists was positive, but their level of knowledge and use was low. However, they thought they would use it frequently in future and needed training.

High-frame-rate reconfigurable diffractive neural network based on superpixels.

The existing implementations of reconfigurable diffractive neural networks rely on both a liquid-crystal spatial light modulator and a digital micromirror device, which results in complexity in the alignment of the optical system and a constrained computational speed. Here, we propose a superpixel diffractive neural network that leverages solely a digital micromirror device to control the neuron bias and connection. This approach considerably simplifies the optical system and achieves a computational speed of 326 Hz per neural layer. We validate our method through experiments in digit classification, achieving an accuracy of 82.6%, and action recognition, attaining a perfect accuracy of 100%. Our findings demonstrate the effectiveness of the superpixel diffractive neural network in simplifying the optical system and enhancing computational speed, opening up new possibilities for real-time optical information processing applications.

The Effects of Social Status and Imposition on the Comprehension of Refusals in Chinese: An ERP Study.

This study aims to examine how real-time processing of information about the social status of interlocutors (high vs. low) and the imposition of making a refusal by manipulating the indirectness of invitation forms (declining direct invitations vs. declining indirect invitations) affects the interpretation of refusals in Chinese. The event-related potentials results showed that high-status invitees who decline invitations from low-status inviters elicited weaker N400 effects followed by late mitigated negative effects, while high imposition refusals elicited stronger N400 effects followed by increased late negativities. The two factors of social status and imposition functioned independently during the comprehension of refusal utterances. These findings suggest that individuals take the social status of interlocutors and the imposition of making a refusal into consideration as an utterance unfolds, while face-threatening contexts create inferential difficulties for reinterpreting the pragmatic implications of an utterance.

Emerging Memtransistors for Neuromorphic System Applications: A Review

The von Neumann architecture with separate memory and processing presents a serious challenge in terms of device integration, power consumption, and real-time information processing. Inspired by the human brain that has highly parallel computing and adaptive learning capabilities, memtransistors are proposed to be developed in order to meet the requirement of artificial intelligence, which can continuously sense the objects, store and process the complex signal, and demonstrate an “all-in-one” low power array. The channel materials of memtransistors include a range of materials, such as two-dimensional (2D) materials, graphene, black phosphorus (BP), carbon nanotubes (CNT), and indium gallium zinc oxide (IGZO). Ferroelectric materials such as P(VDF-TrFE), chalcogenide (PZT), HfxZr1−xO2(HZO), In2Se3, and the electrolyte ion are used as the gate dielectric to mediate artificial synapses. In this review, emergent technology using memtransistors with different materials, diverse device fabrications to improve the integrated storage, and the calculation performance are demonstrated. The different neuromorphic behaviors and the corresponding mechanisms in various materials including organic materials and semiconductor materials are analyzed. Finally, the current challenges and future perspectives for the development of memtransistors in neuromorphic system applications are presented.

Spectrally Resolved Hong-Ou-Mandel Interferometry with Discrete Color Entanglement

Hong-Ou-Mandel (HOM) interferometry is a powerful quantum metrology technique to reconstruct the surface-depth profile of transparent samples with complementary benefits of measurement precision and experimental robustness. It is generally assumed that a trade-off exists between the standard HOM interferometry sensitivity and its ambiguity-free dynamic range. Here we challenge this ``well-known'' assumption through the implementation of spectrally resolved HOM interferometry with discrete color entanglement, and show that the well-separated frequencies embedded in a quantum state suffices to achieve great advantages in enhancing the temporal resolution but without limiting its ambiguity-free dynamic range. Backed by the Fisher-information analysis, spectrally resolved coincidence detection is an optimal measurement strategy as it enables us to recover the quantum Cram\'er-Rao bound. These results may significantly facilitate the use of spectrally resolved quantum interference for real-time quantum information processing, such as ultrafast phase transition and spectral-temporal photography.

Evaluation model of cultural heritage tourist attractions based on network virtual resource sharing and real-time information processing

With the continuous development of modern society, people’s demand for the network is also rising. Because of its powerful function, wide coverage, and other advantages, people in modern society have higher and higher utilization rate of the network, and many things are realized with the help of the network. At the same time, some new business based on the network is emerging. The rapid development of network technology makes many people rush for it. This paper mainly studies how to improve the cache technology, reasonable, and effective allocation of resources, so that people can get a better experience. This paper studies the suitability evaluation of tourism development of regional cultural heritage resources by using Delphi method and AHP analytic hierarchy process and infiltrates the evaluation items into various levels, such as natural environment conditions, social and economic conditions, resource combination levels, and so on. Through the analysis and comparison of different geographic information, we can clearly understand whether the region is suitable for tourism and the spatial differences of some regions. To a certain extent, it lays the foundation for the future development of cultural heritage tourist attractions, and the spatial analysis based on GIS, which is consistent with the tourism suitability evaluation index in the field of regional cultural heritage. When comparing the data with different measurement levels, we can directly determine the characteristics of the spatial differences of different cultural heritage resources, which depends on the value of resources and location conditions of a specific region.

Non-Linear Signal Processing Methods for UAV Detections from a Multi-Function X-Band Radar

This article develops the applicability of non-linear processing techniques such as Compressed Sensing (CS), Principal Component Analysis (PCA), Iterative Adaptive Approach (IAA), and Multiple-input-multiple-output (MIMO) for the purpose of enhanced UAV detections using portable radar systems. The combined scheme has many advantages and the potential for better detection and classification accuracy. Some of the benefits are discussed here with a phased array platform in mind, the novel portable phased array Radar (PWR) by Agile RF Systems (ARS), which offers quadrant outputs. CS and IAA both show promising results when applied to micro-Doppler processing of radar returns owing to the sparse nature of the target Doppler frequencies. This shows promise in reducing the dwell time and increases the rate at which a volume can be interrogated. Real-time processing of target information with iterative and non-linear solutions is possible now with the advent of GPU-based graphics processing hardware. Simulations show promising results.

RadarVerses in Metaverses: A CPSI-Based Architecture for 6S Radar Systems in CPSS

Metaverses have caused significant changes in the industry and their academic foundation can be traced back to the term cyber–physical–social systems (CPSS), which was proposed in 2010. Radar is an important sensor in sensing systems that are widely applied in many fields, especially, in autonomous driving. To deal with the complex environment, smart radars with real-time information processing capabilities are required. Human factors play a critical role in the operation and management of radar systems, thus, digital twins’ radars in cyber–physical systems (CPS) are unable to achieve intelligence in CPSS due to an incomplete consideration of human involvement. For this consideration, we propose a novel framework of RadarVerses for smart radars in metaverses based on ACP-based parallel intelligence, which is also known as cyber–physical–social intelligence (CPSI). RadarVerses consist of five main parts which are physical radars, descriptive radars, predictive radars, prescriptive radars, and deep radars. To construct RadarVerses at the technical level, we introduce four main technical foundations: 1) communication technology; 2) scenarios engineering; 3) foundation models; and 4) digital workers. In addition, we also provide a case study about LiDARs’ predictive maintenance of accumulated snow in RadarVerses.

Rab10 regulates the sorting of internalised TrkB for retrograde axonal transport.

Neurons process real-time information from axon terminals to coordinate gene expression, growth, and plasticity. Inputs from distal axons are encoded as a stream of endocytic organelles, termed signalling endosomes, targeted to the soma. Formation of these organelles depends on target-derived molecules, such as brain-derived neurotrophic factor (BDNF), which is recognised by TrkB receptors on the plasma membrane, endocytosed, and transported to the cell body along the microtubules network. Notwithstanding its physiological and neuropathological importance, the mechanism controlling the sorting of TrkB to signalling endosomes is currently unknown. In this work, we use primary mouse neurons to uncover the small GTPase Rab10 as critical for TrkB sorting and propagation of BDNF signalling from axon terminals to the soma. Our data demonstrate that Rab10 defines a novel membrane compartment that is rapidly mobilised towards the axon terminal upon BDNF stimulation, enabling the axon to fine-tune retrograde signalling depending on BDNF availability at the synapse. These results help clarifying the neuroprotective phenotype recently associated to Rab10 polymorphisms in Alzheimer's disease and provide a new therapeutic target to halt neurodegeneration.

Full-deployed energy management system tested in a microgrid cluster

Under a hierarchical structure perspective, the integration of tools like the internet of things, cloud computing, and machine learning into a microgrid real-time energy management system involves superior capabilities like an autonomous and scalable design, massive storage capabilities, real-time information analysis and processing, and security issues control, to mention a few. This paper evaluates most of them using a cloud-based real-time energy management system integrated into a real-life hardware-in-the-loop testbed. The proposed cloud-based system is tested by solving an economic power dispatch problem including the equalization of the multiple battery-based energy storage systems interacting within a multi-microgrid environment. The test assessment combined reviewing and running microgrid models, incorporating these models into a real-life power-hardware-in-the-loop unit, linking the testbed to a cloud server, and merging the energy management system with on-demand computing tools, primarily machine learning and the internet of things. As established by the experimental evidence, this paper cites the benefits of combining machine learning techniques and internet of things tools with a scalable and autonomous real-life cloud-based energy management system architecture to improve the framework's functionality, enhance the energy forecasting for generation and usage, and cut down the price paid to the service provider.

Photoelectric Memristor-Based Machine Vision for Artificial Intelligence Applications

With the rapid development of next-generation artificial intelligence technology, research on advanced machine vision has received extensive attention. It is well-known that significant progress has been made in artificial vision systems based on light sensors, but the separate light sensor and memory require additional time for information transfer to realize computation due to the limitation of the von Neumann architecture, which delays the computational speed and hinders large-scale integration. In recent years, the emergence of photoelectric memristors has brought new inspiration to the study of machine vision, which is expected to overcome the above problems. Photoelectric memristors can not only respond directly to light stimuli but also perform temporary memory and real-time processing of visual information and sensory data, providing a promising hardware foundation for the development of artificial vision systems. In this review, the background and related theory of photoelectric memristors and machine vision are first introduced. Then, the research progress of photoelectric memristors and machine vision based on them is reviewed. Finally, the existing problems impeding the progress of machine vision based on photoelectric memristors are summarized, and the future development is predicted.