Real-time Services Research Articles

IOT Based Car Parking Using Edge Computing

The promise of autonomous driving is that edge computing can deliver real time services. Using a variety of sensors and communication modules, autonomous cars may function both indoors including in parking lots and outdoors. But because the interior station's ability to communicate with the outer world is limited globe, and the present underpowered autos don't really work; they could have to handle all the data. Autonomously, hence a control mechanism must be put in place. Completely self-sufficient car in the surroundings. In this research, we suggest an edge computing based smart parking system for self driving cars. Smart parking has resolved Boundaries and traffic, and parking spaces are managed using a grid. To assess the effectiveness of the smart parking system, In the current parking system, we contrasted the journey time with the parking time. Furthermore, the resource to make sure that the service quality margin number is offered and maintained; expenses and transaction data are examined.

Resource-constrained edge-based deep learning for real-time person-identification using foot-pad

To build a smart home that assists people with disabilities, we urgently need a non-invasive user identification system capable of recognizing family members in real-time and being easy to use. Existing systems face issues such as privacy concerns from cameras, inconvenience from needing physical contact with sensors, and the requirement to always carry a specific device. Additionally, these systems typically depend on edge nodes to collect data and then transfer it to cloud servers for high-performance inference. This dependency leads to network delays, hindering real-time service, and introducing security issues. To address these requirements, this paper presents a real-time non-invasive user identification system that recognizes users as they step on a foot pad. This study introduces an edge node designed to measure real-time foot pressure distribution data, along with a preprocessing system for data generalization. Additionally, we propose a system that performs real-time user inference using only resource-constrained edge nodes to overcome the challenges of cloud-based systems, including addressing security issues without specific protocols. To achieve this, we optimized various deep learning-based user identification models to be executable on edge nodes and then compared their performance. As a result, using a ResNet18 model through pruning and post quantization training with integer, we achieved inference within 1.5 s with 85% accuracy. Compared to the worst-performing AlexNet, the ResNet18 model shows a substantial reduction in model size by approximately 33%, a decrease in memory usage by about 80%, and a significant increase in inference speed by over tenfold.

An Online Knowledge-Based Support System

Technology advances have touched all the fields of human life, and the educational system is not an exception. Students in Nigerian universities experience challenges in their early or later stages of school due to a lack of a knowledge-based support system. This results in massive failure and confusion in later days in school. A knowledge-based online support system has been developed and implemented in this paper to improve academic support services for students and faculty members in the university. The need for such a support system stems from the various academic problems that the students often experience, including a lack of answers to certain questions, which can result in academic failure, taking too long to graduate, or dropping out of the university altogether. The system is web-based; therefore, users can access the necessary information when needed, and, in most cases, it solves common academic issues in a short time using the web. Following the Agile software development methodology and employing a client-server architecture, the system was designed in a user-friendly and efficient manner. Highlights of the system include an extensive database of FAQs, real-time interactive services, and multi-typed query handling. Testing phases were performed and integrated, and modular testing showed that the system worked as intended, providing valid inputs and also flagging any invalid inputs without causing the system failures, and performing provisions for user corrections. The results suggest that the online support system could enhance the academic environment at AE-FUNAI and therefore would benefit students and staff members alike. Nevertheless, certain limitations still prevail, such as poor internet access, which calls for constant enhancement and provision of support to the system.

Bringing Collaborative Positioning to Native 5G Systems for Enhanced 2D & 3D Location Services

5G's large bandwidths and ubiquity make it a promising candidate for catalyzing future mobile services (emergency, V2X, XR, etc.) that benefit from accurate positioning over a wide area. However, we demonstrate that its heavy reliance on infrastructure base stations significantly limits its positioning ability in urban environments, where multi-story buildings and urban canyons are the norm, resulting in only a small fraction of devices being positioned, often only in 2D. To this end, we propose CAMP, a novel, collaboration-augmented paradigm for 2D and 3D positioning with cellular networks, where UEs leverage 5G NR's peer-peer Sidelink protocol to directly enable positioning measurements between each other in a scalable manner. Leveraging better connectivity (higher degree and SNR) between devices, CAMP overcomes the challenges inherent to infrastructure approaches. It brings algorithmic innovations in distributed, stochastic access and dynamic resource scheduling in a 5G-compliant manner to Sidelink measurements that are otherwise not scalable in conventional gNB-driven Sidelink operation. CAMP is implemented as a near real-time xApp service in the Open-RAN framework. Evaluations in a small cell 5G test-bed and a large-scale RAN on Colosseum, highlight its ability to position over 80% (72%) UEs with under 3m accuracy in 2D (3D) in real-world urban scenarios (compared to 30% with infrastructure approaches) and with 7-25x less overhead compared to baseline collaborative schemes.

Toward real-time operations of modular-vehicle transit services: From rolling horizon control to learning-based approach

Recent technological advancements have opened doors for real-time adjustments and controls during public transport operations. In particular, the introduction of modular vehicles has the potential to significantly enhance public transit service quality. This innovative public transit service with modular vehicles, characterized by its flexible schedules and vehicle formations, allows for the dynamic management of transit capacity to meet the fluctuating passenger demands. This paper proposes to schedule the flexible modular-vehicle transit service in real time considering the varying demands. To jointly optimize the service schedule and vehicle formations, we propose the rolling horizon control approach to decompose the complex problem into subproblems that can be solved efficiently during the process. On top of this, we introduce a learning-based optimization proxy to streamline the optimization process within the rolling horizon framework, enabling near-optimal decisions to be made with minimal execution time without directly solving the optimization problem. Through numerical studies, we demonstrate the effectiveness and efficiency of the proposed methods in terms of solution quality and efficiency. Furthermore, our case studies show that modular vehicles can adapt to the changing demand and effectively reduce the total costs in the transit system.

Flexible Hyper-Distributed IoT–Edge–Cloud Platform for Real-Time Digital Twin Applications on 6G-Intended Testbeds for Logistics and Industry

This paper presents the design and development of a flexible hyper-distributed IoT–Edge–Cloud computing platform for real-time Digital Twins in real logistics and industrial environments, intended as a novel living lab and testbed for future 6G applications. It expands the limited capabilities of IoT devices with extended Cloud and Edge computing functionalities, creating an IoT–Edge–Cloud continuum platform composed of multiple stakeholder solutions, in which vertical application developers can take full advantage of the computing resources of the infrastructure. The platform is built together with a private 5G network to connect machines and sensors on a large scale. Artificial intelligence and machine learning are used to allocate computing resources for real-time services by an end-to-end intelligent orchestrator, and real-time distributed analytic tools leverage Edge computing platforms to support different types of Digital Twin applications for logistics and industry, such as immersive remote driving, with specific characteristics and features. Performance evaluations demonstrated the platform’s capability to support the high-throughput communications required for Digital Twins, achieving user-experienced rates close to the maximum theoretical values, up to 552 Mb/s for the downlink and 87.3 Mb/s for the uplink in the n78 frequency band. Moreover, the platform’s support for Digital Twins was validated via QoE assessments conducted on an immersive remote driving prototype, which demonstrated high levels of user satisfaction in key dimensions such as presence, engagement, control, sensory integration, and cognitive load.

Development of Traffic Scheduling Based on TSN in Smart Substation Devices

Smart substations are an important trend in substation construction. With increasing data traffic, it is difficult for the traditional Ethernet network to meet the real-time requirements of control information in smart substations. Hence, in this paper, a deterministic network architecture for substations based on time-sensitive networks (TSN) has been developed in order to realize the domain-wide time synchronization and efficient real-time communication of the “three-layer and two-network” model in smart substations. Furthermore, a design scheme for substation automation equipment based on TSN is proposed. The proposed device realizes the timely transmission of real-time control information packets by utilizing the Earliest TxTime First (ETF) Qdisc technology of Linux and the timing sending capability of Intel 210 NIC. Furthermore, it collaborates with the time-aware shaper (TAS) traffic scheduling mechanism of TSN switches to ensure the end-to-end deterministic delay of time-sensitive traffic. As a result, it provides efficient real-time communication services with low latency and jitter for smart substation automation systems.

ACORN+: Adaptive Compression-Reconstruction for Device-Cloud Collaboration Video Services

With the improvement of edge-based autonomous systems such as mobile Industrial IoT(IIoT) networks, edge devices can capture and upload videos with increasing bitrates. Massive edge-computing end nodes are eager for adequate multimedia data to satisfy the requirements of real-time video services. However, existing encoding standards for video services in Web 2.0 are specifically designed for something other than IoT video streaming. We have improved our Adaptive Compression-Reconstruction framework ACORN to obtain ACORN+, based on compressed sensing and recent advances in deep learning. At end nodes, we compress multiple sequential video frames into a single frame to reduce video volume. Given that multiple kinds of intelligent tasks are expected to be finished on the device side, we also designed a device-cloud collaboration scheme where deep learning-based algorithms can be executed on both the device and server sides. Experiments reveal that video analytics can be conducted on compressed frames. Taking action recognition as a device-cloud collaboration use case, we find ACORN+ obtains more than 3x speedup on compressed frames. The reconstruction algorithm in ACORN+ is with 1- 4 dB improvements. Moreover, the encoding time cost and the encoded video volume are reduced by more than 4x under the ACORN+ framework. 1

To ride-hail or not to ride-hail? Complementarity and competition between public transit and transportation network companies through the lens of app data

AbstractTransportation Network Company (TNC) services have become a prominent factor in urban transportation in recent years, and there is an ongoing debate regarding their relationship with public transit. While many argue that TNCs draw passengers away from public transportation, others believe the two modes complement each other. However, due to the inadequate sample size of rider surveys as primary data sources, our understanding of how riders choose between these two modalities remains limited. This study uses nine months of trip planning data generated by the Transit App, which captures how travelers engage with multiple options in real time, including TNC and public transit services. We extract measures from Transit describing the travel options and the habits of each individual user for sessions in which the user “tapped” on one of these two modes, indicating consideration of it as an option. Machine learning models predict the likelihood of a rider tapping TNC based on features of the available public transit options and other contextual factors (e.g., time of day, weather conditions). The models find that these taps are driven by factors that highlight the convenience of TNC, such as the waiting time, walking distance, and the number of transfers for public transportation trips. We also find that the majority of TNC trips tapped by app users combine the two modes when using the Transit App, with TNC acting as a connection to or from public transit. These results provide detailed additional evidence for current arguments for both competition and complementarity between TNC and public transit from a population that uses an app to navigate public transit.

Enhancing Urban Surveillance with Fog Computing, Mobile Cloud, and Big Data Analytics in 5G Networks

The new emerging applications in 5G network, in the context of the Internet of Everything (IoE), will introduce high mobility, high scalability, real-time, and low latency requirements that raise new challenges on the services being provided to the users. Fortunately, Fog Computing and Cloud Computing, with their service orchestration mechanisms offer virtually unlimited dynamic resources for computation, storage and service provision, that will effectively cope with the requirements of the forthcoming services. 5G will use the benefits of centralized high performance computing cloud centers, cloud and fog RANs and distributed peer-to-peer mobile cloud that will create opportunities for companies to deploy many new real-time services that cannot be delivered over current mobile and wireless networks. This paper evaluates a model for fog and cloud hybrid environment service orchestration mechanisms for 5G network in terms of energy efficiency per user for different payloads.

Distribution System Security Region with Energy Storage Systems

The high penetration of distributed energy resources (DERs) in distribution systems calls for advanced security management techniques. Hence, this paper proposes the model of the distribution system security region with energy storage systems (DSSR-ESS). The N-1 secure operation range of distribution systems integrating ESS is characterized. Further research reveals the effect and mechanism of ESS on the DSSR properties. Firstly, the ESS in distribution systems is clarified from the perspective of security, and the security service time and real-time security service capability are defined. Secondly, the DSSR-ESS model is formulated by: 1) selecting load and distributed generation (DG) as the state variables, 2) identifying the normal operation security and N-1 security scenarios, and 3) determining the complete security constraints. Then, to solve the time-varying and nonlinear N-1 DSSR-ESS model, an algorithm is developed. The analytical expressions of the region boundaries considering voltage constraints, line loss, and ESS SOC are further obtained. Finally, a modified IEEE-RBTS-BUS4 system and its expanded 104-node test system are used to verify the proposed model. Comparing with the existing research, the proposed DSSR-ESS model considers ESS. This paper concludes that ESS bidirectionally adjusts the power flow of its upstream branches, thereby expanding the security region both in positive and negative directions. But limited by state space boundaries and security boundaries unaffected by ESS, increasing ESS capacity may not lead to DSSR-ESS expansion. Based on this discovery, a sitting strategy for ESS is developed to expand the security region, which demonstrates the practical application of DSSR-ESS in the security management of distribution systems.

Automated Real-Time Tool for Promoting Crisis Resource Use for Suicide Risk (ResourceBot): Development and Usability Study.

Real-time monitoring captures information about suicidal thoughts and behaviors (STBs) as they occur and offers great promise to learn about STBs. However, this approach also introduces questions about how to monitor and respond to real-time information about STBs. Given the increasing use of real-time monitoring, there is a need for novel, effective, and scalable tools for responding to suicide risk in real time. The goal of this study was to develop and test an automated tool (ResourceBot) that promotes the use of crisis services (eg, 988) in real time through a rule-based (ie, if-then) brief barrier reduction intervention. ResourceBot was tested in a 2-week real-time monitoring study of 74 adults with recent suicidal thoughts. ResourceBot was deployed 221 times to 36 participants. There was high engagement with ResourceBot (ie, 87% of the time ResourceBot was deployed, a participant opened the tool and submitted a response to it), but zero participants reported using crisis services after engaging with ResourceBot. The most reported reasons for not using crisis services were beliefs that the resources would not help, wanting to handle things on one's own, and the resources requiring too much time or effort. At the end of the study, participants rated ResourceBot with good usability (mean of 75.6 out of 100) and satisfaction (mean of 20.8 out of 32). This study highlights both the possibilities and challenges of developing effective real-time interventions for suicide risk and areas for refinement in future work.

Estimating the GPS time-varying differential code bias between C1 and P1 observations to provide a model product

Abstract Based on the time-varying characteristics of GPS satellite differential code biases (DCB) between C1 and P1 observations, a method for directly estimating its model coefficients is proposed to streamline processing and enhance product accuracy. Evaluation over three days compares estimated DCB (C1–P1) values and assesses real-time service capabilities in single point positioning (SPP). Three-day DCB (C1–P1) estimates match single-epoch time-varying DCB (C1–P1) series fitting accuracy. SPP results show significant improvement over Orbit Determination in Europe (CODE), reducing average root mean square error in three dimensions by 12 cm and by 4 cm compared to Chinese Academy of Sciences. Real-time DCB (C1–P1) service maintains stable and accurate performance, comparable to fitted values of single-epoch time-varying series. Real-time SPP application shows the method enhances positioning accuracy compared to CODE by up to 35%, confirming the feasibility and effectiveness of the proposed model estimation method.

Wearable Biosensor Smart Glasses Based on Augmented Reality and Eye Tracking.

With the rapid development of wearable biosensor technology, the combination of head-mounted displays and augmented reality (AR) technology has shown great potential for health monitoring and biomedical diagnosis applications. However, further optimizing its performance and improving data interaction accuracy remain crucial issues that must be addressed. In this study, we develop smart glasses based on augmented reality and eye tracking technology. Through real-time information interaction with the server, the smart glasses realize accurate scene perception and analysis of the user's intention and combine with mixed-reality display technology to provide dynamic and real-time intelligent interaction services. A multi-level hardware architecture and optimized data processing process are adopted during the research process to enhance the system's real-time accuracy. Meanwhile, combining the deep learning method with the geometric model significantly improves the system's ability to perceive user behavior and environmental information in complex environments. The experimental results show that when the distance between the subject and the display is 1 m, the eye tracking accuracy of the smart glasses can reach 1.0° with an error of no more than ±0.1°. This study demonstrates that the effective integration of AR and eye tracking technology dramatically improves the functional performance of smart glasses in multiple scenarios. Future research will further optimize smart glasses' algorithms and hardware performance, enhance their application potential in daily health monitoring and medical diagnosis, and provide more possibilities for the innovative development of wearable devices in medical and health management.

Interpretable Mixture of Experts for Decomposition Network on Server Performance Metrics Forecasting

The accurate forecasting of server performance metrics, such as CPU utilization, memory usage, and network bandwidth, is critical for optimizing resource allocation and ensuring system reliability in large-scale computing environments. In this paper, we introduce the Mixture of Experts for Decomposition Kolmogorov–Arnold Network (MOE-KAN), a novel approach designed to improve both the accuracy and interpretability of server performance prediction. The MOE-KAN framework employs a decomposition strategy that breaks down complex, nonlinear server performance patterns into simpler, more interpretable components, facilitating a clearer understanding of how predictions are made. By leveraging a Mixture of Experts (MOE) model, trend and residual components are learned by specialized experts, whose outputs are transparently combined to form the final prediction. The Kolmogorov–Arnold Network further enhances the model’s ability to capture intricate input–output relationships while maintaining transparency in its decision-making process. Experimental results on real-world server performance datasets demonstrate that MOE-KAN not only outperforms traditional models in terms of accuracy but also provides a more trustworthy and interpretable forecasting framework. This makes it particularly suitable for real-time server management and capacity planning, offering both reliability and interpretability in predictive models.

IoT-HGDS: Internet of Things integrated machine learning based hazardous gases detection system for smart kitchen

This paper proposes an Internet of Things (IoT) and Machine Learning (ML) integrated Hazardous Gas Detection System (IoT-HGDS) for smart kitchens. The design incorporates six tin-oxide-based metal–oxide–semiconductor (MOS) gas sensor arrays and one DHT22 (temperature & humidity sensor). This IoT-HGDS can detect different hazardous gases, Volatile Organic Compounds (VOCs), and odors responses released from the kitchen’s materials and transmit them to a Remote Data Processing Centre (RDPC) through Amazon-Web Services (AWS) in real time. In this experiment, we collected 150×9=1350 samples from 9 kitchen materials like ghee, milk, liquid petroleum gas (LPG), bread, mustard oil, compressed natural gas (CNG), pigeon peas, refined oil, and kerosene. The Standardized Independent Component Analysis (SICA) pre-processing technique has been used to clean data, standardize the features, and remove outliers. ML approaches like Logistic Regression (LR), Adaptive Boosting (AdaBoost) and Regularized Discriminant Analysis (RDA) have been applied for accurate identification of gases/VOCs class and provide immediate alerts to improve kitchen safety. The SICA-RDA classifier outperformed (highest accuracy at 97.78 %) as compared to LR and AdaBoost in terms of performance and balanced precision, recall, and F1-Score. LR has the lowest performance in all metrics. LPG has the lowest Mean Squared Error (MSE) of 6.62×10−7, while CNG has the highest MSE of 3.60×10−4. This system can intelligently preserve gases, ensure safety precautions, and prevent accidents in the kitchens.

Hybrid vehicular access protocol and message prioritization for real-time safety messaging

Real-time safety services rely on the exchange of messages to enhance the operations of connected and automated vehicles (CAVs). These safety messages convey vital information about traffic conditions, enabling drivers to take necessary measures to prevent accidents. The timely and reliable delivery of these messages is essential, necessitating efficient channel access. Vehicular Deterministic Access (VDA) is employed as a channel access scheme with distinct priorities and stringent timing guidelines, particularly for urgent safety warnings. In this paper, we propose a hybrid approach that combines VDA and Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) protocols, along with a message prioritization algorithm, to ensure efficient and reliable communication of safety messages in vehicular networks. Our approach leverages the strengths of both VDA and CSMA/CA to avoid message collisions; VDA is more efficient under high traffic loads, while CSMA/CA is better suited for low traffic loads. Additionally, the incorporation of the message prioritization algorithm ensures strict deadline guarantees for high-priority messages, such as Decentralized Environmental Notification Messages (DENMs). We evaluate our proposed solution using the Artery simulation framework. Our results show over a 93% delivery rate for DENM exchanges while maintaining low collision probability across various traffic loads. This research provides practical guidance for the development of efficient and reliable communication systems for CAVs. It also offers a detailed analysis of the trade-offs among different access protocols and message prioritization strategies in vehicular networks.

Application of Real-Time Data Service Technology in Smart oilfield

Abstract Real time data services play an important role in the construction of Smart oilfields, which can help oilfield production managers better understand the production situation and make more accurate decisions. Real time data services based on the Data Lake platform will bring more efficient and safer production methods to oil and gas fields, promoting the upgrading and sustainable development of the Smart oilfield industry. The technical architecture of Smart oilfield real-time data service mainly includes data lake platform, big data processing, edge computing and other core technologies. The data lake platform provides infrastructure, platform services and application services for Smart oilfield. Big data processing technology is responsible for rapid and accurate analysis of massive data. edge computing refers to computing at the data source to reduce data transmission and improve computing efficiency. The key technologies include data collection, transmission, storage, analysis, etc. Among them, data collection mainly uses sensors, PLCs and other equipment to collect production data. Data transmission uses technologies such as the Internet of Things and industrial Ethernet to transmit data to the data lake platform. Data storage adopts distributed storage technology to ensure the safety and reliability of data. Data analysis uses machine learning, deep learning and other technologies to analyze data, Provide decision support.

텍스트 분할을 적용한 KoBART 기반의 실시간 장문 뉴스 요약 시스템 구현

In this study, a real-timelong-news summarization system is implemented based on the model. Owing to its characteristics, the KoBART model cannot summarize news with a token length of 1024 or more. Hence, we implemented a method of dividing long news into paragraphs, summarizing the divided paragraphs, and then resummarizing the summarized sentences. First, we evaluated the performance using an AI Hub dataset to validate the implemented two-stage summarization method. However, because the token length of most of the news provided in the AI Hub dataset is 1024 or less, we analyzed the performance for long news by applying the dataset provided by Hugging Face with a token length of 1024 or more. When summarizing long news with a token length of 1024 or more by dividing it into 512 paragraphs, the average Luge score is 33.99% and the runtime required for summarization is 0.8492 s. Therefore, we confirmed that the implemented long-news summarization system can provide real-time services, even for long news with a token length of 1024 or more.

Constructed encoded data based coded distributed DNN training for edge computing scenario

Deep learning in unmanned aerial vehicle (UAV) deployment encounters two problems: 1, One UAV may fail to store and execute large Deep Neural Network (DNN) model. 2, One UAV fails to accomplish real time services. One possible solution is the group of UAVs (nodes) collectively generate swarm intelligence in the form of edge computing scenario, namely in the distributed computing (DC) mode with clever arrangement, say Coded Distributed Computing (CDC). In CDC systems, the redundant computation introduced by linear coding can compensate stragglers. However, since linear property cannot pass the nonlinear activation function in Deep Neural Network (DNN) training, coding/decoding for CDC need to be applied layer by layer, which slows down the training. To avoid layer-by-layer coding/decoding, we propose a novel DNN training scheme based on constructing encoded data. This construction process lies before the training process (can be done before training without any impact on training efficiency). Based on both the original data and the newly constructed encoded data, the training phase can take advantage of the (n,k) property (Wait for the first k returned data) and hence improve the training speed. The training process does not require encoding/decoding operations, and hence significantly improves the training speed. Experimental results show that the training scheme based on constructed encoded data can achieve prediction accuracy approximating that of the centralized one and significantly reduce the latency compared to the layer-by-layer linear encoding and decoding scheme.