Operational Network Research Articles

Innovative Deep Learning Architecture for the Classification of Lung and Colon Cancer From Histopathology Images

The increasing prevalence of colon and lung cancer presents a considerable challenge to healthcare systems worldwide, emphasizing the critical necessity for early and accurate diagnosis to enhance patient outcomes. The precision of diagnosis heavily relies on the expertise of histopathologists, constituting a demanding task. The health and well‐being of patients are jeopardized in the absence of adequately trained histopathologists, potentially leading to misdiagnoses, unnecessary treatments, and tests, resulting in the inefficient utilization of healthcare resources. However, with substantial technological advancements, deep learning (DL) has emerged as a potent tool in clinical settings, particularly in the realm of medical imaging. This study leveraged the LC25000 dataset, encompassing 25,000 images of lung and colon tissue, introducing an innovative approach by employing a self‐organized operational neural network (Self‐ONN) to accurately detect lung and colon cancer in histopathology images. Subsequently, our novel model underwent comparison with five pretrained convolutional neural network (CNN) models: MobileNetV2‐SelfMLP, Resnet18‐SelfMLP, DenseNet201‐SelfMLP, InceptionV3‐SelfMLP, and MobileViTv2_200‐SelfMLP, where each multilayer perceptron (MLP) was replaced with Self‐MLP. The models’ performance was meticulously assessed using key metrics such as precision, recall, F1 score, accuracy, and area under the receiver operating characteristic (ROC) curve. The proposed model demonstrated exceptional overall accuracy, precision, sensitivity, F1 score, and specificity, achieving 99.74%, 99.74%, 99.74%, 99.74%, and 99.94%, respectively. This underscores the potential of artificial intelligence (AI) to significantly enhance diagnostic precision within clinical settings, portraying a promising avenue for improving patient care and outcomes. The synopsis of the literature provides a thorough examination of several DL and digital image processing methods used in the identification of cancer, with a primary emphasis on lung and colon cancer. The experiments use the LC25000 dataset, which consists of 25,000 photos, for the purposes of training and testing. Various techniques, such as CNNs, transfer learning, ensemble models, and lightweight DL architectures, have been used to accomplish accurate categorization of cancer tissue. Various investigations regularly show exceptional performance, with accuracy rates ranging from 96.19% to 99.97%. DL models such as EfficientNetV2, DHS‐CapsNet, and CNN‐based architectures such as VGG16 and GoogleNet variations have shown remarkable performance in obtaining high levels of accuracy. In addition, methods such as SSL and lightweight DL models provide encouraging outcomes in effectively managing large datasets. In general, the research emphasizes the efficacy of DL methods in successfully diagnosing cancer from histopathological pictures. It therefore indicates that DL has the potential to greatly improve medical diagnostic techniques.

Artificial Intelligence: Real Challenge or Boon for Network Operation Center and Network security

Artificial intelligence has emerged as a game-changer in every domain, and network operation center and network security are no exception. This paper critically evaluates the dual impact of AI as a daunting challenge and a blissful boon in the context of NOCs, network security, and their overlap. As firms increasingly rely on ever more sophisticated and complex networks to run their business, the need for effective network management and security has become a top priority. Correspondingly, controlling such complex networks, an analyzing their performance and status, and safeguarding them from rapidly evolving threats have grown more difficult than ever. AI-powered solutions have the potential to considerably improve NOC efficiency and effectiveness. Introduce automated automation of routine processes, facilitate network performance optimization, and help in forecasting of most network abnormalities. Consequently, these can drastically cut NOC operational costs and network uninsured losses and enhance network dependability. However, and exposure of AI technologies to an entire new category of challenges, such as the potential of AI-based cyberattacks, algorithmic discrimination, and the ethical insinuations of fully autonomous NOCs, also imply that many possible risks and risks are taking into account.

A Data-driven Base Station Sleeping Strategy Based on Traffic Prediction

Due to the rapidly increasing number of base stations (BSs) in the operational cellular networks, their energy consumption is escalating. In this paper, we propose an intelligent data-driven BS sleeping mechanism relying on a wireless traffic prediction model that measures the BSs' capacity in different regions. Firstly, a spatio-temporal cellular traffic prediction model is proposed, where a multi-graph convolutional network (MGCN) is developed to capture the associated spatial features. Furthermore, a multi-channel long short-term memory (LSTM) solution involving hourly, daily and weekly periodic data is used to capture the relevant temporal features. Secondly, the capacities of macro-cell BSs (MBSs) and small-cell BSs (SBSs) having different environment characteristics are modeled, where both clustering and transfer learning algorithms are adopted for quantifying the traffic supported by the MBSs and SBSs. Finally, an optimal BS sleeping strategy is proposed for minimizing the network's power consumption. Experimental results show that the proposed MGCN-LSTM model outperforms the existing models in terms of its cellular traffic prediction accuracy, and the proposed BS sleeping strategy using an approximated non-linear model of the associated capacity function achieves near-maximal energy-saving at a modest complexity.

A Novel Multi-Head Self-Organized Operational Neural Network Architecture for Chronic Obstructive Pulmonary Disease Detection Using Lung Sounds

A Novel Multi-Head Self-Organized Operational Neural Network Architecture for Chronic Obstructive Pulmonary Disease Detection Using Lung Sounds

Development of a stacked hybrid Decision Tree model leveraging the NSL-KDD dataset

Intrusion detection in information technology as well as operational technology networks is highly required in modern day systems due to the increased spate of cyber-attacks in both number and complexity. Anomaly-based intrusion detection systems which have the capacity to detect novel or zero-day attacks are highly employed in this regard. One important component of anomaly-based intrusion detection systems which ensures their behaviour is artificial intelligence in general and machine learning in particular. The burden in modern day cybersecurity research is to investigate and develop models that can outperform existing ones. This paper is aimed at developing a hybrid decision tree model using the stacking ensemble approach. Performances were measured on the basis of recall, precision, accuracy, F1-score, receiver operating characteristics and confusion matrices. The hybrid model presented a precision of 97%, accuracy of 81%, F1-score of 80% and AUC score of 0.96, respectively.

Comparative Evaluation of VAEs, VAE-GANs and AAEs for Anomaly Detection in Network Intrusion Data

With cyberattacks growing in frequency and sophistication, effective anomaly detection is critical for securing networks and systems. This study provides a comparative evaluation of deep generative models for detecting anomalies in network intrusion data. The key objective is to determine the most accurate model architecture. Variational autoencoders (VAEs), VAE-GANs, and adversarial autoencoders (AAEs) are tested on the NSL-KDD dataset containing normal traffic and different attack types. Results show that AAEs significantly outperform VAEs and VAE-GANs, achieving AUC scores up to 0.96 and F1 scores of 0.76 on novel attacks. The adversarial regularization of AAEs enables superior generalization capabilities compared to standard VAEs. VAE-GANs exhibit better accuracy than VAEs, demonstrating the benefits of adversarial training. However, VAE-GANs have higher computational requirements. The findings provide strong evidence that AAEs are the most effective deep anomaly detection technique for intrusion detection systems. This study delivers novel insights into optimizing deep learning architectures for cyber defense. The comparative evaluation methodology and results will aid researchers and practitioners in selecting appropriate models for operational network security.

Maritime Communications—Current State and the Future Potential with SDN and SDR

The rise of the Internet of Things (IoT) has opened up exciting possibilities for new applications. One such novel application is the modernization of maritime communications. Effective maritime communication is vital for ensuring the safety of crew members, vessels, and cargo. The maritime industry is responsible for the transportation of a significant portion of global trade, and as such, the efficient and secure transfer of information is essential to maintain the flow of goods and services. With the increasing complexity of maritime operations, technological advancements such as unmanned aerial vehicles (UAVs), autonomous underwater vehicles (AUVs), and the Internet of Ships (IoS) have been introduced to enhance communication and operational efficiency. However, these technologies also bring new challenges in terms of security and network management. Compromised IT systems, with escalated privileges, can potentially enable easy and ready access to operational technology (OT) systems and networks with the same privileges, with an increased risk of zero-day attacks. In this paper, we first provide a review of the current state and modalities of maritime communications. We then review the current adoption of software-defined radios (SDRs) and software-defined networks (SDNs) in the maritime industry and evaluate their impact as maritime IoT enablers. Finally, as a key contribution of this paper, we propose a unified SDN–SDR-driven cross-layer communications framework that leverages the existing SATCOM communications infrastructure, for improved and resilient maritime communications in highly dynamic and resource-constrained environments.

Drainage ditch network extraction from lidar data using deep convolutional neural networks in a low relief landscape

Drainage networks composed of small, channelized ditches are very common in the eastern United States. These are human-made features commonly constructed for wetland drainage and constitute the headwater portion of permanent hydrographic networks. Accurate information on the drainage ditch location can help define where wetlands have been drained and evaluate impacts of artificial drainage patterns on hydrologic changes. Traditional water channel extraction approaches often cannot accurately identify small ditches especially in low-relief agricultural landscapes. In this study, we employed a state-of-the-art deep learning (DL) approach to extract drainage ditches using light detection and ranging (lidar) data in a low-relief agricultural landscape within the Delmarva area. First, we adopted a deep convolutional neural network based on U-Net architecture to classify ditches from different combinations of aerial optical and lidar derived (i.e., topographic and return intensity) features. The classification results were compared with a typical pixel-oriented machine learning classifier, random forest (RF). Next, we improved the connectivity of ditch networks through a minimum-cost approach and a further incorporation of FA to connect with natural drainage networks. Finally, we evaluated the connected drainage networks against flowlines derived from typical flow routing method (D8), an open-source channel network extraction tool (GeoNet), and the U.S. Geological Survey National Hydrography Dataset High Resolution data at 1:24,000 scale. Our results show that the DL model significantly outperformed the RF model, and the lidar derived topographic features were the most important input for ditch classification. The connected drainage networks extracted with DL exhibited pronouncedly higher precision (0.88) and recall (0.89) and a higher positional accuracy (within one pixel) than other flowline products. Overall, this study demonstrates the utility of DL approaches for automated extraction of ditch networks and the important contribution of lidar-derived topographic data for operational drainage network mapping at local and regional scales.

Mechanism of Regulation on the Marketing of the Breast–Milk Substitutes: a Lesson Learn from one Province in Northern of Thailand

Objectives: To investigate the existing model and mechanism for surveillance of the marketing of food for infants and young children in Phayao province, Thailand according to The Control of Marketing of Infants and Young Children Food Act, B.E. 2560 (2017). Method: This research was a qualitative research study which collected data through an in-depth interview and a focus group with 7 informants: officers of Phayao Provincial Public Health Office, registered nurses, public health technical officers, and consumer network officers. Results: Phayao province had a control model and mechanism on the marketing of food for infants and young children according to The Control of Marketing of Infants and Young Children Food Act, B.E. 2560 (2017) as follows: (1) For passive surveillance, there was a complaint center at the Provincial Public Health Office as the main coordinator to receive complaints and prepare files of violation cases. (2) Active surveillance included visits to target places such as shops and child development centers, and visits to enforce maternal and child health standards. Finally, (3) networking and knowledge dissemination were implemented through the first 1,000 miracle days of life project. The implementational model for controlling the marketing of food for infants and young children should be as follows: (1) Be compliant with laws. (2) Determine clear policies at the departmental levels. (3) Work collaboratively at the provincial, district, and subdistrict levels. (4) Increase participation of related people for surveillance of offences. (5) Empower operational networks, entrepreneurs, and other people to gain knowledge and be compliant with the law in a suitable way. (6) Use communication technology to support implementation. Conclusion: Phayao province has a model and mechanism to control the marketing of food for infants and young children in compliance with the law. The model and mechanism should be developed according to the area contexts to support the implementation to be more effective.

Mortality analysed by propensity score matching: an application to national neonatal audit

Mortality, a key outcome variable in many population studies and studies of healthcare and its interventions, is commonly analysed by regression of the survival status on a set of relevant background variables. We describe an alternative based on the potential outcomes framework, in which we ask how a particular group of subjects, or a population, whose outcomes were realised in one condition, would have fared had they been treated or cared for in different circumstances. The method is applied to neonatal mortality in the operational delivery networks in England and Wales. The performance of a network is assessed by the difference of the mortality rates of the network and of a matched set of babies drawn from the entire domain of the study. The outlier status of a network is established by a decision-theoretical approach.

Self-Organized Operational Neural Networks for The Detection of Atrial Fibrillation

Abstract Atrial fibrillation is a common cardiac arrhythmia, and its incidence increases with age. Currently, numerous deep learning methods have been proposed for AF detection. However, these methods either have complex structures or poor robustness. Given the evidence from recent studies, it is not surprising to observe the limitations in the learning performance of these approaches. This can be attributed to their strictly homogenous conguration, which solely relies on the linear neuron model. The limitations mentioned above have been addressed by operational neural networks (ONNs). These networks employ a heterogeneous network configuration, incorporating neurons equipped with diverse nonlinear operators. Therefore, in this study, to enhance the detection performance while maintaining computational efficiency, a novel model named multi-scale Self-ONNs (MSSelf-ONNs) was proposed to identify AF. The proposed model possesses a significant advantage and superiority over conventional ONNs due to their self-organization capability. Unlike conventional ONNs, MSSelf -ONNs eliminate the need for prior operator search within the operator set library to find the optimal set of operators. This unique characteristic sets MSSelf -ONNs apart and enhances their overall performance. To validate and evaluate the system, we have implemented the experiments on the well-known MIT-BIH atrial fibrillation database. The proposed model yields total accuracies and kappa coefficients of 98% and 0.95, respectively. The experiment results demonstrate that the proposed model outperform the state-of-the-art deep CNN in terms of both performance and computational complexity.

Designing electricity distribution networks: The impact of demand coincidence

With the global effort to reduce carbon emissions, clean technologies such as electric vehicles and heat pumps are increasingly introduced into electricity distribution networks. These technologies considerably increase electricity flows and can lead to more coincident electricity demand. In this paper, we analyze how such increases in demand coincidence impact future distribution network investments. For this purpose, we develop a novel model for designing electricity distribution networks, called the distribution network reconfiguration problem with line-specific demand coincidence (DNRP-LSDC). Our analysis is two-fold: (1) We apply our model to a large sample of real-world networks from a Swiss distribution network operator. We find that a high demand coincidence due to, for example, a large-scale uptake of electric vehicles, requires a substantial amount of new network line construction and increases average network cost by 84% in comparison to the status quo. (2) We use a set of synthetic networks to isolate the effect of specific network characteristics. Here, we show that high coincidence has a more detrimental effect on large networks and on networks with low geographic consumer densities, as present in, e.g., rural areas. We also show that expansion measures are robust to variations in the cost parameters. Our results demonstrate the necessity of designing policies and operational protocols that reduce demand coincidence. Moreover, our findings show that operators of distribution networks must consider the demand coincidence of new electricity uses and adapt investment budgets accordingly. Here, our solution algorithms for the DNRP-LSDC problem can support operators of distribution networks in strategic and operational network design tasks.

Modeling the transition of the multimodal pan-European energy system including an integrated analysis of electricity and gas transport

AbstractMost recently, the European energy system has undergone a fundamental transformation to meet decarbonization targets without compromising the security of the energy supply. The transition involves several energy-generating and consuming sectors emphasizing sector coupling. The increase in the share of renewable energy sources has revealed the need for flexibility in supporting the electricity grid to cope with the resulting high degree of uncertainty. The new technologies accompanying the energy system transition and the recent political crisis in Europe threatening the security of the energy supply have invalidated the experience from the past by drastically changing the conventional scenarios. Hence, supporting strategic planning tools with detailed operational energy network models with appropriate mathematical precision has become more important than ever to understand the impacts of these disruptive changes. In this paper, we propose a workflow to investigate optimal energy transition pathways considering sector coupling. This workflow involves an integrated operational analysis of the electricity market, its transmission grid, and the gas grid in high spatio-temporal resolution. Thus, the workflow enables decision-makers to evaluate the reliability of high-level models even in case of disruptive events. We demonstrate the capabilities of the proposed workflow using results from a pan-European case study. The case study, spanning 2020–2050, illustrates that feasible potential pathways to carbon neutrality are heavily influenced by political and technological constraints. Through integrated operational analysis, we identify scenarios where strategic decisions become costly or infeasible given the existing electricity and gas networks.

Design of a multiparameter data acquisition and control system for in situ seabed observation base stations

AbstractWith the exploration, development, and research of deep‐sea resources, there is an urgent need for long‐term and continuous observation data of the deep‐sea seabed boundary layer. The traditional method of deep‐sea seabed survey and sampling based on scientific research vessels has the discontinuity of observation data in space and time scales. There are some problems in the seabed in situ observation method based on the seabed observation network for low mobility and high operation and maintenance costs, restricting the in‐depth understanding of the dynamic change process of the deep‐sea floor. To solve the above problems, an open and modular data acquisition control system was designed based on an embedded system and signal processing technology. In terms of the physical, chemical, geological, and ecosystem characteristics of the seafloor or near the seafloor boundary layer, various functional sensors and instrumentation were matched to form an independent underwater integrated measurement or experimental device, eventually realizing in situ multiparameter and long‐time series observations of the seafloor. The system data acquisition and control test were completed through laboratory experiments, which verified the feasibility of the system design. The research showed important theoretical and technical reference significance for the exploration and development of resources in the submarine boundary layer and the promotion of deep‐sea scientific research.

A Stackelberg game-based dynamic pricing and robust optimization strategy for microgrid operations

Energy management algorithms of microgrids have attracted growing attention for the efficient operation and coordination of active end-users. In developing these algorithms, leader–follower game approaches could provide effective strategies when the behavioral and operational characteristics of active end-users are integrated into the methodology. However, since these approaches require the exact implementation of the strategy determined by the leader, it is of great importance to deal with the uncertainty of renewable energy generation, charging/discharging energy of storage systems, and load consumption. Thus, possible negative effects of these uncertainties on the outcome can be minimized with an appropriate billing procedure. In this paper, a Stackelberg game approach is proposed for the energy sharing management of a microgrid including prosumers and plug-in electric vehicle (PEV) charging stations (CSs), and a proper billing scheme is designed to increase the robustness of the approach on real-time applications. In the hereby game-theoretic approach, the energy demands of PEV CSs are determined as variable demands considering the operational characteristics of CSs. The energy planning of CSs is based on determining the number of PEVs that each station can accept over the vehicle charging levels for a specified time interval. Also, by the necessary modifications identified for the presented approach, the microgrid can operate independently from the utility grid and satisfy the requirements of islanded mode for the time slots in which the microgrid exchanges energy with the grid based on the end-game results. The energy management strategy proposed in this study presents a comprehensive solution capable of optimizing energy sharing in microgrids while taking into account end-users’ operational characteristics and network requirements. Furthermore, the designed billing scheme ensures that the accuracy of the proposed game approach remains unaffected by the uncertainties, thus preserving the microgrid’s profit.

Pipeline rehabilitation combined strategy for urban water systems

ABSTRACT This paper presents a novel decision support framework for water utilities to automatically create yearly pipeline rehabilitation plans while considering multiple parameters, including pipeline physical properties and other factors like transportation and urban space. The average age of urban water pipelines is growing, and the annual rehabilitation volumes are often lower than needed to improve or even preserve the level of service quality. The existing rehabilitation strategies range from simple pipe-age-based approaches to complex multiparameter strategies. The needed input of specific and detailed data, which are usually not available for water utilities, impedes a straightforward implementation of the latter strategies. Therefore, this study presents a two-step pipe rehabilitation strategy that enables the utility to automatically prioritise the pipes and related street areas/sections based on condition ranking. The strategy was validated on an operational water network in Tallinn, Estonia.

Carbon emissions impacts of operational network constraints: The case of Spain during the Covid-19 crisis

Operating a highly decarbonized power system is technically complex and introduces novel challenges for system operators. To this effect, the hourly day-ahead market schedule must be compliant with all the network security criteria. If required, specific generators should be curtailed or started up after the market clearing via so-called redispatching actions. In this paper, we analyze the bias in emissions intensity of electricity generation that result from not internalizing grid operational limitations in the day-ahead market clearing. In other words, we investigate the incremental emissions resulting from actions by system operators to make the day-ahead market schedule physically feasible. We use hourly data from the Spanish power system between 2019 and 2021. We estimate that while redispatching actions accounts for 2–4% of total annual electricity demand, they represent 6–11% of the annual power sector's CO2 emissions. We find that volumes of fossil generators started up for network security reasons by system operators increase in hours during which the share of renewables in the supply mix is relatively high but, additionally, show that volumes also significantly increase during hours with low energy demand, as during the Covid-19 crisis. These latter actions are not triggered to alleviate grid bottlenecks, but to solve location-specific operational constraints requiring a minimum volume of synchronous generators always running. Because of these actions, we estimate that the emissions from the day-ahead market schedule are downward biased between +0.00391 and + 0.0145 tn of CO2 for each additional MWh of scheduled wind or photovoltaics. We offer several recommendations to better align the day-ahead schedule with the grid operation needs leading to a reduced need for redispatching actions.

Robust Peak Detection for Holter ECGs by Self-Organized Operational Neural Networks.

Although numerous R-peak detectors have been proposed in the literature, their robustness and performance levels may significantly deteriorate in low-quality and noisy signals acquired from mobile electrocardiogram (ECG) sensors, such as Holter monitors. Recently, this issue has been addressed by deep 1-D convolutional neural networks (CNNs) that have achieved state-of-the-art performance levels in Holter monitors; however, they pose a high complexity level that requires special parallelized hardware setup for real-time processing. On the other hand, their performance deteriorates when a compact network configuration is used instead. This is an expected outcome as recent studies have demonstrated that the learning performance of CNNs is limited due to their strictly homogenous configuration with the sole linear neuron model. This has been addressed by operational neural networks (ONNs) with their heterogenous network configuration encapsulating neurons with various nonlinear operators. In this study, to further boost the peak detection performance along with an elegant computational efficiency, we propose 1-D Self-Organized ONNs (Self-ONNs) with generative neurons. The most crucial advantage of 1-D Self-ONNs over the ONNs is their self-organization capability that voids the need to search for the best operator set per neuron since each generative neuron has the ability to create the optimal operator during training. The experimental results over the China Physiological Signal Challenge-2020 (CPSC) dataset with more than one million ECG beats show that the proposed 1-D Self-ONNs can significantly surpass the state-of-the-art deep CNN with less computational complexity. Results demonstrate that the proposed solution achieves a 99.10% F1-score, 99.79% sensitivity, and 98.42% positive predictivity in the CPSC dataset, which is the best R-peak detection performance ever achieved.

Generalized Acoustic Data Analysis Framework for Leakage Detection and Localization in Field Operational Water Distribution Networks

Generalized Acoustic Data Analysis Framework for Leakage Detection and Localization in Field Operational Water Distribution Networks

A model-based approach for converting CGMES power system models into operational networks

This paper presents a model-based approach for converting Common Grid Model Exchange Specification (CGMES) power system models into operational networks. The approach utilizes the Model-Driven Architecture (MDA), borrowed from software engineering, to create a Platform-Independent Model (PIM) saved in a graph database, neo4j. The PIM is then transformed into Platform-Specific Model (PSM) using cypher queries. Finally, from the PSM, a Pandapower network is generated that is used later for running power flow simulations. The Common Information Model (CIM) ontology classes, which form the metamodel behind CGMES, are loaded into the same database, connecting them to the grid instances, allowing for automatic validation and inference. The proposed approach has been successfully applied to large-scale real-world grids that include renewable energy sources.