Robustness of spatial networks and networks of networks

Many complex networks have recently been recognized to involve significant interdependence between different systems. Motivation comes primarily from infrastructures like power grids and communications networks, but also includes areas such as the human brain and finance. Interdependence implies that when components in one system fail, they lead to failures in the same system or other systems. This can then lead to additional failures finally resulting in a long cascade that can cripple the entire system. Furthermore, many of these networks, in particular infrastructure networks, are embedded in space and thus have unique spatial properties that significantly decrease their resilience to failures. Here we present a review of novel results on interdependent spatial networks and how cascading processes are affected by spatial embedding. We include various aspects of spatial embedding such as cases where dependencies are spatially restricted and localized attacks on nodes contained in some spatial region of the network. In general, we find that spatial networks are more vulnerable when they are interdependent and that they are more likely to undergo abrupt failure transitions than interdependent non-embedded networks. We also present results on recovery in spatial networks, the nature of cascades due to overload failures in these networks, and some examples of percolation features found in real-world traffic networks. Finally, we conclude with an outlook on future possible research directions in this area.     

Robustness analysis of static routing on networks

Robustness is one of the crucial properties that needs to be considered in the design of routing strategies on networks. We study the robustness of three typical routing strategies, which are the SP (shortest path), EP (efficient path), and OP (optimal path) strategies, by simulating several different kinds of attacks including random attacks, target attacks and cascading failures on scale-free networks. Results of the average path length, betweenness centrality, network capacity, etc., demonstrate that the EP strategy is more robust than the other two, and the OP strategy is more reliable than the SP strategy in general. However, on the power-grid network, the OP strategy is more resistant against cascading failures than the EP and SP strategies.     

The spatial structure of networks

We study networks that connect points in geographic space, such as transportation networks and the Internet. We find that there are strong signatures in these networks of topography and use patterns, giving the networks shapes that are quite distinct from one another and from non-geographic networks. We offer an explanation of these differences in terms of the costs and benefits of transportation and communication, and give a simple model based on the Monte Carlo optimization of these costs and benefits that reproduces well the qualitative features of the networks studied.     

Robustness of networks against cascading failures

Inspired by other related works, this paper proposes a non-linear load-capacity model against cascading failures, which is more suitable for real networks. The simulation was executed on the B-A scale-free network, E-R random network, Internet AS level network, and the power grid of the western United States. The results show that the model is feasible and effective. By studying the relationship between network cost and robustness, we find that the model can defend against cascading failures better and requires a lower investment cost when higher robustness is required.     

Topological efficiency in three-dimensional gallery networks of termite nests

Transport networks are a key component of human and natural societies that enable efficient communication at a low cost. Here, we study the topological efficiency of the three-dimensional networks of galleries in termite nests and how spatial constraints affect the organisation of these networks. Cubitermes termite nests have far better than random transportation efficiency, but they do not reach theoretical optimal performance. We rather suggest a multiobjective process where a number of additional requirements, such as resilience to external attacks and the presence of spatial constraints, limit the ability of the system to achieve maximal transportation performance.     

Security concerns on machine learning solutions for 6G networks in mmWave beam prediction

6G – sixth generation – is the latest cellular technology currently under development for wireless communication systems. In recent years, machine learning (ML) algorithms have been applied widely in various fields, such as healthcare, transportation, energy, autonomous cars, and many more. Those algorithms have also been used in communication technologies to improve the system performance in terms of frequency spectrum usage, latency, and security. With the rapid developments of ML techniques, especially deep learning (DL), it is critical to consider the security concern when applying the algorithms. While ML algorithms offer significant advantages for 6G networks, security concerns on artificial intelligence (AI) models are typically ignored by the scientific community so far. However, security is also a vital part of AI algorithms because attackers can poison the AI model itself. This paper proposes a mitigation method for adversarial attacks against proposed 6G ML models for the millimeter-wave (mmWave) beam prediction using adversarial training. The main idea behind generating adversarial attacks against ML models is to produce faulty results by manipulating trained DL models for 6G applications for mmWave beam prediction. We also present a proposed adversarial learning mitigation method’s performance for 6G security in mmWave beam prediction application a fast gradient sign method attack. The results show that the defended model under attack’s mean square errors (i.e., the prediction accuracy) are very close to the undefended model without attack.     

Do topological models provide good information about electricity infrastructure vulnerability?

In order to identify the extent to which results from topological graph models are useful for modeling vulnerability in electricity infrastructure, we measure the susceptibility of power networks to random failures and directed attacks using three measures of vulnerability: characteristic path lengths, connectivity loss, and blackout sizes. The first two are purely topological metrics. The blackout size calculation results from a model of cascading failure in power networks. Testing the response of 40 areas within the Eastern U.S. power grid and a standard IEEE test case to a variety of attack/failure vectors indicates that directed attacks result in larger failures using all three vulnerability measures, but the attack-vectors that appear to cause the most damage depend on the measure chosen. While the topological metrics and the power grid model show some similar trends, the vulnerability metrics for individual simulations show only a mild correlation. We conclude that evaluating vulnerability in power networks using purely topological metrics can be misleading.     

一种全局同质化相依网络耦合模式

相依网络的相依模式(耦合模式)是影响其鲁棒性的重要因素之一.本文针对具有无标度特性的两个子网络提出一种全局同质化相依网络耦合模式.该模式以子网络的总度分布均匀化为原则建立相依网络的相依边,一方面压缩度分布宽度,提高其对随机失效的抗毁性,另一方面避开对度大节点(关键节点)的相依,提高其对蓄意攻击的抗毁性.论文将其与常见的节点一对一的同配、异配及随机相依模式以及一对多随机相依模式作了对比分析,仿真研究其在随机失效和蓄意攻击下的鲁棒性能.研究结果表明,本文所提全局同质化相依网络耦合模式能大大提高无标度子网络所构成的相依网络抗级联失效能力.本文研究成果能够为相依网络的安全设计等提供指导意义.     

Fast long-range connections in transportation networks

Multidimensional scaling is applied in order to visualize an analogue of the small-world effect implied by edges having different displacement velocities in transportation networks. Our findings are illustrated for two real-world systems, namely the London urban network (streets and underground) and the US highway network enhanced by some of the main US airlines routes. We also show that the travel time in these two networks is drastically changed by attacks targeting the edges with large displacement velocities.     

基于复杂网络理论的多元混合空管技术保障系统网络特征分析

为提高空管技术保障系统应对突发事件的能力,本文以空管技术保障系统导航、通信、监视设备覆盖的航路结构为基础,构建系统对应的空间网络模型.提出从灵活性、鲁棒性、高效性三个方面度量空管技术保障系统网络特性,对北京、上海、广州、昆明、沈阳、兰州飞行情报区的空管技术保障系统网络的平均度、度分布、度-度相关性、聚集系数、平均路径长度、直径等进行分析.分析结果显示,各飞行情报区空管技术保障系统的平均聚集系数在0.25—0.39之间,网络聚集程度偏低;网络平均路径长度为3.4,表现出小世界网络特征;度值3时服从幂律分布,度-度分布不表现出正相关或负相关.对网络进行基于度优先的和随机的抗毁性测度,空管技术保障系统网络抗毁性较差,网络的可靠性由少数核心节点决定,应对核心节点进行目标免疫,提高网络的抗毁性.这些规律为空管技术保障系统能力提升、新建扩建提供理论依据,对降低突发事件对空管系统保障能力的影响,保障空中交通持续安全具有现实意义.     

Mitigation strategies on scale-free networks against cascading failures

According to the dynamic characteristics of the cascading propagation, we introduce a mitigation mechanism and propose four mitigation methods on four types of nodes. By the normalized average avalanche size and a new measure, we demonstrate the efficiencies of the mitigation strategies on enhancing the robustness of scale-free networks against cascading failures and give the order of the effectiveness of the mitigation strategies. Surprisingly, we find that only adopting once mitigation mechanism on a small part of the overload nodes can dramatically improve the robustness of scale-free networks. In addition, we also show by numerical simulations that the optimal mitigation method strongly depends on the total capacities of all nodes in a network and the distribution of the load in the cascading model. Therefore, according to the protection strength for scale-free networks, by the distribution of the load and the protection price of networks, we can reasonably select how many nodes and which mitigation method to efficiently protect scale-free networks at the lower price. These findings may be very useful for avoiding various cascading-failure-induced disasters in the real world and for leading to insights into the mitigation of cascading failures.     

Proactive robustness control of heterogeneously loaded networks

A proactive measure to increase the robustness of heterogeneously loaded networks against cascades of overload failures is proposed. It is based on load-dependent weights. Compared to simple hop weights, respective shortest flow paths turn a previously heterogeneous load distribution into a more homogeneous one for the nodes and links of the network. The use of these flow paths increases the networks robustness and at the same time reduces the investment costs into the networks capacity layout. These findings are of relevance for critical infrastructures like communication and transportation networks.     

Evolution of the Internet AS-level topology:From nodes and edges to components

Studying the topology of infrastructure communication networks(e.g., the Internet) has become a means to understand and develop complex systems. Therefore, investigating the evolution of Internet network topology might elucidate disciplines governing the dynamic process of complex systems. It may also contribute to a more intelligent communication network framework based on its autonomous behavior. In this paper, the Internet Autonomous Systems(ASes) topology from 1998 to 2013 was studied by deconstructing and analysing topological entities on three different scales(i.e., nodes,edges and 3 network components: single-edge component M1, binary component M2 and triangle component M3). The results indicate that: a) 95% of the Internet edges are internal edges(as opposed to external and boundary edges); b) the Internet network consists mainly of internal components, particularly M2 internal components; c) in most cases, a node initially connects with multiple nodes to form an M2 component to take part in the network; d) the Internet network evolves to lower entropy. Furthermore, we find that, as a complex system, the evolution of the Internet exhibits a behavioral series,which is similar to the biological phenomena concerned with the study on metabolism and replication. To the best of our knowledge, this is the first study of the evolution of the Internet network through analysis of dynamic features of its nodes,edges and components, and therefore our study represents an innovative approach to the subject.     

Comparison of cascading failures in small-world and scale-free networks subject to vertex and edge attacks

Complex networks may undergo a global cascade of overload failures when a single highly loaded vertex or edge is intentionally attacked. Here we use the recent load model of cascading failures to investigate the performance of the small-world (SW) and scale-free (SF) networks subject to deliberate attacks on vertex and edge. Simulation results suggest that compared with the SW network, the SF network is more vulnerable to deliberate vertex attacks and more robust to deliberate edge attacks. In the SF network, deliberate vertex attacks can result in larger cascading failures than deliberate edge attacks; however, in the SW network the situation is opposite. Furthermore, with the increase of the rewiring probability the SW network becomes more and more robust to deliberate vertex and edge attacks.     

Enhancement of scale-free network attack tolerance

Despite the large size of most communication and transportation systems,there are short paths between nodes in these networks which guarantee the efficient information,data and passenger delivery;furthermore these networks have a surprising tolerance under random errors thanks to their inherent scale-free topology.However,their scale-free topology also makes them fragile under intentional attacks,leaving us a challenge on how to improve the network robustness against intentional attacks without losing their strong tolerance under random errors and high message and passenger delivering capacity.Here we propose two methods (SL method and SH method) to enhance scale-free network’s tolerance under attack in different conditions.     

Effects of the cascading failures on scale-free traffic networks

In this paper, we consider the artificial scale-free traffic network with dynamic weights (cost) and focus on how the removal strategies (flow-based removal, betweenness-based removal and mix-based removal) affect the damage of cascading failures based on the user-equilibrium (UE) assignment, which ensures the balance of flow on the traffic network. Experiment simulation shows that different removal strategies can bring large dissimilarities of the efficiency and damage after the intentional removal of an edge. We show that the mix-based removal of a single edge might reduce the damage of cascading failures and delay the breakdown time, especially for larger reserve capacity coefficient α. This is particularly important for real-world networks with a highly hetereogeneous distribution of flow, i.e., traffic and transportation networks, logistics networks and electrical power grids.     

Two-peak and three-peak optimal complex networks

A central issue in complex networks is tolerance of random failures and intentional attacks. Current literature emphasizes the dichotomy between networks with a power-law node connectivity distribution, which are robust to random failures but fragile to targeted attacks, versus networks with an exponentially decaying connectivity distribution, which are less tolerant to failures but more resilient to attacks. We prove analytically that the optimal network configuration under a classic measure of robustness is altogether different from both of the above: in all cases, failure and/or attack, there are no more than three distinct node connectivities in the optimal network.     

An Optimized Black-Box Adversarial Simulator Attack Based on Meta-Learning

Much research on adversarial attacks has proved that deep neural networks have certain security vulnerabilities. Among potential attacks, black-box adversarial attacks are considered the most realistic based on the the natural hidden nature of deep neural networks. Such attacks have become a critical academic emphasis in the current security field. However, current black-box attack methods still have shortcomings, resulting in incomplete utilization of query information. Our research, based on the newly proposed Simulator Attack, proves the correctness and usability of feature layer information in a simulator model obtained by meta-learning for the first time. Then, we propose an optimized Simulator Attack+ based on this discovery. Our optimization methods used in Simulator Attack+ include: (1) a feature attentional boosting module that uses the feature layer information of the simulator to enhance the attack and accelerate the generation of adversarial examples; (2) a linear self-adaptive simulator-predict interval mechanism that allows the simulator model to be fully fine-tuned in the early stage of the attack and dynamically adjusts the interval for querying the black-box model; and (3) an unsupervised clustering module to provide a warm-start for targeted attacks. Results from experiments on the CIFAR-10 and CIFAR-100 datasets clearly show that Simulator Attack+ can further reduce the number of consuming queries to improve query efficiency while maintaining the attack.     

On the propagation of congestion waves in the Internet

Traffic modeling of communication networks such as the Internet has become a very important field of research. A number of interesting phenomena are found in measurements and traffic simulations. One of them is the propagation of congestion waves opposite to the main packet flow direction. The purpose of this paper is to model and analyze packet congestion on a given route and to provide a possible explanation for this phenomenon.     

Lightweight authentication scheme for massive MIMO on Internet of Things connectivity

Massive MIMO is an essential technology in developing 5G networks and a concept that may be applied to other wireless systems. However, the advantages of adopting massive MIMO for broadband communication are well-established. Recently researcher has been devoted to building communication systems sustaining high communication rates with security. While massive MIMO for Internet-of-Things (IoT) connectivity is still a developing issue, IoT connectivity has requirements and limitations that differ significantly from broadband connections. Although IoT makes people’s lives easier by allowing physical devices to flow through, the interaction of open wireless channels such as Bluetooth, ZigBee, LoRa, Narrowband-Internet of Things (NBIoT), and WiFi, has produced various security and privacy difficulties. Identity authentication is one of the effective solutions for addressing the Internet of Things security and privacy concerns. The typical point-to-point authentication technique ignores the internet of things’ massive number of nodes and limited node resources. Group authentication is an authentication method that simultaneously confirms the identity of a group of members, offering a novel approach to identity identification for the internet of things nodes. However, existing group authentication systems appropriate for the internet of things scenarios pose security issues. They cannot withstand malicious attacks such as forging and replay and cannot prevent group managers from fooling group members. Most existing group authentication schemes are computationally expensive and cannot be applied to resource-constrained IoT scenarios. At the same time, existing systems based on secret-sharing technology cannot resist forgery attacks and replay attacks. The attacker can forge a legal token by modifying the Lagrangian coefficient in the authentication token to pass the group authentication. This work employs verifiable secret sharing technology to create a lightweight verifiable group authentication method (L-IoT-GS) suitable for Internet of Things situations to resist group managers’ deceptive group behavior. Nodes in the Internet of Things scenario can frequently join and leave the network. Because of this, this article proposes a critical update link based on a verified group authentication system for updating group Member rights. According to security analysis, the suggested L-IoT-GS scheme meets accuracy and confidentiality requirements and can withstand malicious attacks such as replay, forgery, and impersonation. Furthermore, performance study and experimental simulation reveal that the L-IoT-GS technique minimizes group members’ computing costs compared to existing standard IoT group authentication schemes.