Computational Capacity Research Articles

Probabilistic Protection for Both Computing and Transmission Capacities of Virtual Networks Under Multiple Facility Node Failures

Probabilistic Protection for Both Computing and Transmission Capacities of Virtual Networks Under Multiple Facility Node Failures

Intelligent detection method with 3D ranging for external force damage monitoring of power transmission lines

The rapid development of smart grid requires more and more reliable power supply. The external force damage to transmission lines can significantly impact the safety and stability of the power grid, potentially causing electric shock accidents. Existing methods for monitoring external force damage in transmission corridors lack the ability to provide effective warnings based on the real-time distance between potential hazards and power lines. Additionally, limited computational power and storage capacity at edge terminals restrict the efficient deployment of high-precision (high complexity) visual algorithms. This study presents, for the first time, a lightweight intelligent detection method integrating detection and three-dimensional (3D) ranging. A regression loss optimized for small objects is introduced to compensate for the shortcomings of lightweight networks in detection accuracy. Simultaneously, based on the influence of convolutions in various modules of the baseline model on performance, lightweight improvements are made to the detector architecture using Omni-Dimensional Dynamic Convolution and Distribution Shifting Convolution. Finally, a 3D ranging module is integrated into the detector, involving operations such as 2D3D information matching and back-projection transformation. This method innovatively achieves automated ranging and hierarchical warning. Its effectiveness is validated in transmission corridor scenarios under various weather conditions and surveillance video. The results demonstrate that our method outperforms other algorithms in hazard detection accuracy and lightweight performance. Moreover, the distance prediction error rate is below 1.6%. The hierarchical warning solutions can be applied to more scenarios.

Experimental backward integration for state-dependent differential Riccati equation (SDDRE): A case study on flapping-wing flying robot

Backward integration (BI) is a classical approach for solving optimal control arising from linear quadratic regulator (LQR) design in differential form. It proposes a two-round solution, the first one starting from a final boundary condition to the initial one to generate the optimal gain, and the next one, solving the control system in a forward loop. Implementation of the BI on the nonlinear optimal control, the state-dependent differential Riccati equation (SDDRE), is a challenge since in the backward loop, the state information is missing and is not straightforward like the LQR. Hence, a control for backward motion is required to regulate the system from the terminal to the initial desired states. While there have been some valuable works on the theoretical implementation of the BI in simulations for the SDDRE, this approach has not been reported in experiments for the best knowledge of the authors. Here in this work, the SDDRE is solved using the BI and two backward and forward solution steps, and it is experimentally applied for a flapping-wing flying robot (FWFR). The execution of the control system is done onboard using Raspberry Pi 4B as the processor which has limited computational capacity. The trajectory tracking of a line in a closed limited space is proposed for the FWFR flight. The objective is to position the robot bird at the corner of the rectangular limited space with minimum error in translation. The results have been presented for 21 flights to show the repeatability and presented the best-case minimum error of 10 (cm) at the end of the trajectory, in the YZ-plane. Considering approximately 12 (m) flight path, the error is found less than 1 percent of the travel distance. The results were compared with forward integration to confirm the correctness of the computation. The experimental flight dataset, MATLAB simulation codes, and experimental Python codes are available as supplementary material for this work in the online version of the paper.

KLD: a program to elucidate the localization of the Fermi and Coulomb holes in molecular systems.

The electron localization is a concept that allows scientists to better understand the physical and chemical properties of electronic systems. It is associated with the propensity of electron pairs with opposite spins to accumulate as well as with their response to external perturbations. This paper contains a detailed description of the design and implementation of the program KLD, which was primarily developed in our research group to elucidate electron localization in molecular systems by evaluating the information content of electron-pair density functions. KLD employs two information-based functions as a real space measure of the Fermi and Coulomb holes for same-spin electrons and shows a better resolution as compared to other methods (i.e., ELF). Information about the acceleration of the code is also included in the present work, being noticeable the reduction of wall-time calculation and the error calculation between versions. KLD was designed to be easy to use, extend, and maintain; thus, many principles of modern software development, extensive testing, and package management were adopted. The latest version of the KLD program was created utilizing the Compute Unified Device Architecture (CUDA) version, which allows it to use the computational capacity of NVIDIA Graphics Processing Units (GPUs) for processing purposes. The electron-pair conditional density was calculated from the canonical molecular orbitals obtained at the HF/6-31G(2df,p) level, or alternatively the natural orbitals in the case of explicit correlated wavefunctions computed at the MP2/6-31G(2df,p)//HF/6-31G(2df,p) level.

A novel ECG compression algorithm using Pulse-Width Modulation integrated quantization for low-power real-time monitoring

Cardiac monitoring systems in Internet of Things (IoT) healthcare, reliant on limited battery and computational capacity, need efficient local processing and wireless transmission for comprehensive analysis. Due to the power-intensive wireless transmission in IoT devices, ECG signal compression is essential to minimize data transfer. This paper presents a real-time, low-complexity algorithm for compressing electrocardiogram (ECG) signals. The algorithm uses just nine arithmetic operations per ECG sample point, generating a hybrid Pulse Width Modulation (PWM) signal storable in a compact 4-bit resolution format. Despite its simplicity, it performs comparably to existing methods in terms of Percentage Root-Mean-Square Difference (PRD) and space-saving while significantly reducing complexity and maintaining robustness against signal noise. It achieves an average Bit Compression Ratio (BCR) of 4 and space savings of 90.4% for ECG signals in the MIT-BIH database, with a PRD of 0.33% and a Quality Score (QS) of 12. The reconstructed signal shows no adverse effects on QRS complex detection and heart rate variability, preserving both the signal amplitude and periodicity. This efficient method for transferring ECG data from wearable devices enables real-time cardiac activity monitoring with reduced data storage requirements. Its versatility suggests potential broader applications, extending to compression of various signal types beyond ECG.

Diagnostic analysis of cancer based on machine learning

Cancer, as the second leading cause of death in the world, caused an estimated 9.6 million deaths in 2018, accounting for one-sixth of all deaths. Early detection and early treatment is the best solution for cancer, and now, through machine learning methods, we can achieve accurate judgment of cancer so as to realize precise treatment and reduce the mortality rate of cancer. During this discussion, we will focus on the applications of machine learning methods to diagnose breast cancer, prostate cancer, oral cancer, which use machine learning methods including neural convolutional networks, K-clustering, support vector machine (SVM), and so on. As of now, machine learning has achieved better results than other methods, but due to the importance and complexity of cancer diagnosis and the cost of human computational capacity for diagnosis, we still hope to find a more accurate and effective method to realize the accurate judgment of cancer, and then introduce it into real-life applications.

Pre-sensor computing with compact multilayer optical neural network.

Moving computation units closer to sensors is becoming a promising approach to addressing bottlenecks in computing speed, power consumption, and data storage. Pre-sensor computing with optical neural networks (ONNs) allows extensive processing. However, the lack of nonlinear activation and dependence on laser input limits the computational capacity, practicality, and scalability. A compact and passive multilayer ONN (MONN) is proposed, which has two convolution layers and an inserted nonlinear layer, performing pre-sensor computations with designed passive masks and a quantum dot film for incoherent light. MONN has an optical length as short as 5 millimeters, two orders of magnitude smaller than state-of-the-art lens-based ONNs. MONN outperforms linear single-layer ONN across various vision tasks, off-loading up to 95% of computationally expensive operations into optics from electronics. Motivated by MONN, a paradigm is emerging for mobile vision, fulfilling the demands for practicality, miniaturization, and low power consumption.

Trust criteria for artificial intelligence in health: normative and epistemic considerations

Rapid advancements in artificial intelligence and machine learning (AI/ML) in healthcare raise pressing questions about how much users should trust AI/ML systems, particularly for high stakes clinical decision-making. Ensuring that...

Air computing: A survey on a new generation computation paradigm

There is an ever-growing race between what novel applications demand from the infrastructure and what the continuous technological breakthroughs bring in. Especially after the proliferation of smart devices and diverse IoT requirements, we observe the dominance of cutting-edge applications with ever-increased user expectations in terms of mobility, pervasiveness, and real-time response. Over the years, to meet the requirements of those applications, cloud computing provides the necessary capacity for computation, while edge computing ensures low latency. However, these two essential solutions would be insufficient for next-generation applications since computational and communicational bottlenecks are inevitable due to the highly dynamic load. On the other hand, inadequate infrastructure considering rural areas and disaster sites makes the utilization of those solutions difficult. Therefore, a 3D networking structure using different air layers including Low Altitude Platforms, High Altitude Platforms, and Low Earth Orbits in a harmonized manner for both urban and rural areas should be applied to satisfy the requirements of the dynamic environment. In this perspective, we put forward a novel, next-generation paradigm called Air Computing that presents a dynamic, responsive, and high-resolution computation environment for all spectrum of applications. In this survey, we define the components of air computing, investigate its architecture in detail, and discuss its essential use cases and the advantages it brings for next-generation application scenarios. We provide a detailed and technical overview of the benefits and challenges of air computing as a novel paradigm and spot the important future research directions.

Online adaptation of fourier series-based acoustic transfer function model and its application to sound source localization and separation

In this paper, we propose an online adaptation method for Fourier series-based acoustic transfer function (FS-ATF) models for robot audition systems using microphone array signal processing. The ATF represents the characteristics of signal propagation from a sound source to a microphone, which is essential for sound source localization (SSL) and sound source separation (SSS). For practical applications in dynamically changing real-world environments, ATF models must satisfy two criteria: (1) they must be adaptable to changes in the acoustic environment; and (2) they must be lightweight to be suitable for resource-constrained systems, such as robots with limited memory and computational capacity. The proposed method addresses these challenges using Fourier series expansions for interpolation, which reduces the memory footprint of the ATF model and facilitates online adaptation to acoustic environmental changes. The experimental results demonstrate that the proposed online adaptation method both improves the SSL and SSS performance while reducing the size of the ATF model, which represents a significant improvement over existing online ATF adaptation methods.

Non-symbolic estimation of big and small ratios with accurate and noisy feedback

The ratio of two magnitudes can take one of two values depending on the order they are operated on: a ‘big’ ratio of the larger to smaller magnitude, or a ‘small’ ratio of the smaller to larger. Although big and small ratio scales have different metric properties and carry divergent predictions for perceptual comparison tasks, no psychophysical studies have directly compared them. Two experiments are reported in which subjects implicitly learned to compare pairs of brightnesses and line lengths by non-symbolic feedback based on the scaled big ratio, small ratio or difference of the magnitudes presented. Results of Experiment 1 showed all three operations were learned quickly and estimated with a high degree of accuracy that did not significantly differ across groups or between intensive and extensive modalities, though regressions on individual data suggested an overall predisposition towards differences. Experiment 2 tested whether subjects learned to estimate the operation trained or to associate stimulus pairs with correct responses. For each operation, Gaussian noise was added to the feedback that was constant for repetitions of each pair. For all subjects, coefficients for the added noise component were negative when entered in a regression model alongside the trained differences or ratios, and were statistically significant in 80% of individual cases. Thus, subjects learned to estimate the comparative operations and effectively ignored or suppressed the added noise. These results suggest the perceptual system is highly flexible in its capacity for non-symbolic computation, which may reflect a deeper connection between perceptual structure and mathematics.

Evaluating flood hazards in data-sparse coastal lowlands: highlighting the Ayeyarwady Delta (Myanmar)

Coastal lowlands and river deltas worldwide are increasingly exposed to coastal, pluvial and fluvial flooding as well as relative sea-level rise (RSLR). However, information about both single and multiple flood-type hazards, their potential impact and the characteristics of areas, population and assets at risk is often still limited as high-quality data either does not exist or is not accessible. This often constitutes a main barrier for generating sound assessments, especially for scientific and public communities in the so-called Global South. We provide a standardised, integrative approach for the first-order assessment of these single and multiple flood-type hazards and show how this can be conducted for data-sparse, hardly accessible and inaccessible coastal lowlands such as the Ayeyarwady Delta in Myanmar by using only open accessible and freely available datasets of satellite imagery, global precipitation estimates, satellite-based river discharge measurements, elevation, land use, and population data. More than 70% of the delta, mainly used for agriculture, and about 40% of its present population are prone to flooding due to either monsoon precipitation and runoff, storm surge, and RSLR, or their combination, jeopardising food security and economic development in the region. The approach allows for the integration and combination of various datasets, combined in a highly flexible workflow that performs at low computational capacities, supporting the evaluation of flood-prone areas on regional and local scale for data-sparse coastal lowlands worldwide. It thereby allows to attribute different types of flood hazards, complements concepts of vulnerability and risk, and supports risk-informed decision making and development of effective multi-flooding adaptation strategies.

Design of Area Efficient Network-On-Chip Router: A Comprehensive Review

The number of uses for cutting-edge technologies has led to a further growth in a single chip's computational capacity. In this case, several applications want to build on a single chip for computing resources. As a result, connecting the IP cores becomes yet another difficult chore. The many-core System-On-Chips (SoCs) are being replaced by Network-On-Chip (NoC) as an on-chip connectivity option. As a result, the Network on Chip was created as a cutting-edge framework for those networks inside the System on Chip. Modern multiprocessor architectures would benefit more from a NoC architecture as its communication backbone. The most important components of any network structure are its topologies, routing algorithms, and router architectures. NoCs use the routers on each node to route traffic. Circuit complexity, high critical path latency, resource usage, timing, and power efficiency are the primary shortcomings of conventional NoC router architecture. It has been difficult to build a high-performance, low-latency NoC with little area overhead. This paper surveys previous methods and strategies for NoC router topologies and study of general router architecture and its components. Analysis is carried out to understand and work for a low latency, low power consumption, and high performance NoC router design that can be employed with a wide range of FPGA families. In the current work, we are structuring a modified four port router with the goals of low area and high performance operation.

Enhancing mobile robot navigation: integrating reactive autonomy through deep learning and fuzzy behavior

Objective: This study aimed to develop a control architecture for reactive autonomous navigation of a mobile robot by integrating Deep Learning techniques and fuzzy behaviors based on traffic signal recognition. Materials: The research utilized transfer learning with the Inception V3 network as a base for training a neural network to identify traffic signals. The experiments were conducted using a Donkey-Car, an Ackermann-steering-type open-source mobile robot, with inherent computational limitations. Results: The implementation of the transfer learning technique yielded a satisfactory result, achieving a high accuracy of 96.2% in identifying traffic signals. However, challenges were encountered due to delays in frames per second (FPS) during testing tracks, attributed to the Raspberry Pi's limited computational capacity. Conclusions: By combining Deep Learning and fuzzy behaviors, the study demonstrated the effectiveness of the control architecture in enhancing the robot's autonomous navigation capabilities. The integration of pre-trained models and fuzzy logic provided adaptability and responsiveness to dynamic traffic scenarios. Future research could focus on optimizing system parameters and exploring applications in more complex environments to further advance autonomous robotics and artificial intelligence technologies.

A Placement Method of the 5G Edge Nodes Based on the Hotspot Distribution of Mobile Users

Due to the emergence of various new applications, such as short videos and online games, higher requirements of their computing and storage capacity are demanded of mobile networks. The traditional cloud computing paradigm has the shortcomings of large latency and high bandwidth demand of the core network. Therefore, how to mine the hotspot distribution of these applications and reasonably configure 5G edge nodes to reduce latency and core network bandwidth are facing great challenges. To address these issues, we designed a placement method for the 5G edge nodes based on mobile hotspots. In this method, we first cluster all locations from the user trajectories to obtain the cluster areas. Further, we extract the features, such as the number of users and duration time in all cluster areas, and extract the hotspots from all cluster areas based on the features of each cluster. Then, we introduce the base station’s high load utilization rate and the core network’s bandwidth reduction rate as the optimization parameters to construct the mathematical model of multi-objective optimization. Finally, we formalize the model into a 0–1 integer programming problem and design a greedy algorithm to solve this model. We also complete a series of experiments to evaluate our proposed methods using the GeoLife dataset. The experimental results show that the high load utilization rate can be increased up to 7.69%, and the bandwidth reduction rate of the core network can be improved up to 6.34%.

A Novel Data Analysis Architectural Framework for Data Reduction on Edges for Smart City Healthcare

The most difficult task is managing and digesting the massive amount of heterogeneous data created by IoT. It questioned the efficiency and computational storage capacity of today's infrastructure resources. Furthermore, because there is so much data, it becomes difficult to give the user enough information to make a decision. A technology known as Complex Event Processing (CEP) has been created to glean important insights from massive real-time data streams. The goal of this research is to create a system that can extract complicated events from large amounts of sensor data and only transmit those events to a cloud server. Global healthcare services have been expanding at an exponential rate because of the vast amount of data that clinical and medical organizations generate daily. Healthcare systems can use e-health services to meet the medical and assistive requirements of people. Edge technology is proven in this direction. Mixing the old classic concepts with the innovative we have proposed the Novel data analysis architecture for smart heart disease prediction. We have created the fuzzy-based inference system at the edges for the data analysis and used the resulting Complex Events at the end predictions.

Performance research of edge computing nodes based on heterogeneous environment

Abstract Edge computing realizes real-time parallel storage and computing in a distributed way, which can reduce the data transmission cost. The edge nodes, which are composed of different heterogeneous physical devices and hardware resources, assume the computing and storage capacity of the edge network. At present, the resource characteristics and performance of heterogeneous edge nodes are difficult to characterize. Therefore, a resource-aware performance feature detection model for heterogeneous edge compute nodes is proposed in this study. This model is based on a resource utilization matrix and resource relative performance quantization algorithm to extract the performance characteristics of the same type of resources of the edge nodes, to realize multi-dimensional and systematic feature detection and description of heterogeneous edge nodes. On this basis, a weight-based edge node task scheduling algorithm is designed. The final verification results show that the CPU relative performance ratios of Node 1-Node 2, Node 1-Node 3, and Node 2-Node 3 are 6.23, 3.81, and 0.81 respectively. The performance relationship among the three nodes is Node 1> Node 3> Node 2. The test group and the verification group tend to be consistent overall. This method has a good application prospect.

Deep learning techniques to process 3D chest CT

The idea of using X–rays and Computed Tomography (CT) images as diagnostic method has been explored in several studies. Most of these studies work with slices of CT image in 2D, requiring less computational capacity and less time to process them than 3D. The processing of volumetric data (the complete CT images in 3D) adds an extra dimension of information. However, the magnitude of the data is considerably larger than working with slices in 2D, so extra computational processing is required. In this study a model capable of performing a classification of a 3D input that represents the volume of the CT scan is proposed. The model is able to classify the 3D input between COVID–19 and Non–COVID–19, but reducing the use of resources when performing the classification. The proposed model is the ResNet–50 model with a new dimension of information added, which is a simple autoencoder. This autoencoder is trained on the same dataset, and a vector representation of each exam is generated and used together with the exams to feed the ResNet–50. To validate the proposal, the same proposed model is compared with and without the autoencoder module that provides more information to the proposed model. The proposed model obtains better metrics than the same model without the autoencoder, confirming that extracting relevant features from the dataset helps improve the performance of the model.

DNN acceleration in vehicle edge computing with mobility-awareness: A synergistic vehicle–edge and edge–edge framework

In recent years, vehicular networks have seen a proliferation of applications and services such as image tagging, lane detection, and speech recognition. Many of these applications rely on Deep Neural Networks (DNNs) and demand low-latency computation. To meet these requirements, Vehicular Edge Computing (VEC) has been introduced to augment the abundant computation capacity of vehicular networks to complement limited computation resources on vehicles. Nevertheless, offloading DNN tasks to MEC (Multi-access Edge Computing) servers effectively and efficiently remains a challenging topic due to the dynamic nature of vehicular mobility and varying loads on the servers. In this paper, we propose a novel and efficient distributed DNN Partitioning And Offloading (DPAO), leveraging the mobility of vehicles and the synergy between vehicle–edge and edge–edge computing. We exploit the variations in both computation time and output data size across different layers of DNN to make optimized decisions for accelerating DNN computations while reducing the transmission time of intermediate data. In the meantime, we dynamically partition and offload tasks between MEC servers based on their load differences. We have conducted extensive simulations and testbed experiments to demonstrate the effectiveness of DPAO. The evaluation results show that, compared to offloaded all tasks to MEC server, DPAO reduces the latency of DNN tasks by 2.4x. DPAO with queue reservation can further reduce the task average completion time by 10%.

Multi-species identification and number counting of fish passing through fishway at hydropower stations with LigTraNet

Fishway monitoring can verify the effectiveness of the fishway, optimise the operation mode, and achieve scientific management of fishway operations. Traditional fishway monitoring approaches, hindered by their inefficiency and substantial disruption of fish, are ill-suited for long-term surveillance; thus, employing video monitoring coupled with object detection technology presents an alternative or complementary solution. However, challenges such as the constrained computational capacity of onsite equipment in fishways, complexities involved in model deployment, and sluggish pace of detection are significant hurdles. In this study, by utilising the YOLOv8n model as a benchmark, we engineered a cross-stage partial module with a single convolution (C1) module to replace the existing C2f module with the aim of enhancing performance. We replaced the conventional 2D convolutions in the bottleneck configuration with depthwise separable convolutions and integrated the SimAM module to extract the detailed characteristics of the fish species. By amalgamating LigObNet detection with the DeepSORT algorithm, we established LigTraNet, which is designed to enable precise tracking, identification, and counting of individual fish. The results showed that LigObNet exhibited the lowest complexity and fastest detection speed for underwater fish among similar object recognition and detection models. Compared with the benchmark YOLOv8n model, there were reductions of 8.9% in the network layers, 40.5% in the parameter count, 39.3% in the memory footprint, and 35.8% in the giga floating-point operations and a 38.1% improvement in the inference speed. LigTraNet achieved a total count accuracy rate of 91.8%, demonstrating superior species quantification capabilities over other models with minimal resource usage and rapid inference capabilities, thus offering enhanced practicality for deployment on devices in real-world engineering contexts. This represents a departure from traditional manual monitoring methods for assessing fishway effectiveness, revolutionising aquatic ecological monitoring tools and methodologies and fostering the collaborative advancement of water resource project operations and ecological conservation.