Showing papers on "Network topology published in 2019"

A Survey on Network Embedding

Abstract: Network embedding assigns nodes in a network to low-dimensional representations and effectively preserves the network structure. Recently, a significant amount of progresses have been made toward this emerging network analysis paradigm. In this survey, we focus on categorizing and then reviewing the current development on network embedding methods, and point out its future research directions. We first summarize the motivation of network embedding. We discuss the classical graph embedding algorithms and their relationship with network embedding. Afterwards and primarily, we provide a comprehensive overview of a large number of network embedding methods in a systematic manner, covering the structure- and property-preserving network embedding methods, the network embedding methods with side information, and the advanced information preserving network embedding methods. Moreover, several evaluation approaches for network embedding and some useful online resources, including the network data sets and softwares, are reviewed, too. Finally, we discuss the framework of exploiting these network embedding methods to build an effective system and point out some potential future directions.

The Marabou Framework for Verification and Analysis of Deep Neural Networks

Abstract: Deep neural networks are revolutionizing the way complex systems are designed. Consequently, there is a pressing need for tools and techniques for network analysis and certification. To help in addressing that need, we present Marabou, a framework for verifying deep neural networks. Marabou is an SMT-based tool that can answer queries about a network’s properties by transforming these queries into constraint satisfaction problems. It can accommodate networks with different activation functions and topologies, and it performs high-level reasoning on the network that can curtail the search space and improve performance. It also supports parallel execution to further enhance scalability. Marabou accepts multiple input formats, including protocol buffer files generated by the popular TensorFlow framework for neural networks. We describe the system architecture and main components, evaluate the technique and discuss ongoing work.

Connecting the Dots: Identifying Network Structure via Graph Signal Processing

Abstract: Network topology inference is a significant problem in network science. Most graph signal processing (GSP) efforts to date assume that the underlying network is known and then analyze how the graph?s algebraic and spectral characteristics impact the properties of the graph signals of interest. Such an assumption is often untenable beyond applications dealing with, e.g., directly observable social and infrastructure networks; and typically adopted graph construction schemes are largely informal, distinctly lacking an element of validation. This article offers an overview of graph-learning methods developed to bridge the aforementioned gap, by using information available from graph signals to infer the underlying graph topology. Fairly mature statistical approaches are surveyed first, where correlation analysis takes center stage along with its connections to covariance selection and high-dimensional regression for learning Gaussian graphical models. Recent GSP-based network inference frameworks are also described, which postulate that the network exists as a latent underlying structure and that observations are generated as a result of a network process defined in such a graph. A number of arguably more nascent topics are also briefly outlined, including inference of dynamic networks and nonlinear models of pairwise interaction, as well as extensions to directed (di) graphs and their relation to causal inference. All in all, this article introduces readers to challenges and opportunities for SP research in emerging topic areas at the crossroads of modeling, prediction, and control of complex behavior arising in networked systems that evolve over time.

Prescribed-Time Consensus and Containment Control of Networked Multiagent Systems

Abstract: In this paper, we present a new prescribed-time distributed control method for consensus and containment of networked multiple systems. Different from both regular finite-time control (where the finite settling time is not uniform in initial conditions) and the fixed-time control (where the settling time cannot be preassigned arbitrarily), the proposed one is built upon a novel scaling function, resulting in prespecifiable convergence time (the settling time can be preassigned as needed within any physically allowable range). Furthermore, the developed control scheme not only ensures that all the agents reach the average consensus in prescribed finite time under undirected connected topology, but also ensures that all the agents reach a prescribed-time consensus with the root’s state being the group decision value under the directed topology containing a spanning tree with the root as the leader. In addition, we extend the result to prescribed-time containment control involving multiple leaders under directed communication topology. Numerical examples are provided to verify the effectiveness and the superiority of the proposed control.

Optimal Design of a New Cascaded Multilevel Inverter Topology With Reduced Switch Count

Abstract: Multilevel inverters (MLIs) are a great development for industrial and renewable energy applications due to their dominance over conventional two-level inverter with respect to size, rating of switches, filter requirement, and efficiency. A new single-phase cascaded MLI topology is suggested in this paper. The proposed MLI topology is designed with the aim of reducing the number of switches and the number of dc voltage sources with modularity while having a higher number of levels at the output. For the determination of the magnitude of dc voltage sources and a number of levels in the cascade connection, three different algorithms are proposed. The optimization of the proposed topology is aimed at achieving a higher number of levels while minimizing other parameters. A detailed comparison is made with other comparable MLI topologies to prove the superiority of the proposed structure. A selective harmonic elimination pulse width modulation technique is used to produce the pulses for the switches to achieve high-quality voltage at the output. Finally, the experimental results are provided for the basic unit with 11 levels and for cascading of two such units to achieve 71 levels at the output.

Deep Mesh Reconstruction From Single RGB Images via Topology Modification Networks

Abstract: Reconstructing the 3D mesh of a general object from a single image is now possible thanks to the latest advances of deep learning technologies. However, due to the nontrivial difficulty of generating a feasible mesh structure, the state-of-the-art approaches often simplify the problem by learning the displacements of a template mesh that deforms it to the target surface. Though reconstructing a 3D shape with complex topology can be achieved by deforming multiple mesh patches, it remains difficult to stitch the results to ensure a high meshing quality. In this paper, we present an end-to-end single-view mesh reconstruction framework that is able to generate high-quality meshes with complex topologies from a single genus-0 template mesh. The key to our approach is a novel progressive shaping framework that alternates between mesh deformation and topology modification. While a deformation network predicts the per-vertex translations that reduce the gap between the reconstructed mesh and the ground truth, a novel topology modification network is employed to prune the error-prone faces, enabling the evolution of topology. By iterating over the two procedures, one can progressively modify the mesh topology while achieving higher reconstruction accuracy. Moreover, a boundary refinement network is designed to refine the boundary conditions to further improve the visual quality of the reconstructed mesh. Extensive experiments demonstrate that our approach outperforms the current state-of-the-art methods both qualitatively and quantitatively, especially for the shapes with complex topologies.

Topologies and Control Schemes of Bidirectional DC–DC Power Converters: An Overview

Abstract: Bidirectional DC-DC power converters are increasingly employed in diverse applications whereby power flow in both forward and reverse directions are required. These include but not limited to energy storage systems, uninterruptable power supplies, electric vehicles, and renewable energy systems, to name a few. This paper aims to review these converters from the point of view of topology as well as control schemes. From the point of view of topology, these converters are divided into two main categories, namely non-isolated and isolated configurations. Each category is divided into eight groups along with their respective schematics and a table of summary. Furthermore, the common control schemes and switching strategies for these converters are also reviewed. Some of the control schemes are typically applied to all DC-DC power converters such as PID, sliding mode, fuzzy, model predictive, digital control, etc. In this context, it should be noted that some switching strategies were designed specifically for isolated bidirectional DC-DC converters in order to improve their performance such as single phase shift, dual phase shift, triple phase shift, etc. The features of each topology and control scheme along with their typical applications are discussed, in order to provide a ground of comparison for realizing new configurations or finding the appropriate converter for the specific application.

Consistency and differences between centrality measures across distinct classes of networks.

Abstract: The roles of different nodes within a network are often understood through centrality analysis, which aims to quantify the capacity of a node to influence, or be influenced by, other nodes via its connection topology. Many different centrality measures have been proposed, but the degree to which they offer unique information, and whether it is advantageous to use multiple centrality measures to define node roles, is unclear. Here we calculate correlations between 17 different centrality measures across 212 diverse real-world networks, examine how these correlations relate to variations in network density and global topology, and investigate whether nodes can be clustered into distinct classes according to their centrality profiles. We find that centrality measures are generally positively correlated to each other, the strength of these correlations varies across networks, and network modularity plays a key role in driving these cross-network variations. Data-driven clustering of nodes based on centrality profiles can distinguish different roles, including topological cores of highly central nodes and peripheries of less central nodes. Our findings illustrate how network topology shapes the pattern of correlations between centrality measures and demonstrate how a comparative approach to network centrality can inform the interpretation of nodal roles in complex networks.

Novel self-adaptive routing service algorithm for application in VANET

Abstract: As a special MANET (mobile ad hoc network), VANET (vehicular ad-hoc network) has two important properties: the network topology changes frequently, and communication links are unreliable. Both properties are caused by vehicle mobility. To predict the reliability of links between vehicles effectively and design a reliable routing service protocol to meet various QoS application requirements, in this paper, details of the motion characteristics of vehicles and the reasons that cause links to go down are analyzed. Then a link duration model based on time duration is proposed. Link reliability is evaluated and used as a key parameter to design a new routing protocol. Quick changes in topology make it a huge challenge to find and maintain the end-to-end optimal path, but the heuristic Q-Learning algorithm can dynamically adjust the routing path through interaction with the surrounding environment. This paper proposes a reliable self-adaptive routing algorithm (RSAR) based on this heuristic service algorithm. By combining the reliability parameter and adjusting the heuristic function, RSAR achieves good performance with VANET. With the NS-2 simulator, RSAR performance is proved. The results show that RSAR is very useful for many VANET applications.

On Constructing Multiple Lyapunov Functions for Tracking Control of Multiple Agents With Switching Topologies

Abstract: Distributed consensus tracking for linear multiagent systems (MASs) with directed switching topologies and a dynamic leader is investigated in this paper. By fully considering the special feature of Laplacian matrices for topology candidates, several new classes of multiple Lyapunov functions (MLFs) are constructed in this paper for leader-following MASs with, respectively, an autonomous leader and a nonautonomous leader. Under the condition that each possible topology graph contains a spanning tree rooted at the leader node, some efficient criteria for achieving consensus tracking in the considered MASs are provided. Specifically, it is proven that consensus tracking in the closed-loop MASs can be ensured if the average dwell time for switching among different topologies is larger than a derived positive quantity and the control parameters in tracking protocols are appropriately designed. It is further theoretically shown that the present Lyapunov inequality based criteria for consensus tracking with an autonomous leader are much less conservative than the existing ones derived by the $M$ -matrix theory. The results are then extended to the case where the topology graph only frequently contains a directed spanning tree as the MASs evolve over time. At last, numerical simulations are performed to illustrate the effectiveness of the analytical analysis and the advantages of the proposed MLFs.

Secure and Privacy-Preserving Consensus

Abstract: Consensus is fundamental for distributed systems since it underpins key functionalities of such systems ranging from distributed information fusion, decision making, to decentralized control. In order to reach an agreement, existing consensus algorithms require each agent to exchange explicit state information with its neighbors. This leads to the disclosure of private state information, which is undesirable in cases where privacy is of concern. In this paper, we propose a novel approach for undirected networks, which can enable secure and privacy-preserving average consensus in a decentralized architecture in the absence of an aggregator or third party. By leveraging partial homomorphic cryptography to embed secrecy in pairwise interaction dynamics, our approach can guarantee convergence to the consensus value (subject to a quantization error) in a deterministic manner without disclosing a node's state to its neighbors. We provide a new privacy definition for dynamical systems, and give a new framework to rigorously prove that a node's privacy can be protected as long as it has at least one legitimate neighbor, which follows the consensus protocol faithfully without attempts to infer other nodes’ states. In addition to enabling resilience to passive attackers aiming to steal state information, the approach also allows easy incorporation of defending mechanisms against active attackers who try to alter the content of exchanged messages. Furthermore, in contrast to existing noise-injection-based privacy-preserving mechanisms that have to reconfigure the entire network when the topology or number of nodes varies, our approach is applicable to dynamic environments with time-varying coupling topologies. This secure and privacy-preserving approach is also applicable to weighted average consensus as well as maximum/minimum consensus under a new update rule. Numerical simulations and comparison with existing approaches confirm the theoretical results. Experimental results on a Raspberry-Pi board based microcontroller network are also presented to verify the effectiveness and efficiency of the approach.

A New Multilevel Inverter Topology With Reduce Switch Count

Abstract: Multilevel inverters are a new family of converters for dc-ac conversion for the medium and high voltage and power applications. In this paper, two new topologies for the staircase output voltage generations have been proposed with a lesser number of switch requirement. The first topology requires three dc voltage sources and ten switches to synthesize 15 levels across the load. The extension of the first topology has been proposed as the second topology, which consists of four dc voltage sources and 12 switches to achieve 25 levels at the output. Both topologies, apart from having lesser switch count, exhibit the merits in terms of reduced voltage stresses across the switches. In addition, a detailed comparative study of both topologies has been presented in this paper to demonstrate the features of the proposed topologies. Several experimental results have been included in this paper to validate the performances of the proposed topologies with different loading condition and dynamic changes in load and modulation indexes.

A Robust Consensus Algorithm for Current Sharing and Voltage Regulation in DC Microgrids

Abstract: In this paper, a novel distributed control algorithm for current sharing and voltage regulation in DC microgrids is proposed. The DC microgrid is composed of several distributed generation units, including buck converters and current loads. The considered model permits an arbitrary network topology and is affected by an unknown load demand and modeling uncertainties. The proposed control strategy exploits a communication network to achieve proportional current sharing using a consensus-like algorithm. Voltage regulation is achieved by constraining the system to a suitable manifold. Two robust control strategies of sliding mode type are developed to reach the desired manifold in a finite time. The proposed control scheme is formally analyzed, proving the achievement of proportional current sharing, while guaranteeing that the weighted average voltage of the microgrid is identical to the weighted average of the voltage references.

Delay-Minimization Routing for Heterogeneous VANETs With Machine Learning Based Mobility Prediction

Abstract: Establishing and maintaining end-to-end connections in a vehicular ad hoc network (VANET) is challenging due to the high vehicle mobility, dynamic inter-vehicle spacing, and variable vehicle density. Mobility prediction of vehicles can address the aforementioned challenge, since it can provide a better routing planning and improve overall VANET performance in terms of continuous service availability. In this paper, a centralized routing scheme with mobility prediction is proposed for VANET assisted by an artificial intelligence powered software-defined network (SDN) controller. Specifically, the SDN controller can perform accurate mobility prediction through an advanced artificial neural network technique. Then, based on the mobility prediction, the successful transmission probability and average delay of each vehicle's request under frequent network topology changes can be estimated by the roadside units (RSUs) or the base station (BS). The estimation is performed based on a stochastic urban traffic model in which the vehicle arrival follows a non-homogeneous Poisson process. The SDN controller gathers network information from RSUs and BS that are considered as the switches. Based on the global network information, the SDN controller computes optimal routing paths for switches (i.e., BS and RSU). While the source vehicle and destination vehicle are located in the coverage area of the same switch, further routing decision will be made by the RSUs or the BS independently to minimize the overall vehicular service delay. The RSUs or the BS schedule the requests of vehicles by either vehicle-to-vehicle or vehicle-to-infrastructure communication, from the source vehicle to the destination vehicle. Simulation results demonstrate that our proposed centralized routing scheme outperforms others in terms of transmission delay, and the transmission performance of our proposed routing scheme is more robust with varying vehicle velocity.

Scalability Aspects of Cell-Free Massive MIMO

Abstract: Ubiquitous cell-free massive MIMO (multiple-input multiple-output) combines massive MIMO technology and user-centric transmission in a distributed architecture. All the access points (APs) in the network cooperate to jointly and coherently serve a smaller number of users in the same time-frequency resource. However, this coordination needs significant amounts of control signalling which introduces additional overhead, while data co-processing increases the back/front-haul requirements. Hence, the notion that the “whole world” could constitute one network, and that all APs would act as a single base station, is not scalable. In this study, we address some system scalability aspects of cell-free massive MIMO that have been neglected in literature until now. In particular, we propose and evaluate a solution related to data processing, network topology and power control. Results indicate that our proposed framework achieves full scalability at the cost of a modest performance loss compared to the canonical form of cell-free massive MIMO.

Robustness Optimization Scheme With Multi-Population Co-Evolution for Scale-Free Wireless Sensor Networks

Abstract: Wireless sensor networks (WSNs) have been the popular targets for cyberattacks these days. One type of network topology for WSNs, the scale-free topology, can effectively withstand random attacks in which the nodes in the topology are randomly selected as targets. However, it is fragile to malicious attacks in which the nodes with high node degrees are selected as targets. Thus, how to improve the robustness of the scale-free topology against malicious attacks becomes a critical issue. To tackle this problem, this paper proposes a Robustness Optimization scheme with multi-population Co-evolution for scale-free wireless sensor networKS (ROCKS) to improve the robustness of the scale-free topology. We build initial scale-free topologies according to the characteristics of WSNs in the real-world environment. Then, we apply our ROCKS with novel crossover operator and mutation operator to optimize the robustness of the scale-free topologies constructed for WSNs. For a scale-free WSNs topology, our proposed algorithm keeps the initial degree of each node unchanged such that the optimized topology remains scale-free. Based on a well-known metric for the robustness against malicious attacks, our experiment results show that ROCKS roughly doubles the robustness of initial scale-free WSNs, and outperforms two existing algorithms by about 16% when the network size is large.

Routing Protocols for Unmanned Aerial Vehicle Networks: A Survey

Abstract: Unmanned aerial vehicles (UAVs) have gained popularity for diverse applications and services in both the military and civilian domains. For cooperation and collaboration among UAVs, they can be wirelessly interconnected in an ad hoc manner, resulting in a UAV network. UAV networks have unique features and characteristics that are different from mobile ad hoc networks and vehicular ad hoc networks. The dynamic behavior of rapid mobility and topology changes in UAV networks makes the design of a routing protocol quite challenging. In this paper, we review the routing protocols for UAV networks, in which the topology-based, position-based, hierarchical, deterministic, stochastic, and social-network-based routing protocols are extensively surveyed. The routing protocols are then compared qualitatively in terms of their major features, characteristics, and performance. Open issues and research challenges are also discussed in the perspective of design and implementation.

A Fully Data-Driven Method Based on Generative Adversarial Networks for Power System Dynamic Security Assessment With Missing Data

Abstract: This paper proposes a fully data-driven approach for PMU-based pre-fault dynamic security assessment (DSA) with incomplete data measurements. The generative adversarial network (GAN), which is an emerging unsupervised deep learning technique based on two contesting deep neural networks, is used to address the missing data. While the state-of-the-art methods for missing data are dependent on PMU observability, they are limited by the placement of PMU and network topologies. Distinguished from existing methods, the proposed approach is fully data-driven and can fill up incomplete PMU data independent on PMU observability and network topologies. Therefore, it is more generalized and extensible. Simulation results show that, under any PMU missing conditions, the proposed method can maintain a competitively high DSA accuracy with a much less computation complexity.

Deep Neural Network Approximation Theory

Abstract: This paper develops fundamental limits of deep neural network learning by characterizing what is possible if no constraints are imposed on the learning algorithm and on the amount of training data. Concretely, we consider Kolmogorov-optimal approximation through deep neural networks with the guiding theme being a relation between the complexity of the function (class) to be approximated and the complexity of the approximating network in terms of connectivity and memory requirements for storing the network topology and the associated quantized weights. The theory we develop establishes that deep networks are Kolmogorov-optimal approximants for markedly different function classes, such as unit balls in Besov spaces and modulation spaces. In addition, deep networks provide exponential approximation accuracy - i.e., the approximation error decays exponentially in the number of nonzero weights in the network - of the multiplication operation, polynomials, sinusoidal functions, and certain smooth functions. Moreover, this holds true even for one-dimensional oscillatory textures and the Weierstrass function - a fractal function, neither of which has previously known methods achieving exponential approximation accuracy. We also show that in the approximation of sufficiently smooth functions finite-width deep networks require strictly smaller connectivity than finite-depth wide networks.

An Improved Flower Pollination Algorithm for Optimizing Layouts of Nodes in Wireless Sensor Network

Abstract: The arrangement of nodes impacts the quality of connectivity and energy consumption in wireless sensor network (WSN) for prolonging the lifetime. This paper presents an improved flower pollination algorithm based on a hybrid of the parallel and compact techniques for global optimizations and a layout of nodes in WSN. The parallel enhances diversity pollinations for exploring in space search and sharing computation load. The compact can save storing variables for computation in the optimization process. In the experimental section, the selected test functions and the network topology issue WSN are used to test the performance of the proposed approach. Compared results with the other methods in the literature show that the proposed algorithm achieves the practical way of reducing the number of its stored memory variables and running times.

GCN-GAN: A Non-linear Temporal Link Prediction Model for Weighted Dynamic Networks

Abstract: In this paper, we generally formulate the dynamics prediction problem of various network systems (e.g., the prediction of mobility, traffic and topology) as the temporal link prediction task. Different from conventional techniques of temporal link prediction that ignore the potential non-linear characteristics and the informative link weights in the dynamic network, we introduce a novel non-linear model GCN-GAN to tackle the challenging temporal link prediction task of weighted dynamic networks. The proposed model leverages the benefits of the graph convolutional network (GCN), long short-term memory (LSTM) as well as the generative adversarial network (GAN). Thus, the dynamics, topology structure and evolutionary patterns of weighted dynamic networks can be fully exploited to improve the temporal link prediction performance. Concretely, we first utilize GCN to explore the local topological characteristics of each single snapshot and then employ LSTM to characterize the evolving features of the dynamic networks. Moreover, GAN is used to enhance the ability of the model to generate the next weighted network snapshot, which can effectively tackle the sparsity and the wide-value-range problem of edge weights in real-life dynamic networks. To verify the model’s effectiveness, we conduct extensive experiments on four datasets of different network systems and application scenarios. The experimental results demonstrate that our model achieves impressive results compared to the state-of-the-art competitors.

Low Switching Frequency Based Asymmetrical Multilevel Inverter Topology With Reduced Switch Count

Abstract: The inceptions of multilevel inverters (MLI) have caught the attention of researchers for medium and high power applications. However, there has always been a need for a topology with a lower number of device count for higher efficiency and reliability. A new single-phase MLI topology has been proposed in this paper to reduce the number of switches in the circuit and obtain higher voltage level at the output. The basic unit of the proposed topology produces 13 levels at the output with three dc voltage sources and eight switches. Three extentions of the basic unit have been proposed in this paper. A detailed analysis of the proposed topology has been carried out to show the superiority of the proposed converter with respect to the other existing MLI topologies. Power loss analysis has been done using PLECS software, resulting in a maximum efficiency of 98.5%. Nearest level control (NLC) pulse-width modulation technique has been used to produce gate pulses for the switches to achieve better output voltage waveform. The various simulation results have been performed in the PLECS software and a laboratory setup has been used to show the feasibility of the proposed MLI topology.

Coordinated Optimal Network Reconfiguration and Voltage Regulator/DER Control for Unbalanced Distribution Systems

Abstract: Network reconfiguration has long been used by distribution system operators to achieve certain operation objectives such as reducing system losses or regulating bus voltages. In emerging distribution systems with a proliferation of distributed energy resources (DERs), co-optimizing network topology and DERs’ dispatches could further enhance such operational benefits. This paper focuses on the optimal network reconfiguration problem of distribution systems via an unbalanced ac optimal power flow framework, which rigorously addresses operation characters of unbalanced network, DERs, and voltage regulators (VRs). Two VR models with continuous and discrete tap ratios are studied and compared. The proposed co-optimization problem is formulated as a mixed-integer chordal relaxation-based semidefinite programming model with binary variables indicating line-switching statuses and tap positions. Several acceleration strategies by studying the structure of distribution networks are explored for reducing the number of binary variables and enhancing the computational performance. Case studies on modified IEEE 34-bus and 392-bus systems illustrate the effectiveness of the proposed approach.

Topology Optimization Accelerated by Deep Learning

Abstract: The computational cost of topology optimization based on the stochastic algorithm is shown to be greatly reduced by deep learning. In the learning phase, the cross-sectional image of an interior permanent magnet motor, represented in RGB, is used to train a convolutional neural network (CNN) to infer the torque properties. In the optimization phase, all the individuals are approximately evaluated by the trained CNN, while finite element analysis for accurate evaluation is performed only for a limited number of individuals. It is numerically shown that the computational cost for the topology optimization can be reduced without the loss of optimization quality.

New Family of Boost Switched-Capacitor Seven-Level Inverters (BSC7LI)

Abstract: This paper proposes a new family of multilevel inverter topology that is able to generate seven voltage levels by utilizing one or two floating capacitors and 10 power switches. This novel boost switched-capacitor seven-level inverter possesses voltage boosting capability with an achievable maximum voltage level 1.5 times the input direct current (dc) voltage. The generation of higher output voltage does not incur high-voltage stress on any power switch in this topology, as the peak inverse voltages of all power switches do not exceed the input source voltage. In addition, capacitor voltage balancing is not essential since the floating capacitors are effectively balanced during the charging and discharging processes. Furthermore, the proposed topology eliminates the need for multiple isolated dc sources, and a single dc source is sufficient in both its single-phase and three-phase topologies. The operating principle and steady-state analysis of the proposed topology are elaborated. Experimental results from a single-phase prototype are then presented to verify the validity of the proposed topology.