Search

Most real-world networks are time-varying, and many are subject to the stochastic functioning of their nodes and edges. Examples can be seen in the human brain undergoing an epileptic seizure, spontaneous infection and recovery in epidemics, and intermittent functioning of devices in the Internet of Things. Moreover, such networks are becoming increasingly large due to rapid technological advances. However, little has been done to study time-varying, large-scale, stochastic networks (SNs) from a reliability engineering perspective. Toward this goal, this article develops a fault-tolerance model for a type of time-varying network in which nodes (and/or edges) stochastically switch between active and inactive states. It considers fault tolerance from a global connectivity point of view, which has applications in many natural and engineered networks. Specifically, this article presents a Markov chain framework that models the dynamic behavior of nodes and allows for the computation of quantitative measures, including availability and time-to-failure metrics. To accommodate large-scale networks and emphasize global connectivity, this framework utilizes percolation theory, which has recently been of interest in the reliability engineering discipline, to characterize network failure. This article makes several contributions: it proposes a Markov chain framework for computing fault-tolerance metrics that is tractable for large-scale networks, it shows the existence of a phase transition in network availability of a time-varying SN, and it accounts for finite-size effects of percolation in the fault-tolerance model. The proposed methodology is applied to Erdos-Renyi random graphs and a real, large-scale power grid. Experimental results provide insights into network design, maintenance, and failure prevention of time-varying SNs.

Online and omnichannel retailers are proposing increasingly tight delivery deadlines, moving toward instant on-demand delivery. To operate last-mile distribution systems with such tight delivery deadlines efficiently, defining the right strategic distribution network design is of paramount importance. However, this problem exceeds the complexity of the strategic design of traditional last-mile distribution networks for two main reasons: (1) the reduced time available for order handling and delivery and (2) the absence of a delivery cut-off time that clearly separates order collection and delivery periods. This renders state-of-the-art last-mile distribution network design models inappropriate, as they assume periodic order fulfillment based on a delivery cutoff. In this study, we propose a metamodel simulation-based optimization (SO) approach to strategically design last-mile distribution networks with tight delivery deadlines. Our methodology integrates an in-depth simulator with traditional optimization techniques by extending a traditional black-box SO algorithm with an analytical model that captures the underlying structure of the decision problem. Based on a numerical study inspired by the efforts of a global fashion company to introduce on-demand distribution with tight delivery deadlines in Manhattan, we show that our approach outperforms contemporary SO approaches as well as deterministic and stochastic programming methods. In particular, our method systematically yields network designs with superior expected cost performance. Furthermore, it converges to good solutions with a lower computational budget and is more consistent in finding high-quality solutions. We show how congestion effects in the processing of orders at facilities negatively impact the network performance through late delivery of orders and reduced potential for consolidation. In addition, we show that the sensitivity of the optimal network design to congestion effects in order processing at the facilities increases as delivery deadlines become increasingly tight.

Mobile edge computing (MEC)-enabled Internet of Things (IoT) networks have been deemed a promising paradigm to support massive energy-constrained and computation-limited IoT devices. Energy harvesting (EH) further enhances the operating capabilities of IoT devices that normally only possess very limited energy support. Nevertheless, many studies show that IoT devices using EH can experience uncertainty and unpredictability, which can complicate EH-based IoT network design. Furthermore, with many new services in 5G and the forthcoming 6G eras such as autonomous driving and vehicular communications, mobility consideration in IoT networks becomes more and more important. In this paper, we study the computing offloading and resource allocation problems in an IoT network that supports both mobility and energy harvesting. The long-term average sum service cost of all the mobile IoT devices (MIDs) is minimized by optimizing the harvested energy, task-partition factors, the CPU frequencies, the transmit power, and the association vector of MIDs. An online mobility-aware offloading and resource allocation (OMORA) algorithm is proposed based on Lyapunov optimization and Semi-Definite Programming (SDP). This online algorithm optimizes the offloading scheme without the need to have prior knowledge of user mobility, the EH model, and the channel condition. Theoretical analysis shows that the proposed OMORA algorithm can achieve asymptotic optimality. Simulation results demonstrate that the proposed algorithm can effectively balance the system service cost and energy queue length, and outperform other offloading benchmark algorithms on the system service cost and packet losses.

Recent advances in networking and computing technologies have led to an increased interest in considering computation and communication in a joint and distributed mode according to the computing in the network paradigm. Furthermore, the emergence of the sixth generation (6G) networks is profiling the ever more challenging applications' requirements imposed by the new era service classes. In order to guarantee efficient implementation and availability to the upcoming applications, characterized by stringent quality of experience requirements, an accurate network design and delay analysis is becoming mandatory to pursue efficient 6G network dimensioning and enable computing in the network-based applications. Toward this goal, this article proposes a suitable end-to-end delay performance analysis and reliability evaluation in the case of a 6G network offering virtual reality services, which is considered one of the most challenging technologies for the coming new era service classes. More in depth, the aim of the article is the formulation of the stochastic end-to-end delay bound by applying supermartingale envelopes in order to allow accurate VR reliability prediction in relation to the number of users linked to the same computing node with a specified service profile. The goodness and effectiveness of the proposed approach is validated by providing comparison with simulation results and analytical predictions derived by resorting to the use of the classical Markov queueing theory.

Cellular network traffic increases rapidly, and new services are introduced every year. For proper planning and design of such networks, exact requirements must be known with good accuracy. Dimensioning is an important part of network planning and design. Dimensioning is essential to determine the network requirements. In the coming years, fifth-generation (5G) will be deployed widely. 5G infrastructure is hybrid of wireless and optical components. For 5G network dimensioning, there is a need of a hybrid model. In this paper, the authors develop mathematical expressions for 5G network dimensioning. They use ITU proposed typical 5G network provisions to estimate bandwidth, network capacity, coverage, and capital expenditures. They also establish the correlation between the optical and the wireless parts. The expressions developed in this work can be used for the fast estimation of network coverage. So, this model can play important roles for 5G network planning and design.

One of the goals of cloud service providers is to satisfy service-level agreements without significant over-provisioning in data center clusters. Efforts to meet these requirements have been mainly based on resource over-provisioning rather than identifying performance bottlenecks. While increasing parallelism tends to reduce the average and tail latency, the joint impact of concurrent job scheduling and parallel task processing is a challenging problem to analytically model, particularly when compared to the models developed without the notion of concurrency. This article presents an analytical model for distributed schedulers in data center cluster networks. The model can be used to investigate how latency can affect a data center network design and how many resources should be allocated to meet service-level agreements. To get better insight, we build upon ideas from queuing networks, which provide a framework to measure expected latency versus resource provisioning. The model is based on tandem queuing networks and fork–join systems to compute expected latency in closed forms at various stages of data center cluster networks. Theoretical analysis and simulations have been conducted to demonstrate the effectiveness of the proposed model and to strike a balance between expected latency and resource utilization. Results obtained from various simulation scenarios on different data center traffic traces confirm the soundness of the model.

We address the problem of distributed state estimation of a linear dynamical process in an attack-prone environment. Recent attempts to solve this problem impose stringent redundancy requirements on the measurement and communication resources of the network. In this paper, we take a step towards alleviating such strict requirements by exploring two complementary directions: (i) making a small subset of the nodes immune to attacks, or "trusted", and (ii) incorporating diversity into the network. We define graph-theoretic constructs that formally capture the notions of redundancy, diversity, and trust. Based on these constructs, we develop a resilient estimation algorithm and demonstrate that even relatively sparse networks that either exhibit node-diversity, or contain a small subset of trusted nodes, can be just as resilient to adversarial attacks as more dense networks. Finally, given a finite budget for network design, we focus on characterizing the complexity of (i) selecting a set of trusted nodes, and (ii) allocating diversity, so as to achieve a desired level of robustness. We establish that, unfortunately, each of these problems is NP-complete.

Wireless Sensor Network (WSN) based Communication has been devised for exchanging data with low cost, minimal maintenance, and for more convenience. These communications are used in various applications like monitoring, surveillance, defence, healthcare, automation etc. Several routing protocols had been proposed for energy efficient wireless communication and to prolong the network life time. In such protocols, the network design plays a vital role in improving the network performance. The communication strategy among the sensor nodes depend on network design criteria. The paper therefore, presents a novice hierarchical structure based network design and energy efficient routing method for WSN. Here the clusters are formed using fuzzy-multi-criteria decision approach and cluster heads are optimally selected using Analytical Hierarchy Process. The Penguin Search Optimization Algorithm is used for diversified and intensified strategic planning for both intra-cluster and inter-cluster communication. The proposed technique is compared with existing techniques on multiple parameters and found to perform better. It shows a throughput of 0.95 Mbps and minimal energy consumption of 0.35mj and proving it to be energy efficient.

Data center networks provide the physical infrastructure that hosts Internet‐related services and cloud computing. Designing data center networks properly is imperative for Internet‐related service and cloud computing providers to gain competitive edges through cost efficiency and service quality. In this study, we formulate a mathematical programming model to address the data center network design problem, in which the objective is to minimize total operating cost and the service delay penalty by optimizing data center location, footprint allocation, and resource provisioning decisions, while incorporating essential features, such as latency, power, multiple resources, configuration limits, and interdependent footprints. We employ a queueing model to approximate the service latency and provide tractable reformulations. To enhance computational efficiency for large‐scale problems, we further develop Lagrangian relaxation methods and generate strengthening cuts by exploiting the structural properties of the problem. Our numerical studies demonstrate that the proposed model, which jointly optimizes location, allocation, and resource provisioning, can achieve significant cost reductions and improvements in service quality compared with a hierarchical approach that optimizes these decisions sequentially. Moreover, our proposed solution methods outperform state‐of‐the‐art commercial software in terms of computational efficiency. Based on real‐world datasets, the proposed model selects data centers that have been chosen by major cloud computing infrastructure providers. We also draw managerial insights that can be used as design guidelines in practice.

Due to rapid growth of demanding data rate, many emerged technologies has been developed for wireless communication. Consequently, the perspective of network design has been shifted to new scale, wherein massive number of machines and people can be handled. Ultra-dense networks (UDNs) represents the bottleneck of 5G system capacity. In this paper, we investigate design criteria of UDNs and discuss the relative technologies which are regarded as the main axes of the current research. Also, coverage performance analysis of dense network is introduced based on stochastic geometry, besides the available software packages of 5G networks. Finally, we introduce the most relevant novel trends and open issues for future research directions in this track. Hence, this survey represents complete paradigm for going through UDNs.

Software-defined networking (SDN) has become the technology of choice for designing the next-generation network infrastructure that is featured with high-volume traffics, rapidly increased scale, and dynamic nature. Furthermore, to deploy multiple controllers in the control plane of SDN is widely considered with the aim of improving the stability and reliability of the network. This paper presents an integrated framework for a comprehensive multi-controller management in SDN. The proposed framework is comprised of a network planning phase and a runtime maintenance phase. Novel algorithms are proposed in the network planning phase to estimate the required number of controllers in the network, to determine the nodes for placing the controllers, and to assign the switch to its managing controller. Moreover, these algorithms are designed by mitigating the problems of device isolation and controller overload such that the reliability and stability of the control plane can be enhanced. In addition, a mechanism based on the State Behavior Tree is proposed in the runtime maintenance phase of the framework. This mechanism dynamically manages the loading of the controller during the execution time so that the occurrence of the controller overload is minimized. The experimental results show that, compared to the prior arts, the proposed framework reduces the isolation probability by up to 89% and increases the device connectivity by up to 34%. The occurrence of the controller overload during runtime is also significantly decreased.

Intelligent Internet of Things (IIoT) is comprised of various wireless and wired networks for industrial applications, which makes it complex and heterogeneous.The openness of IIoT has led to the intractable problems of network security and management. Many network security and management functions rely on network traffic prediction techniques, such as anomaly detection and predictive network planning. Predicting IIoT network traffic is significantly difficult because its frequently updated topology and diversified services lead to irregular network traffic fluctuations. Motivated by these observations, we proposed a reinforcement learning-based mechanism in this article. We modeled the network traffic prediction problem as a Markov decision process, and then, predicted network traffic by Monte Carlo $Q$ -learning. Furthermore, we addressed the real-time requirement of the proposed mechanism and we proposed a residual-based dictionary learning algorithm to improve the complexity of Monte Carlo $Q$ -learning. Finally, the effectiveness of our mechanism was evaluated using the real network traffic.

A wireless rechargeable sensor network was proposed to extend the lifetime of the wireless sensor network. In this article, a charger is combined together with a self-propelled vehicle to provide a more flexible result of charger deployment. The dynamic chargers path selection problem is defined and mapped into the traveling salesman problem. Four metaheuristic algorithms for Internet-of-Things (IoT) applications are designed, and the higher fitness value between the charging path and the number of dead IoT devices is achieved. However, metaheuristic approaches may spend more time on searching solutions so that many IoT devices overuse limited power and fail to be charged for a long time, leading to power exhaustion. In this article, the edge computing technique is applied to accelerate the obtainment of charging paths with the well-defined edge/centralized unit switching. Moreover, to assure the calculated path trustworthy and will not be tampered with, the blockchain technology is adopted. The proposed architecture maintains high-level information credibility while transmitting the information of charging paths within the cloud and edge. The simulation results showed that the proposed method is capable of achieving better charging efficiency and less deployment cost.

We consider the problem of restoring services provided by an interdependent set of infrastructures after they were disrupted from an extreme event. Specifically, we select the set of damaged infrastructure arcs for immediate restoration and schedule these on a set of machines (work crews). Our novel contribution is that when we determine the selection and scheduling of these damaged arcs, we explicitly consider the movement of machines through a damaged transportation network that is currently being restored. Previous works failed to consider how machine movement greatly influences the ability to conduct timely restoration due to the interdependence on the transportation network. To model this restoration construct, we propose an interdependent integrated network design and scheduling problem with movement of machines (IINDS-MM). In an IINDS-MM problem, we have a base transportation network and at least one additional infrastructure network layer. For each network layer, we determine what damaged arcs are selected for restoration, which machine will conduct the restoration, and the sequence of tasks assigned to each machine when explicitly considering machine movement through the changing transportation network. We propose a mixed integer programming formulation of the IINDS-MM problem and solve it using a rolling horizon solution procedure. Using realistic data representing Juan Diaz, Panama and the customizable artificial community CLARC data set, we simulate different storm surge levels and possible damage scenarios. We then solve the IINDS-MM problem and deduce insights about machine starting locations, machine capabilities, and the performance of IINDS-MM compared to existing restoration models.

Many wireless Internet-of-Things applications require extended battery life ranging from a few months to a few years. Such applications have motivated the recent developments in low power wide area networks, including the rise of Long Range (LoRa) technology. LoRa has a simple modulation scheme designed for extended converge, low battery consumption, and resistance to high interference levels. Thus LoRa is primarily targeted for shared spectrum applications where interference levels are typically higher than controlled spectrum applications where a single operator usually has a dominant control on the quality of service. As a result, it is of paramount importance to carefully design IoT networks while taking into account the impending impacts of interference and propagation environments. This paper presents a novel LoRa network design framework that utilizes a developed open-source emulator to provide a reliable network coverage estimation. The framework is tested in one of the largest open-access IoT network designs in Australia, which enabled the deployment of 294 sensors and 48 gateways. Both the framework and the emulator are implemented using MATLAB scripting, enabling integration with built-in and external radio planning tools. The framework leverages real interference measurements captured using software defined radio that records the spectrotemporal behavior of the existing traffic in the shared band.

Nowadays, legacy electrical distribution systems are migrating to a new modern electric grid with the capability of supporting different applications such as advanced metering infrastructure (AMI), distributed energy resources, and electric vehicles. Among these applications, AMI is playing an important role in delivering data from customers to power utilities, supporting reliable real-time monitoring, and remote operation of power quality data and voltage profile. The AMI consists of smart meters, data aggregation points (DAPs), the utility control center, and communication networks. Appropriate network planning plays an essential role in facilitating the exchange of data between consumers and power utilities as well as accommodating new smart grid applications and future growth. This work proposes an optimal placement of DAPs for AMI based on machine learning clustering techniques in residential grids. Network partitioning is introduced to create sub-networks and graph algorithms generate a deployment topology given optimization constraints. A new measurement metric called coverage density is considered to indicate neighborhood area networks (NAN) zones with the appropriate coverage. Three real scenarios of NAN are considered: urban, suburban, and rural. The proposed algorithm is evaluated and compared with conventional heuristic optimization methods with respect to average and maximum distance between smart meters and DAPs, coverage density, and execution time.

This paper focuses on how to increase the availability of a backbone network with minimal cost. In particular, the new framework focuses on resilience against natural disasters and is an evolution of the FRADIR/FRADIR-II framework. It targets three different directions, namely: network planning, failure modeling, and survivable routing. The steady state network planning is tackled by upgrading a sub-network (a set of links termed the spine) to achieve the targeted availability threshold. A new two-stage approach is proposed: a heuristic algorithm combined with a mixed-integer linear problem to optimize the availability upgrade cost. To tackle the disaster-resilient network planning problem, a new integer linear program is presented for the optimal link intensity tolerance upgrades together with an efficient heuristic scheme to reduce the running time. Failure modeling is improved by considering more realistic disasters. In particular, we focus on earthquakes using the historical data of the epicenters and the moment magnitudes. The joint failure probabilities of the multi-link failures are estimated, and the set of shared risk link groups is defined. The survivable routing aims to improve the network's connectivity during these shared risk link group failures. Here, a generalized dedicated protection algorithm is used to protect against all the listed failures. Finally, the experimental results demonstrate the benefits of the refined eFRADIR framework in the event of disasters by guaranteeing low disconnection probabilities even during large-scale natural disasters. This article is based on work from COST Action CA15127 ("Resilient communication services protecting end-user applications from disaster-based failures" - RECODIS), supported by COST (European Cooperation in Science and Technology); http://www.cost.eu. This work was supported in part by the High Speed Networks Laboratory (HSNLab); in part by the National Research, Development, and Innovation Fund of Hungary, financed through the FK_17, KH_18, K_17, FK_20 and K_18 funding schemes, respectively, under Project 123957, Project 129589, Project 124171, Project 134604, and Project 128062; and in part by the BME through the TKP2020, Institutional Excellence Program of the National Research Development and Innovation Office in the field of Artificial Intelligence under Grant BME IE-MI-SC TKP2020. The work of Rita Girão-Silva and Teresa Gomes was supported in part by the Fundação para a Ciência e a Tecnologia (FCT), I.P. under Project UIDB/00308/2020, and in part by the ERDF Funds through the Centre's Regional Operational Program and by National Funds through FCT under Project CENTRO-01-0145-FEDER-029312.

Over the coming years, the foresighted enormous increase in smart devices supporting Internet-of-Things (IoT) applications demand novelty in network design. A promising solution to the ever-increasing low-latency requirement of IoT applications is the development of fog network architecture. However, the presence of an enormous number of smart devices in fog networks affects the performance of the network. To harvest the benefits of fog networking necessitates finding optimal cloudlet selection strategies. This article formulates a mixed-integer non-linear programming (MINLP) problem that has the objective of latency minimization. An exhaustive search on our cache-enabled (CE) fog architecture cannot be applied because of the problem’s combinatorial and NP-hard nature. Similarly, the genetic algorithm (GA) cannot be used to find the solution because of the calculation of the number of generations. The increase in the number of IoT and fog nodes increases the solution search space, hence an Outer Approximation Algorithm (OAA) is proposed to arrive at the solution. Low complexity, convergence, and effectiveness of the proposed algorithm ensures the $\epsilon $ -optimal solution = 10−3, obtained through standard problem solvers.

In mobile network deployments of growing size, the optimum and fast planning of radio resources are a key task. Cloud services enable efficient and scalable implementation of procedures and algorithms. In this paper, a proof of concept implementation of a cloud-based network planning work pattern using Amazon Web Services (AWS) is presented, containing new and efficient radio resource planning algorithms for 3G, 4G and 5G systems. It extracts configuration and performance data from the network, enabling to accurately estimate cells coverage, identify neighboring cells and optimally plan scrambling codes (SCs) and physical cell identity (PCI) in 3G and 4G/5G networks, respectively. This implementation was integrated and is available in the commercial Metric Software-as-a-Service (SaaS) monitoring and planning tool. The cloud-based planning system is demonstrated in various canonical and realistic Universal Mobile Telecommunications System (UMTS) and Long Term Evolution (LTE) scenarios, and compared to an algorithm previously used by Metric. For a small LTE realistic scenario consisting of 9 sites and 23 cells, it takes less than 0.6 seconds to perform the planning. For an UMTS realistic scenario with 12 484 unplanned cells, the planning is efficiently achieved, taking less than 8 seconds, and guaranteeing no collisions between first order neighboring cells. The proposed concept is proved, as this system, capable of automatically planning 3/4/5G realistic networks of multi-vendor equipment, makes Metric more attractive to the market. info:eu-repo/semantics/publishedVersion

The energy consumption and coverage range of unmanned aerial vehicles (UAVs) are major challenges in UAV-based postdisaster communications. To address these challenges, energy harvesting is employed to power communication devices and prolong the lifetime of the wireless communication network during a disaster. In addition, clustering techniques and device-to-device (D2D) communication are needed to increase the overall network coverage and provide sustainable connectivity during the disaster and postdisaster phases. We have proposed a novel emergency communication system (ECS) using the optimal cluster head (CH) technique to improve the energy transfer efficiency for sustainable network connectivity. We have developed a UAV deployment model assisted by the clustering technique and D2D links that is capable of harvesting energy to increase the network lifetime. This new approach is expected to enhance the reliability of the network in disaster situations. The proposed methods have been evaluated by measuring the energy efficiency performance and the network outage probability. The simulation results demonstrate improved performance with the deployment of optimal CHs, while the outage probability has been effectively reduced. Moreover, the proposed approach has been proven to reduce the computational complexity. In conclusion, UAV deployment with the optimal CH algorithm is a suitable network design to recover from natural disasters and potentially save many lives.

Long-Range Wide-Area Network (LoRaWAN) enables flexible long-range service communications with low power consumption which is suitable for many IoT applications. The densification of LoRaWAN, which is needed to meet a wide range of IoT networking requirements, poses further challenges. For instance, the deployment of gateways and IoT devices are widely deployed in urban areas, which leads to interference caused by concurrent transmissions on the same channel. In this context, it is crucial to understand aspects such as the coexistence of IoT devices and applications, resource allocation, Media Access Control (MAC) layer, network planning, and mobility support, that directly affect LoRaWAN’s performance. We present a systematic review of state-of-the-art works for LoRaWAN optimization solutions for IoT networking operations. We focus on five aspects that directly affect the performance of LoRaWAN. These specific aspects are directly associated with the challenges of densification of LoRaWAN. Based on the literature analysis, we present a taxonomy covering five aspects related to LoRaWAN optimizations for efficient IoT networks. Finally, we identify key research challenges and open issues in LoRaWAN optimizations for IoT networking operations that must be further studied in the future.

In this paper, we propose novel resource allocation algorithms for ultra-reliable and low-latency communication (URLLC) in distributed antenna systems (DASs) and discuss the effectiveness of the receive antenna deployment and multiple access strategies in fulfilling the stringent reliability and latency requirements. We analyze achievable rates and reliabilities of the DASs with orthogonal multiple access (OMA) and non-orthogonal multiple access (NOMA) by taking into account the influence of the latency constraint and the non-identical channel distributions of the distributed antennas. Based on the performance analysis, we propose novel resource allocation algorithms for fulfilling the stringent performance requirement in OMA and NOMA. The numerical simulation results provide insightful information for understanding the impacts of multiple access strategy and antenna deployment on the spectral efficiency under the reliability and latency constraints. In particular, NOMA makes it possible to deal with more transmit nodes than OMA, and more dispersed antenna deployment facilitates to achieve a target URLLC performance with less spectrum resource as the ratio of transmit nodes to receive antennas increases. Consequently, the analysis and simulation results provide the meaningful information on the network design for mission-critical applications.

The design of Very High Throughput Satellite (VHTS) systems will be an important milestone for future telecommunications systems in the light of the ongoing integration between satellite and 5G technologies as well as with the upcoming 6 G evolution. To this end, satellite systems will be operating at EHF frequency bands, hence requiring a special attention to the overall network design in terms of ground and space segment optimisation. In particular, the mitigation of feeder link outage events through advanced prediction techniques and the exploitation of smart gateway diversity concepts will be combined with an overarching network resources’ orchestration, jointly conceived to meet the QoS requirements of the end users. In this respect, this paper proposes a novel optimisation framework able to achieve very high throughput system figures, building on an adaptive QoS management concept. The proposed framework has been compared with other existing solutions from the literature and validated by means of extensive simulation campaigns, whose collected results show significant performance improvements.

Network modeling is a key enabler to achieve efficient network operation in future self-driving Software-Defined Networks. However, we still lack functional network models able to produce accurate predictions of Key Performance Indicators (KPI) such as delay, jitter or loss at limited cost. In this paper we propose RouteNet, a novel network model based on Graph Neural Network (GNN) that is able to understand the complex relationship between topology, routing, and input traffic to produce accurate estimates of the per-source/destination per-packet delay distribution and loss. RouteNet leverages the ability of GNNs to learn and model graph-structured information and as a result, our model is able to generalize over arbitrary topologies, routing schemes and traffic intensity. In our evaluation, we show that RouteNet is able to predict accurately the delay distribution (mean delay and jitter) and loss even in topologies, routing and traffic unseen in the training (worst case MRE=15.4%). Also, we present several use cases where we leverage the KPI predictions of our GNN model to achieve efficient routing optimization and network planning., Comment: 12 pages

A key concern in the design of 5G is the radio access network, which is expected to be significantly denser and more advanced, with considerably higher infrastructure and power consumption cost than that of conventional mobile network standards. Novel algorithms/approaches for optimal planning of the radio access network are required for tackling the additional complexity of the problem of cost‐efficient radio access planning in 5G, which cannot be properly handled by conventional approaches. This study proposes a novel optimisation framework for the cost‐efficient deployment and configuration of 5G base stations. The main idea of the proposed optimisation framework is to exploit heterogeneity in three key 5G technologies, heterogeneous base station architecture, cell range extension and multiple‐input–multiple‐output spatial multiplexing, by jointly optimising their configurations during network design. In addition, the proposed optimisation framework includes generic steps for applying meta‐heuristic algorithms to the problem, which are necessary to overcome the problem's complexity, especially for large problem instances. The authors’ results show that their novel optimisation framework improves the cost efficiency of the network planning both in terms of power and infrastructural cost to operators.

Driven by recent advances in networking and computing technologies, distributed application scenarios are increasingly deployed on resource-constrained processing platforms. This includes networked embedded and cyber-physical systems as well as edge computing in mobile applications and the Internet of Things (IoT). In such resource-constrained Networks-of-Systems (NoS), computation and communication workloads need to be carefully co-optimized yet are tightly coupled. How to optimally partition and schedule application tasks among an appropriately designed NoS architecture requires a simultaneous consideration of design parameters from applications and processing platforms all the way to network configurations. Traditionally, however, systems and networks are designed in isolation and combined in an ad hoc manner, which ignores joint effects and optimization opportunities. To systematically explore and optimize NoS design spaces, a higher level of design abstraction on top of traditional system and network design is required. In this article, we propose a novel network-level design methodology for resource-constrained NoS optimization and design space exploration. A key component in such a design flow is fast yet accurate network/system co-simulation to rapidly evaluate NoS parameters with high fidelity. We first introduce a novel NoS simulator (NoSSim) that integrates source-level simulation models of applications with a host-compiled system simulation platform and a reconfigurable network simulation backplane to accurately capture system and network interactions. The co-simulation platform is further combined with model generation tools and a multi-objective genetic search algorithm to provide a comprehensive and fully automated NoS design space exploration framework. Finally, we apply our network-level design flow on several state-of-art IoT/mobile design case studies. Results show that NoSSim can achieve more than 86% simulation accuracy on average as compared to a real-world edge device cluster, where sensitivities to various design parameters are faithfully captured with high fidelity. When applying our network-level design space exploration methodology, design decisions are automatically optimized, where non-obvious NoS configurations are discovered outperforming manually designed solutions by more than 45%.

The fog-computing paradigm has appeared as a geo distributed response to a growing focus on latency-sensitive Internet-of-Things (IoT) applications and the long delay that may be provided by cloud data centers. Although many researchers have investigated how IoT can interact with a fog, very few have tackled the question of how to construct a fog infrastructure for the expected IoT traffic. This article addresses the design and dimensioning of a fog infrastructure via a mixed-integer linear program (MILP) to construct a physical fog network design by mapping IoT virtual networks to dimensioned fog nodes. Due to the exponential nature of this MILP formulation, we also propose a column generation model with near-optimal results at a significantly reduced design and dimensioning cost. The numerical results show the viability of the column generation method in its proximity to the optimal solution and in its reasonable solution time.

In this paper, we envision a hybrid opto-acoustic network design for the internet of underwater things (IoUT). Software-defined underwater networking (SDUN) is presented as an enabler of hybridizing benefits of optic and acoustic systems and adapting IoUT nodes to the challenging and dynamically changing underwater environment. We explain inextricably interwoven relations among functionalities of different layers and analyze their impacts on key network attributes. Network function virtualization (NFV) concept is then introduced to realize application specific cross-layer protocol suites through an NFV management and orchestration system. We finally discuss how SDUN and NFV can slice available network resources as per the diverging service demands of different underwater applications. Such a revolutionary architectural paradigm shift is not only a cure for chronicle underwater networking problems but also a way of smoothly integrating IoUT and IoT ecosystems., Submitted to IEEE Communications Magazine

We describe two mixed-integer linear programming formulations, one a faster version of a previous proposal, the other a slower but better performing new model, for the design of Quantum Key Distribution (QKD) sub-networks dimensioned to secure existing core fiber plants. We exploit existing technologies, including non-quantum repeater nodes and multiple disjoint QKD paths to overcome reach limitations while maintaining security guarantees. We examine the models&rsquo, performance using simulations on both synthetic and real topologies, quantifying their time and resulting QKD network cost compared to our previous proposal.

Flow-guided nanocommunication networks comprise a new paradigm that enables among other things, new medical applications for monitoring, information gathering, and data transmission inside the human body. Although promising, this type of network is conceived to operate in a very hostile environment (e.g. blood circulatory system) where nano-node mobility, along with reduced coverage hinder successful transmissions. In addition to these concerns, it is necessary to take into account the scarce amount of energy available in nano-nodes and the important limitations to harvesting energy. Thus, it is essential to carefully select network parameters to tailor network design and improve performance or satisfy specific application requirements and objectives. To this end, this article presents an optimization framework to maximize throughput by selecting the appropriate frame size as a function of the available energy, nano-router location, and transmission rate. This optimization is conducted for different scenarios where the study can reveal whether a general setup that can work in almost all situations is desirable or, depending on the application, a pre-setting customization of network parameters before deployment is more advantageous in terms of performance. Maximizing throughput ensures information transmission and establishes the maximum amount of data that a nano-router can receive. These two questions are critical for the development of reliable medical applications that can operate within the capabilities that flow-guided nano-networks can offer. Thus, the main technical contributions of this paper are: i) the extension of the flow-guided nano-network analytical model, and ii) the derivation of a throughput optimization framework for this type of networks, which provides a suitable frame size as a function of different network parameters such as flow velocity, nano-router location, and transmission power.

With the advent of Internet of Vehicles (IoV), coupled with enormous number of devices performing computational and storage tasks between the cloud and users, Vehicular Fog Computing (VFC) can be an answer to the surging challenges in today’s Intelligent Transportation Systems (ITS). However, the decentralized and heterogeneous nature of VFC infrastructures makes Vehicular Fog Network Planning (VFNP) problem complex and challenging. To deal with this problem, we propose an Integer Linear Programming (ILP) model that determines the optimal location, the capacity and the number of Fog Computing Nodes (FCN) towards minimizing the overall network delay and energy consumption. By running an example problem on default settings of GAMS CPLEX solver, we demonstrate the working of VFNP model and the associated constraints. We also analyzed the delay and energy variation for different problem sizes. The results show that, as the input size increases the overall delay increases linearly, the energy consumption follows parabolic path and the solution time shows a non-deterministic polynomial (NP) behavior.

Distributed Energy Resources (DER), mainly residential solar PV, are embedded deep within the power distribution network and their adoption is fast increasing globally. As more customers participate, these power generation units cause Reverse Power Flow (RPF) at the edge of the grid, directed upstream into the network, thus violating one of the traditional design principles for power networks. The effects of a single residential solar PV system is negligible, but as the adoption by end-consumers increases to high percentages, the aggregated effect is no longer negligible and must be considered in the design and configuration of power networks. This article proposes a framework that helps to predict the RPF intensity probability for any given scenario of DER penetration within the distribution network. The considered scenario parameters are the number and location of each residential DERs, their capacity and the daily net-load profiles. Classical simulation-based approach for this is not scalable as it relies on solving the load-flow equations for each individual scenario. The framework leverages machine learning techniques to make fast and precise RPF prediction within the network for each scenario. The framework enables the Distribution Network Service Providers (DNSPs) to assess DERs penetration scenarios at a granular level, derive and localise the RPF risks and assess the respective impacts on the installed assets for network planning purpose. The framework is illustrated with scenario analysis conducted on an IEEE 123 bus system and OpenDSS and shown that it can lead to multiple orders of magnitude savings in computational time while retaining an accuracy of 94% or above compared to classical brute force simulations.

In this paper, we consider a power allocation optimization technique for a time-varying fading broadcast channel in energy harvesting communication systems, in which a transmitter with a rechargeable battery transmits messages to receivers using the harvested energy. We first prove that the optimal online power allocation policy for the sum-rate maximization of the transmitter is a monotonically increasing function of harvested energy, remaining battery, and the channel gain of each user. We then construct a lightweight neural network architecture to take advantage of the monotonicity of the optimal policy. This two-step approach, which relies on effective function approximation to provide a mathematical guideline for neural network design, can prevent us from wasting the representational capacity of neural networks. The tailored neural network architectures enable stable learning and eliminate the heuristic network design. For performance assessment, the proposed approach is compared with the closed-form optimal policy for a partially observable Markov problem. Through additional experiments, it is observed that our online solution achieves a performance close to the theoretical upper bound of the performance in a time-varying fading broadcast channel.

Cloud-based computing technology is one of the most significant technical advents of the last decade and extension of this facility towards access networks by aggregation of cloudlets is a step further. To fulfill the ravenous demand for computational resources entangled with the stringent latency requirements of computationally-heavy applications related to augmented reality, cognitive assistance and context-aware computation, installation of cloudlets near the access segment is a very promising solution because of its support for wide geographical network distribution, low latency, mobility and heterogeneity. In this paper, we propose a novel framework, Cloudlet Cost OptiMization over PASSIve Optical Network (CCOMPASSION), and formulate a nonlinear mixed-integer program to identify optimal cloudlet placement locations such that installation cost is minimized whilst meeting the capacity and latency constraints. Considering urban, suburban and rural scenarios as commonly-used network deployment models, we investigate the feasibility of the proposed model over them and provide guidance on the overall cloudlet facility installation over optical access network. We also study the percentage of incremental energy budget in the presence of cloudlets of the existing network. The final results from our proposed model can be considered as fundamental cornerstones for network planning with hybrid cloudlet network architectures., This paper is published in 2018 IEEE Conference on Computer Communications (INFOCOM). Copyright @ IEEE

Telecommunications networks are facing increasing demand for Internet services. Therefore, the problem of telecommunications network design with the objective to maximize service data flows and provide fair treatment of all services is very up-to-date. In this application, the so-called maxmin fair (MMF) solution concept is widely used to formulate the resource allocation scheme. It assumes that the worst service performance is maximized and the solution is additionally regularized with the lexicographic maximization of the second worst performance, the third one, etc. In this paper we discuss solution algorithms for MMF problems related to telecommunications network design. Due to lexicographic maximization of ordered quantities, the MMF solution concept cannot be tackled by the standard optimization model (mathematical programme). However, one can formulate a sequential lexicographic optimization procedure. The basic procedure is applicable only for convex models, thus it allows to deal with basic design problems but fails if practical discrete restrictions commonly arriving in telecommunications network design are to be taken into account. Then, however, alternative sequential approaches allowing to solve non-convex MMF problems can be used. Keywords—network design, resource allocation, fairness, lexicographic optimization, lexicographic max-min.

The emergence of electromobility along with recent developments in wireless power transfer (WPT) technology offer potentials to improve the carbon footprint of bus transport, while offering quality services. Indeed, the deployment of fast charging stations and dynamic charging roadway segments (lanes) can ensure fast energy transmission to electricity-powered buses, mitigating existing energy-related concerns and limitations. Existing models for public transport network design cannot adequately capture the dependence between electric vehicle charging infrastructure requirements and route operational characteristics. In this context, this paper investigates the combined Transit Route Network Design and Charging Infrastructure Location Problem and proposes a bi-level formulation to handle both planning stages. At the upper level, candidate route sets are generated and evaluated, while at the lower-level wireless charging infrastructures are optimally deployed. A multi-objective Particle Swarm Optimization (MO-PSO) algorithm embedded with an integer programming solver is employed to handle the complexity of the problem and the conflicting design objectives related to passengers and operators. The resulting model is applied to an established benchmark network to assess the tradeoffs arising between user-oriented and operator-oriented solutions and highlight the complex decision process associated with the deployment of electric public transport networks.

We address the problem of designing a 3-architecture telecommunication access networks that combines optical fiber- and copper-based access technologies with Free Space Optics (FSO) technology. We propose a mixed integer linear programming model for representing the optimal design problem, which can be traced back to a multi-architecture variant of the well-known Connected Facility Location problem. In order to protect design solutions against reduction in capacity that may occur in FSO connections due to adverse weather conditions, we adopt a robust cutting plane procedure. Preliminary computational tests are reported to show the ability of our robust optimization approach to produce weather-resistant design solutions.

Video conferencing applications have seen explosive growth both in the number of available applications and their use. However, there have been few studies on the detailed analysis of video conferencing applications with respect to network dynamics, yet understanding these dynamics is essential for network design and improving these applications. In this paper, we carry out an in-depth measurement and modeling study on the rate control algorithms used in six popular commercial video conferencing applications. Based on macroscopic behaviors commonly observed across these applications in our extensive measurements, we construct a unified architecture to model the rate control mechanisms of individual applications. We then reconstruct each application's rate control by inferring key parameters that closely follow its rate control and quality adaptation behaviors. To our knowledge, this is the first work that reverse-engineers rate control algorithms of popular video conferencing applications, which are often unknown or hidden as they are proprietary software. We confirm our analysis and models using an end-to-end testbed that can capture the dynamics of each application under a variety of network conditions. We also show how we can use these models to gain insights into the particular behaviors of an application in two practical scenarios.

Throughput prediction remains a relevant and challenging problem in the area of wireless networking. Both users and service providers would benefit from the predictive Quality of Service (QoS) which has the potential to improve perceived service quality by users but also provide valuable insights for the network planning, optimization, and deployment. There are many factors that can hinder the adoption of Machine Learning (ML) models, such as the data collection overhead for the network operator, prediction uncertainties of an ML model and the lack of transparency in the ML inference process. In this paper, we provide results on the instantaneous throughput prediction using ML techniques. The data for the throughput prediction was collected in a live network of one of the biggest operators in Asia which improves confidence of the results. We also discuss how custom loss functions can extract value from prediction errors. Finally, we look at explainability methods providing more transparency on the predictions of ML algorithms. Various explainability methods prove that without being explicitly programmed, ML algorithms can exploit and learn from the underlying physical phenomena and operating principles of wireless networks.

Distribution networks are exposed to new challenges due to the evolution of technology, requiring planning engineers to perform studies with greater regularity with more detailed electric models, to improve network decision-making processes. This paper proposes an approach to periodically build, a validated OpenDSS distribution network model for network planning, converting electric elements of Geographic Information System to Common Information Model files, which can be used to feed Distribution System Management and OpenDSS platforms, closing gaps between these two systems. The proposal enables a fast conversion of 100% of a distribution network, reducing time for decision making and giving more robustness to planning studies and further analysis. Simulation results demonstrate the effectiveness of the proposed approach in creating suitable and open-source network models, with high accuracy, for a variety of studies for network planning helping Smart Grid implementation and power quality enhancements.

Data representation, storage and processing in Internet of Things (IoT) devices must be done such that the network lifetime and the quality of service (QoS) for the network is minimally affected. In order solve this constraint researchers use sleep scheduling, aggregation, effective data representation, pre-processing, and other techniques. Each of these techniques has its own nuances and advantages. Techniques which involve sleep scheduling improve the network lifetime, but reduce other QoS parameters like speed of operation, while techniques involving pre-processing of data guarantees high throughput but increases energy consumption in the network. A statistical survey of these techniques is mentioned in this text. Moreover, this text also recommends a set of approaches that can be applied in order to improve the performance of IoT networks via better data representation, enhanced data storage structures and effective data analysis techniques. Use of Machine learning and blockchain technologies for improving QoS and security performance of the network are also evaluated. Using this text researchers in the IoT domain can evaluate techniques needed for effective IoT network design and improve the QoS parameters of the network.

Mobile Edge Computing (MEC) is a key technology for the deployment of next generation (5G and beyond) mobile networks. The computational power it provides at the edge could allow providers to fulfill the requirements of use cases in need of ultra-low latency, high bandwidth, as well as real-time access to the radio network. However, this potential needs to be carefully administered as the edge is certainly limited in terms of computation capability, as opposed to the cloud which holds the promise of a virtually infinite power. MEC nodes, though, could still try to exploit not only their local capacity, but also the one that the neighbor MEC nodes could offer. Considering that the 5G scenario assumes an ultra-dense distribution of MEC nodes, this possibility could be feasible, provided that we find an effective way to carefully allocate the resources available at each edge node. In this paper, we provide an optimization framework that considers several key aspects of the resource allocation problem with cooperating MEC nodes. We carefully model and optimize the allocation of resources, including computation and storage capacity available in network nodes as well as link capacity. Specifically, our proposed model jointly optimizes (1) the user requests admission decision (2) their scheduling, also called calendaring (3) and routing as well as (4) the decision of which nodes will serve such user requests and (5) the amount of processing and storage capacity reserved on the chosen nodes. Both an exact optimization model and an effective heuristic, based on sequential fixing, are provided. Furthermore, we propose a distributed approach for our problem, based on the Alternating Direction Method of Multipliers (ADMM), so that resource allocation decisions can be made in a distributed fashion by edge nodes with limited overhead. We perform an extensive numerical analysis in several real-size network scenarios, using real positions for radio access points of a mobile operator in the Milan area. Results demonstrate that the heuristic performs close to the optimum in all considered network scenarios, while exhibiting a low computing time. This provides an evidence that our proposal is an effective framework for optimizing resource allocation in next-generation mobile networks.

Over the years, the need for communication networks capable of providing an ever-increasing set of services has grown. In order to satisfy user requirements and provide guarantees of reliability of the network itself, efficient techniques are required for analysis, evaluation and design. For this reason, the need arises to have models able to represent the peculiar characteristics of network traffic and to produce reliable predictions of its behavior in an adequate period of time. Therefore, network traffic prediction plays an important role by supporting many practical applications, ranging from network planning and provisioning to security. Several works so far have focused on building app-specific models. However, this choice produces multiple models that need to be properly managed and deployed across network devices. Therefore, in this paper, we explore different training strategies to reduce the number of models, adopting the Markov Chains to model mobile video apps traffic at packet-level. We discuss and experimentally evaluate the prediction effectiveness of the proposed approaches by comparing the performance of app models with models trained on a specific category of video apps and a model trained on the mix of all video traffic.

A passive in-building distributed antenna system (IB-DAS) is often used to enhance indoor mobile data coverage by introducing indoor antennas inside buildings. Such systems are created to ensure that traffic generated indoors does not heavily depend on base stations installed outdoor, as penetration issues of wireless signals can affect the quality of connection. The focus of this paper is on the automation of IB-DAS design. Particularly, it provides an extensive analysis of the performance of four binary Evolutionary Algorithms (EAs) for this problem and shows that the two tested Estimation of Distribution Algorithms (EDAs) performs well on this problem. Furthermore, it investigates the effect of different genetic operators on the performance of the considered EAs. The practice outcome of this is to select the best algorithm among others to be implemented in a DAS network planning tool and to help our industrial partner reduce both the design time and deployment cost.

The core networks of current telecommunication infrastructures are typically engineered as multi-layer networks. The uppermost layer is defined by the virtual topology, which determines the logical connections between core network routers. This topology is realized by optical paths in the lower layer, which is defined by optical fiber connections between network nodes. Minimizing the hardware cost incurred by these optical paths for a given set of traffic demands is a common combinatorial optimization problem in network planning, often approached by Mixed-Integer Linear Programming. However, increasing network densities and the introduction of additional constraints will impact tractability of future network problems. In order to provide a more scalable method, we suggest a Genetic Algorithm-based approach that optimizes the virtual topology and subsequently derives the remaining parameters from it. Our genetic encoding utilizes a combination of spanning trees and augmentation links to quickly form meaningful topologies. We compare the results of our approach to known linear programming solutions in simple scenarios and to a competing heuristic based on Simulated Annealing in large-scale problems.

Emerging optical communication technologies support the dynamic reconfiguration of datacenter network topologies depending on the traffic they serve. However, to reap the benefits of such demand-aware networks, a control logic is required which allows to quickly learn and adapt to traffic patterns. This paper presents CacheNet, a novel approach to efficiently control demand-aware networks. CacheNet leverages temporal and spatial locality in the traffic by managing the reconfigurable links of the optical switches as a links-cache. Network traffic, in turn, can be served either by a link from the link-cache component or by a demand-oblivious topology component. We study several classic caching algorithms and provide an analytical model which captures their performance benefits compared to an all demand-oblivious topology. Our analytical results show that based on the hit ratios and the links-cache size, our hybrid design can outperform designs that are based only on demand-oblivious topology.

Traffic Matrix (TM) clearly describes the volume and the distribution of traffic flows inside a network. TM plays an important role in many network management fields, such as traffic accounting, short-time traffic scheduling or re-routing, network design, anomaly detection, etc. Hence, an accurate TM prediction strategy is essential to handle those tasks effectively. Fortunately, Artificial Intelligence (AI) has been developing very strongly, thanks to computer technology developments such as GPU and TPU. That offers an opportunity to apply AI to TM prediction methods. However, applying Machine Learning techniques in traditional networks encounters some issues due to the distributed control and restricted local view of network nodes. For this purpose, a centralized control architecture, e.g., Software-defined Network (SDN), is a promising candidate. In this paper, we apply Long Short-Term Memory (LSTM) and its two variants, Bidirectional LSTM (BiLSTM) and Gate Recurrent Unit (GRU), for TM prediction mechanisms of an SDN architecture network. The prediction models have been evaluated using two datasets: the popular GEANT backbone network traffic data and our dataset generated through a testbed. The experimental results show that our approach yielded promising traffic prediction accuracy.

Large Low Earth Orbit broadband satellite networks are quickly becoming viable alternatives to ground-based or terrestrial networks as a number of implementations are currently being constructed by commercial service providers. Unlike previous systems, these new constellations are comprised of 100’s to 1,000’s of interconnected satellites. This paper presents a novel simulation tool that enables the study of large satellite constellations with focus on inter-satellite and satellite-to-ground station network design. We present the design and implementation of the simulator tool and demonstrate its use to analyze network designs and constellation structures with regards to routing performance.

Deploying Sub-6GHz networks together with millimeter wave (mmWave) is a promising solution to achieve high data rates in traffic hotspots while guaranteeing sufficient coverage, where mmWave small cells are densely deployed to provide high quality of service. In this paper, we propose an analytical framework to investigate the integrated Sub-6GHz-mmWave cellular networks, in which the Sub-6GHz base stations (BSs) are modeled as a Poisson point process, and the mmWave BSs are clustered following a Poisson cluster process in traffic hotspots. We conduct stochastic geometry-based analysis and derive the performance metrics including the association probability, signal-to-interference-plus-noise ratio coverage probability and average achievable rate, which are validated to be accurate by Monte Carlo simulations. We analyze the impact of various deployment parameters on the network performance to give insights on the network design. In particular, it is shown that deploying mmWave small cells in traffic hotspots will outperform both traditional Sub-6GHz heterogeneous network and isolated mmWave system in terms of the coverage probability. It can also be shown that extremely high and extremely small association weight for mmWave BSs will deteriorate the performance for cell edge users and cell interior users, respectively. Moreover, there exists an optimal pre-decided dispersion parameter of mmWave BSs that contributes to the maximum coverage probability., 28 pages, 16 figures, to appear in IEEE Transactions on Communications