Network Solution Research Articles

Demonstration of 200 Gb/s/λ PON in O-Band With High-Bandwidth TFLN Modulator

As the broadband applications continue rising, the solutions of high-speed passive optical network (PON) with large power budget are highly desired. Direct detection scheme combined with digital signal procession (DSP) has been considered as a promising candidate due to its low cost, low power consumption and simple architecture. In this paper, we experimentally demonstrate 200 Gb/s/λ and 240 Gb/s/λ time division multiplexed passive optical network (TDM-PON) downstream transmissions based on four-level pulsed amplitude (PAM-4) modulation with direct detection in O-band. A high bandwidth thin film lithium niobate modulator is adopted to generate high Baudrate optical signal at the optical line terminal (OLT) side. A pre-amplified semiconductor optical amplifier SOA is employed at the optical network unit (ONU) side to improve the power budget. Several DSP algorithms are studied to compensate the linear and nonlinear distortions. We also propose a simplified Volterra nonlinear equalizer (S-VNLE) to reduce the computational complexity. Considering the soft-decision FEC (SD-FEC) threshold (1 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−2</sup> ), over 29 dB and 28 dB power budgets are achieved respectively for 200 Gb/s/λ and 240 Gb/s/λ PON system after 20 km standard single mode fiber (SSMF) based on our proposed S-VNLE.

Simulation Studies of Link Group in Elastic Optical Networks Used in Internet of Things Solutions

Abstract The elastic optical network (EON) solution is the most flexible in terms of allocation of resources for various traffic streams transported in core part (edge-cloud) of Internet of Things (IoT) networks. This paper presents the results of a study on the determination of the occupancy distribution and the loss probability for particular traffic flows offered in EONs supporting IoT systems. The purpose of the research was to determine the impact of: the number of links making up a group, the numbers of frequency slot units (FSUs) in individual EON link, the traffic intensity of individual traffic flows, the capacity of the EON link (number of FSUs) and the number of FSUs requested by different traffic streams. The developed simulation program of link group of EON allows the calculation of the values of the occupancy distribution and loss probability. The results obtained from the simulation are presented on the graphs in the form of points with confidence intervals.

Enabling Proportionally-Fair Mobility Management With Reinforcement Learning in 5G Networks

Mobility management in 5G is challenging, and at higher frequencies, a larger number of cells is needed to provide similar coverage to that in 4G. Consequently, Base Stations (BSs) are placed much more densely and users experience frequent handovers, reducing network capacity. Advanced handover techniques are needed in 5G to perform smooth network operation. In this paper, we formulate an optimization problem, whose goal is to strive for fairness in data rates among users and to reduce handovers. To accomplish that, we consider jointly the decisions when to handover and to which BS a user is to be assigned. This is an integer nonlinear program, and by relaxing it, we obtain an upper bound. Further, due to its NP-hardness, we propose a centralized and a multi-agent Deep Q Network (DQN)-based algorithm, which both find near-optimal user-to-BS assignments. We evaluate our Reinforcement Learning-based solutions for networks of different sizes and users with different velocities. We compare our approaches with baselines and show that they outperform them considerably in terms of fairness and radio link failures while being within 95% of the optimum. Our DQN algorithms also reduce the handover rate by up to 93% and avoid ping-pong handovers almost completely.

Automatic Spike Neural Technique for Slicing Bandwidth Estimated Virtual Buffer-Size in Network Environment

The Next-generation networks, such as 5G and 6G, need capacity and requirements for low latency, and high dependability. According to experts, one of the most important features of (5 and 6) G networks is network slicing. To enhance the Quality of Service (QoS), network operators may now operate many instances on the same infrastructure due to configuring able slicing QoS. Each virtualized network resource, such as connection bandwidth, buffer size, and computing functions, may have a varied number of virtualized network resources. Because network resources are limited, virtual resources of the slices must be carefully coordinated to meet the different QoS requirements of users and services. These networks may be modified to achieve QoS using Artificial Intelligence (AI) and machine learning (ML). Developing an intelligent decision-making system for network management and reducing network slice failures requires reconfigurable wireless network solutions with machine learning capabilities. Using Spiking Neural Network (SNN) and prediction, we have developed a 'Buffer-Size Management' model for controlling network load efficiency by managing the slice's buffer size. To analyze incoming traffic and predict the network slice buffer size; our proposed Buffer-Size Management model can intelligently choose the best amount of buffer size for each slice to reduce packet loss ratio, increase throughput to 95% and reduce network failure by about 97%.

Recurrent Network Solutions for Human Posture Recognition Based on Kinect Skeletal Data.

Ambient Assisted Living (AAL) systems are designed to provide unobtrusive and user-friendly support in daily life and can be used for monitoring frail people based on various types of sensors, including wearables and cameras. Although cameras can be perceived as intrusive in terms of privacy, low-cost RGB-D devices (i.e., Kinect V2) that extract skeletal data can partially overcome these limits. In addition, deep learning-based algorithms, such as Recurrent Neural Networks (RNNs), can be trained on skeletal tracking data to automatically identify different human postures in the AAL domain. In this study, we investigate the performance of two RNN models (2BLSTM and 3BGRU) in identifying daily living postures and potentially dangerous situations in a home monitoring system, based on 3D skeletal data acquired with Kinect V2. We tested the RNN models with two different feature sets: one consisting of eight human-crafted kinematic features selected by a genetic algorithm, and another consisting of 52 ego-centric 3D coordinates of each considered skeleton joint, plus the subject's distance from the Kinect V2. To improve the generalization ability of the 3BGRU model, we also applied a data augmentation method to balance the training dataset. With this last solution we reached an accuracy of 88%, the best we achieved so far.

The Diverse Technology of MANETs: A Survey of Applications and Challenges

In recent years, mobile ad hoc networks (MANETs) have shown much potential in providing wireless network solutions for a variety of situations. MANETs are made up of a variety of devices (both mobile and internet of things (IoT)) which form a wireless network that provides wireless network functionality and cloud services and can be used to provide wireless services to devices in areas where fixed infrastructure is not accessible. Although it is not necessary for a node to be connected to fixed wireless or cellular infrastructure for the network, even a single node connected to infrastructure opens up a wide range of new research opportunities and applications. Due to the variety of compatible devices, MANETs are the perfect technology for a wide variety of scenarios, such as: military, search and rescue operations, and more. Our contribution is a complete survey of the applications of MANETs, which to our knowledge has not been previously documented. This survey will discuss the industries MANETs are used in (e.g. education, smart city, etc.), the benefits and services they can provide users, and will also explore the challenges this technology still experiences

МЕТОДЫ И СРЕДСТВА СЛЕЖЕНИЯ ЗА ПЕРЕМЕЩЕНИЕМ И ВЗАИМОДЕЙСТВИЕМ СОТРУДНИКОВ И ПОКУПАТЕЛЕЙ ПО ВИДЕОИЗОБРАЖЕНИЮ

Due to the rapid development of the sphere of trade, the means of automatic control of thework of employees providing services to customers are gaining particular popularity. At the moment,there are many modern approaches, methods and algorithms for automatically trackingbuyers and sellers in the store. Modern companies are trying to solve this problem in differentways: counting visitors, monitoring devices, various neural network solutions, and so on. Afterreviewing the solutions with the necessary functionality, the main disadvantages were identified,such as, for example, high cost, inconvenience in use, and so on. As a result, the authors set agoal: to improve the quality of tracking the movement of employees / customers through the developmentof automated means and methods of movement control, inter-chamber tracking and identificationof the individual. The article describes a method for automatic recognition and tracking ofemployees of stores and firms. The method is based on a cascade of neural networks and algorithmsthat allow recognizing customers and employees in uniform, as well as evaluating thequality of employees' work and customer satisfaction by voice. As the results of the research, thisarticle presents models and methods for classifying customers and sellers by uniform, methods fordetermining the level of interaction between sellers and customers based on algorithms for determiningthe satisfaction of visitors and customers by voice and face, and algorithms for determiningthe quality of employees' work. The developed methods can improve the efficiency of employees, aswell as increase the quality of services provided. Based on the results of the work, testing wascarried out and a conclusion was made about the satisfactory performance of the presented methodsand algorithms.

Target-oriented deep learning-based image registration with individualized test-time adaptation.

A classic approach in medical image registration is to formulate an optimization problem based on the image pair of interest, and seek a deformation vector field (DVF) to minimize the corresponding objective, often iteratively. It has a clear focus on the targeted pair, but is typically slow. In contrast, more recent deep-learning-based registration offers a much faster alternative and can benefit from data-driven regularization. However, learning is a process to "fit" the training cohort, whose image or motion characteristics or both may differ from the pair of images to be tested, which is the ultimate goal of registration. Therefore, generalization gap poses a high risk with direct inference alone. In this study, we propose an individualized adaptation to improve test sample targeting, to achieve a synergy of efficiency and performance in registration. Using a previously developed network with an integrated motion representation prior module as the implementation backbone, we propose to adapt the trained registration network further for image pairs at test time to optimize the individualized performance. The adaptation method was tested against various characteristics shifts caused by cross-protocol, cross-platform, and cross-modality, with test evaluation performed on lung CBCT, cardiac MRI, and lung MRI,respectively. Landmark-based registration errors and motion-compensated image enhancement results demonstrated significantly improved test registration performance from our method, compared to tuned classic B-spline registration and network solutions without adaptation. We have developed a method to synergistically combine the effectiveness of pre-trained deep network and the target-centric perspective of optimization-based registration to improve performance on individual testdata.

Multi-agent deep reinforcement learning for user association and resource allocation in integrated terrestrial and non-terrestrial networks

Integrating the terrestrial network with non-terrestrial networks to provide radio access as anticipated in the beyond 5G networks calls for efficient user association and resource allocation strategies. In this work, a weighted sum single objective optimization problem that maximizes the total network data rate while minimizing the mobility-induced handoffs and prioritizing service provisioning of mission-critical users in the integrated terrestrial and non-terrestrial network is formulated. The problem’s complexity is reduced by solving the defined problem in two phases; the user association followed by the resource distribution phase. Several proposed approaches to the user association sub-problem utilize a central node that requires nearly-complete information, which may not be available in real time. On the contrary, this paper proposes a centralized training and distributed execution multi-agent dueling double deep Q network (MA3DQN) solution. Each user collects the channel state and access node loading information in this approach and makes an association decision that considers its quality of service requirements. The algorithm’s performance is validated through comparison with the genetic algorithm (GA), the integer linear programming (ILP), a heuristic approximation-based solution, the greedy approach, and the random user association (RUA) algorithm. Moreover, the paper simulates the multi-agent deep Q network solution as an additional benchmark algorithm. Simulation results reveal that as the number of users in the network increases, the acquired data rate of the MA3DQN is within 0.48% and 0.4% of that achieved by the GA and ILP, respectively, and outperforms all other algorithms. Notably, the proposed MA3DQN algorithm presents the best running time, attaining a gain of 99.9% over the GA algorithm, which performs the poorest among the algorithms characterized by polynomial worst-case time complexity. Besides, the MA3DQN approach maintains a handoff probability of zero, unlike the ILP, the approximation-based, greedy, and RUA solutions.

Demodulation Reference Signal (DM-RS) based channel estimation for non-terrestrial networks to support high mobility

This paper provides an adaptive 5G-NR based solution for non-terrestrial networks (NTN) to support the high mobility of the user equipment (UE). A front-loaded demodulation reference signal (DM-RS) symbol pattern with additional orthogonal symbols is proposed to support UE mobility 500 km/h. We develop a 5G-NR link-level simulator (LLS) to analyze the performance of physical downlink shared channel (PDSCH) under the proposed front-loaded DM-RS symbol pattern, tapped delay line (TDL) channel model, and precoding-based beamforming/MIMO schemes. The simulation results show that the proposed channel estimation method based on front-loaded DM-RS helps to improve the link reliability and achieves the maximum spectral efficiency up to 5.05 bps/Hz under 64-QAM, which can also contribute to the evolution of NTN in 6G networks to support the transonic speed mobility of UE.

Generating Defective Epoxy Drop Images for Die Attachment in Integrated Circuit Manufacturing via Enhanced Loss Function CycleGAN

In integrated circuit manufacturing, defects in epoxy drops for die attachments are required to be identified during production. Modern identification techniques based on vision-based deep neural networks require the availability of a very large number of defect and non-defect epoxy drop images. In practice, however, very few defective epoxy drop images are available. This paper presents a generative adversarial network solution to generate synthesized defective epoxy drop images as a data augmentation approach so that vision-based deep neural networks can be trained or tested using such images. More specifically, the so-called CycleGAN variation of the generative adversarial network is used by enhancing its cycle consistency loss function with two other loss functions consisting of learned perceptual image patch similarity (LPIPS) and a structural similarity index metric (SSIM). The results obtained indicate that when using the enhanced loss function, the quality of synthesized defective epoxy drop images are improved by 59%, 12%, and 131% for the metrics of the peak signal-to-noise ratio (PSNR), universal image quality index (UQI), and visual information fidelity (VIF), respectively, compared to the CycleGAN standard loss function. A typical image classifier is used to show the improvement in the identification outcome when using the synthesized images generated by the developed data augmentation approach.

Selecting IOT WSNIDS Based on Metrics User Necessity Weight Approach

Internet of Things (IoT) WSNIDS is a network security software that is built for detection of vulnerability exploits against attacks. The choice of WSNIDS depends on the IoT architecture and application. The administrator is the one who will decide which WSNIDS will be the best solution for the sensor network. Not one solution is possible that is going to work for all, so administrator has to equate the abilities that individual WSNIDS provide along with economical, information and desires to find one that works top for them. The concept in this paper offers a user requirement weight-based method to WSNIDS choice for IoT. We primary discuss user WSNIDS requirements and WSNIDS metrics, after that for each WSNIDS necessity we match the metric(s) that are concerned. User are required to lists their WSNIDS necessities in a limited collection that are significant ranging from least to most. User necessities are frequently indicated in a positive form or transformed to the optimistic form. The initial requisite (i.e., minimum significant) is allocated the lowest weight (e.g., one) while the remaining requirements are allocated growing weights in proportion to their relative importance. Once the necessities are weighted, each WSNIDS metric is allocated a weight that is identical to the totality of the weights of the necessities it is going to contribute. WSNIDS metrics are organized in downward sequence where metric having maximum weight is at the topmost position. Suitable WSNIDS means may be selected after match is done between the metrics weight and WSNIDS features.

Corneal endothelial image segmentation training data generation using GANs. Do experts need to annotate?

Objective.Current solutions for corneal endothelial image segmentation use convolutional neural networks. However, their potential is not exploited due to the scarcity of labeled corneal endothelial data caused by an expensive cell delineation process. Therefore, this work proposes synthesizing cell edges and corresponding images using generative adversarial neural networks. To our knowledge, such work has not yet been reported in the context of corneal endothelial image segmentation. Methods.A two-step pipeline for synthesizing training patches is proposed. First, a custom mosaic generator creates a grid that mimics a binary map of endothelial cell edges. The synthetic cell edges are then input to the generative adversarial neural network, which is trained to generate corneal endothelial image patches from the corresponding edge labels. Finally, pairs of synthetic patches are used to train the patch-based U-Net model for corneal endothelial image segmentation. Results.Experiments performed on three datasets of corneal endothelial images show that using purely synthetic data for U-Net training allows image segmentation with comparable accuracy to that obtained when using original images annotated by experts. Depending on the dataset, the segmentation quality decreases only from 1% to 4%. Conclusions:Our solution provides a cost-effective source of diverse training data for corneal endothelial image segmentation. Significance.Due to the simple graphical user interface wrapping the proposed solution, many users can easily obtain unlimited training data for corneal endothelial image segmentation and use it in various scenarios.

Supervised Machine Learning Method for Ontology-based Financial Decisions in the Stock Market

For changing semantics, ontological and information presentation, as well as computational linguistics for Asian social networks, are one of the most essential platforms for offering enhanced and real-time data mapping, as well as huge data access across diverse big data sources on the web architecture, information extraction mining, statistical modeling and data modeling, database control, and so on. The concept of opinion or sentiment analysis is often used to predict or classify the textual data, sentiment, affect, subjectivity, and other emotional states in online text. Recognizing the message's positive and negative thoughts or opinions by examining the author's goals will aid in a better understanding of the text's content in terms of the stock market. An intelligent ontology and knowledge Asian social network solution can improve the effectiveness of a company's decision making support procedures by deriving important information about users from a wide variety of web sources. However, ontology is concerned primarily with problem-solving knowledge discovery. The utilization of Internet-based modernizations welcomed a significant effect on the Indian stock exchange. News related to the stock market in the most recent decade plays a vital role for the brokers or users. This article focuses on predicting stock market news sentiments based on their polarity and textual information using the concept of ontological knowledge-based Convolution Neural Network (CNN) as a machine learning approach. Optimal features are essential for the sentiment classification model to predict the stock's textual reviews' exact sentiment. Therefore, the swarm-based Artificial Bee Colony (ABC) algorithm is utilized with the Lexicon feature extraction approach using a novel fitness function. The main motivation for combining ABC and CNN is to accelerate model training, which is why the suggested approach is effective in predicting emotions from stock news.

Plug and Charge Solutions with Vehicle-to-Grid Communication

With electro-mobility, the demand for charging services is increasing. In Germany alone, the Federal Ministry for Digital Affairs and Transport required for at least 843,000 new charging stations by 2030. Charging service is key for electro-mobility and requires a new innovative technology for the future. Plug and Charge (PnC) enables easy, convenient, and secure charging process compared in existing technologies (RFID, SMS, or App). This secure and convenient charging process is based on digital certificates from PnC ecosystem. Today, validation of certificates between electric vehicle (EV) and charging station (EVSE) in PnC ecosystem is very simple. For the future, there are many new requirements for PnC ecosystem from new standard (ISO 15118-20), Federal Ministry of Economics and Climate Protection, new certificates (V2G root CAs) and new Pool&Service (e.g., Root Certificate Pool). This will make the validation of certificates between EV and EVSE more and more difficult. These new and existing requirements are identified, analyzed and evaluated here. With this article, for the first time, these requirements are derived on entire PnC ecosystem, on cloud components (e.g., MO) and on individual functions (e.g., mTLS, certificate authentication, and authorization). From these requirements, a three-level solution (1-EV, 2-PKI, and 3-Cloud components (e.g., MO, CPO)) is generated. These three level solutions enable successful validation of the certificate with or without interoperability of the certificate. New and various solutions for future networks (e.g., contract and root certificate pools) for either interoperable or non-interoperable networks are illustrated here.

Modeling Minimum Cost Network Flows With Port‐Hamiltonian Systems

AbstractWe give a short overview of advantages and drawbacks of the classical formulation of minimum cost network flow problems and solution techniques, to motivate a reformulation of classical static minimum cost network flow problems as optimal control problems constrained by port‐Hamiltonian systems (pHS). The first‐order optimality system for the port‐Hamiltonian system‐constrained optimal control problem is formally derived. Then we propose a gradient‐based algorithm to find optimal controls. The port‐Hamiltonian system formulation naturally conserves flow and supports a wide array of further modeling options as, for example, node reservoirs, flow dependent costs, leaking pipes (dissipation) and coupled sub‐networks (ports). They thus provide a versatile alternative to state‐of‐the art approaches towards dynamic network flow problems, which are often based on computationally costly time‐expanded networks. We argue that this opens the door for a plethora of modeling options and solution approaches for network flow problems.

A Bayesian Multi-Armed Bandit Algorithm for Dynamic End-to-End Routing in SDN-Based Networks with Piecewise-Stationary Rewards

To handle the exponential growth of data-intensive network edge services and automatically solve new challenges in routing management, machine learning is steadily being incorporated into software-defined networking solutions. In this line, the article presents the design of a piecewise-stationary Bayesian multi-armed bandit approach for the online optimum end-to-end dynamic routing of data flows in the context of programmable networking systems. This learning-based approach has been analyzed with simulated and emulated data, showing the proposal’s ability to sequentially and proactively self-discover the end-to-end routing path with minimal delay among a considerable number of alternatives, even when facing abrupt changes in transmission delay distributions due to both variable congestion levels on path network devices and dynamic delays to transmission links.

Widening Blockchain Technology toward Access Control for Service Provisioning in Cellular Networks.

The attention on blockchain technology (BCT) to create new forms of relational reliance has seen an explosion of new applications and initiatives, to assure decentralized security and trust. Its potential as a game-changing technology relates to how data gets distributed and replicated over several organizations and countries. This paper provides an introduction to BCT, as well as a review of its technological aspects. A concrete application of outsource access control and pricing procedures in cellular networks, based on a decentralized access control-as-a-service solution for private cellular networks, is also presented. The application can be used by service and content providers, to provide new business models. The proposed method removes the single point of failure from conventional centralized access control systems, increasing scalability while decreasing operational complexity, regarding access control and pricing procedures. Design and implementation details of the new method in a real-world scenario using a private cellular network and a BCT system that enables smart contracts are also provided.

Robustness Analysis of BAM Cellular Neural Network with Deviating Arguments of Generalized Type

By generating equivalent integral equations, we analyze the existence and uniqueness of solutions of bidirectional associative memory cellular neural network (BAMCNN) with deviating arguments firstly. Secondly, the question of robustness of stability (RoS) of BAMCNN with deviating argument is studied. Using the Gronwall inequality, we calculate the upper bounds of the interference intensities that can maintain the initial stability of system. The perturbed BAMCNN will maintain its original stability if the strength of one or more perturbations is less than the upper bounds that we calculated in this study. To demonstrate the validity of the conjectural values, a variety of numerical illustrations are provided.

Request delay and survivability optimization for software defined‐wide area networking (SD‐WAN) using multi‐agent deep reinforcement learning

AbstractData exchange between headquarters and local branches represents a major challenge issue for business success. For this issue, traditional solutions applied to wide area networks (WAN) remain unrealistic and require a good knowledge of the systems. Recently, software‐defined wide area networking (SD‐WAN) plays a pivotal role and constitutes, in general, a reliable solution for wide area networking. Compared to the classical WAN, SD‐WAN decouples the control plane from gateway devices. Moreover, SD‐WAN are based on centralized WAN management with dynamic reconfiguration, allowing it to control the entire service requirements. Nevertheless, traffic management is the most usual metric in SD‐WAN. It gives a clear indication on propagation latency, request delay and survivability; that is, network connectivity. In this article, to guarantee an efficient quality of service and deal with the increased delay problem, we investigate the problem of joint optimization of average request delay and survivability in the proposed SD‐WAN model. The optimization approaches are formulated so as to minimize the average request delay and maximize the network connectivity. Subsequently, we adopt a multi‐agent deep Q‐Network algorithm to solve it, where the reward function is reformulated based on the optimization objective. Simulation results show that our strategy improves the system's performance significantly in terms of request delay and survivability, compared to traditional baseline algorithms.