arXiv subjects

Reuven Cohen

Publications and source records attributed to Reuven Cohen.

At least 55 records · Page 3Linked to original sources

arXiv2014

Efficiency of message transmission using biased random walks in complex networks in the presence of traps

We study the problem of a particle/message that travels as a biased random walk towards a target node in a network in the presence of traps. The bias is represented as the probability $p$ of the particle to travel along the shortest path to the target node. The efficiency of the transmission process is expressed through the fraction $f_g$ of particles that succeed to reach the target without being trapped. By relating $f_g$ with the number $S$ of nodes visited before reaching the target, we firstly show that, for the unbiased random walk, $f_g$ is inversely proportional to both the concentration $c$ of traps and the size $N$ of the network. For the case of biased walks, a simple approximation of $S$ provides an analytical solution that describes well the behavior of $f_g$, especially for $p>0.5$. Also, it is shown that for a given value of the bias $p$, when the concentration of traps is less than a threshold value equal to the inverse of the Mean First Passage Time (MFPT) between two randomly chosen nodes of the network, the efficiency of transmission is unaffected by the presence of traps and almost all the particles arrive at the target. As a consequence, for a given concentration of traps, we can estimate the minimum bias that is needed to have unaffected transmission, especially in the case of Random Regular (RR), Erdős-Rényi (ER) and Scale-Free (SF) networks, where an exact expression (RR and ER) or an upper bound (SF) of the MFPT is known analytically. We also study analytically and numerically, the fraction $f_g$ of particles that reach the target on SF networks, where a single trap is placed on the highest degree node. For the unbiased random walk, we find that $f_g \sim N^{-1/(γ-1)}$, where $γ$ is the power law exponent of the SF network.

physics.soc-ph↗

arXiv2014

Simultaneous first and second order percolation transitions in interdependent networks

In a system of interdependent networks, an initial failure of nodes invokes a cascade of iterative failures that may lead to a total collapse of the whole system in a form of an abrupt first order transition. When the fraction of initial failed nodes $1-p$ reaches criticality, $p=p_c$, the abrupt collapse occurs by spontaneous cascading failures. At this stage, the giant component decreases slowly in a plateau form and the number of iterations in the cascade, $τ$, diverges. The origin of this plateau and its increasing with the size of the system remained unclear. Here we find that simultaneously with the abrupt first order transition a spontaneous second order percolation occurs during the cascade of iterative failures. This sheds light on the origin of the plateau and on how its length scales with the size of the system. Understanding the critical nature of the dynamical process of cascading failures may be useful for designing strategies for preventing and mitigating catastrophic collapses.

cond-mat.stat-mech↗

arXiv2013

Coping with Physical Attacks on Random Network Structures

Communication networks are vulnerable to natural disasters, such as earthquakes or floods, as well as to physical attacks, such as an Electromagnetic Pulse (EMP) attack. Such real-world events happen at specific geographical locations and disrupt specific parts of the network. Therefore, the geographical layout of the network determines the impact of such events on the network's physical topology in terms of capacity, connectivity, and flow. Recent works focused on assessing the vulnerability of a deterministic network to such events. In this work, we focus on assessing the vulnerability of (geographical) random networks to such disasters. We consider stochastic graph models in which nodes and links are probabilistically distributed on a plane, and model the disaster event as a circular cut that destroys any node or link within or intersecting the circle. We develop algorithms for assessing the damage of both targeted and non-targeted (random) attacks and determining which attack locations have the expected most disruptive impact on the network. Then, we provide experimental results for assessing the impact of circular disasters to communications networks in the USA, where the network's geographical layout was modeled probabilistically, relying on demographic information only. Our results demonstrates the applicability of our algorithms to real-world scenarios. Our algorithms allows to examine how valuable is public information about the network's geographical area (e.g., demography, topography, economy) to an attacker's destruction assessment capabilities in the case the network's physical topology is hidden or examine the affect of hiding the actual physical location of the fibers on the attack strategy. Thereby, our schemes can be used as a tool for policy makers and engineers to design more robust networks and identifying locations which require additional protection efforts.

eess.SY↗

arXiv2013

Anomalous biased diffusion in networks

We study diffusion with a bias towards a target node in networks. This problem is relevant to efficient routing strategies in emerging communication networks like optical networks. Bias is represented by a probability $p$ of the packet/particle to travel at every hop towards a site which is along the shortest path to the target node. We investigate the scaling of the mean first passage time (MFPT) with the size of the network. We find by using theoretical analysis and computer simulations that for Random Regular (RR) and Erdős-Rényi (ER) networks, there exists a threshold probability, $p_{th}$, such that for $p p_{th}$ the MFPT scales logarithmically with $N$. The threshold value $p_{th}$ of the bias parameter for which the regime transition occurs is found to depend only on the mean degree of the nodes. An exact solution for every value of $p$ is given for the scaling of the MFPT in RR networks. The regime transition is also observed for the second moment of the probability distribution function, the standard deviation.

physics.soc-ph↗

arXiv2011

Percolation in Interdependent and Interconnected Networks: Abrupt Change from Second to First Order Transition

Robustness of two coupled networks system has been studied only for dependency coupling (S. Buldyrev et. al., Nature, 2010) and only for connectivity coupling (E. A. Leicht and R. M. D'Souza, arxiv:09070894). Here we study, using a percolation approach, a more realistic coupled networks system where both interdependent and interconnected links exist. We find a rich and unusual phase transition phenomena including hybrid transition of mixed first and second order i.e., discontinuities like a first order transition of the giant component followed by a continuous decrease to zero like a second order transition. Moreover, we find unusual discontinuous changes from second order to first order transition as a function of the dependency coupling between the two networks.

physics.soc-ph↗

arXiv2009

Dynamic networks and directed percolation

We introduce a model for dynamic networks, where the links or the strengths of the links change over time. We solve the model by mapping dynamic networks to the problem of directed percolation, where the direction corresponds to the evolution of the network in time. We show that the dynamic network undergoes a percolation phase transition at a critical concentration $p_c$, which decreases with the rate $r$ at which the network links are changed. The behavior near criticality is universal and independent of $r$. We find fundamental network laws are changed. (i) For Erdős-Rényi networks we find that the size of the giant component at criticality scales with the network size $N$ for all values of $r$, rather than as $N^{2/3}$. (ii) In the presence of a broad distribution of disorder, the optimal path length between two nodes in a dynamic network scales as $N^{1/2}$, compared to $N^{1/3}$ in a static network.

cond-mat.dis-nn↗

arXiv2008

Fractal Boundaries of Complex Networks

We introduce the concept of boundaries of a complex network as the set of nodes at distance larger than the mean distance from a given node in the network. We study the statistical properties of the boundaries nodes of complex networks. We find that for both Erdös-Rényi and scale-free model networks, as well as for several real networks, the boundaries have fractal properties. In particular, the number of boundaries nodes {\it B} follows a power-law probability density function which scales as $B^{-2}$. The clusters formed by the boundary nodes are fractals with a fractal dimension $d_{f} \approx 2$. We present analytical and numerical evidence supporting these results for a broad class of networks. Our findings imply potential applications for epidemic spreading.

math-ph↗

arXiv2008

Unusual percolation in simple small-world networks

We present an exact solution of percolation in a generalized class of Watts-Strogatz graphs defined on a 1-dimensional underlying lattice. We find a non-classical critical point in the limit of the number of long-range bonds in the system going to zero, with a discontinuity in the percolation probability and a divergence in the mean finite-cluster size. We show that the critical behavior falls into one of three regimes depending on the proportion of occupied long-range to unoccupied nearest-neighbor bonds, with each regime being characterized by different critical exponents. The three regimes can be united by a single scaling function around the critical point. These results can be used to identify the number of long-range links necessary to secure connectivity in a communication or transportation chain. As an example, we can resolve the communication problem in a game of "telephone".

cond-mat.dis-nn↗

arXiv2007

Throughput Optimal On-Line Algorithms for Advanced Resource Reservation in Ultra High-Speed Networks

Advanced channel reservation is emerging as an important feature of ultra high-speed networks requiring the transfer of large files. Applications include scientific data transfers and database backup. In this paper, we present two new, on-line algorithms for advanced reservation, called BatchAll and BatchLim, that are guaranteed to achieve optimal throughput performance, based on multi-commodity flow arguments. Both algorithms are shown to have polynomial-time complexity and provable bounds on the maximum delay for 1+epsilon bandwidth augmented networks. The BatchLim algorithm returns the completion time of a connection immediately as a request is placed, but at the expense of a slightly looser competitive ratio than that of BatchAll. We also present a simple approach that limits the number of parallel paths used by the algorithms while provably bounding the maximum reduction factor in the transmission throughput. We show that, although the number of different paths can be exponentially large, the actual number of paths needed to approximate the flow is quite small and proportional to the number of edges in the network. Simulations for a number of topologies show that, in practice, 3 to 5 parallel paths are sufficient to achieve close to optimal performance. The performance of the competitive algorithms are also compared to a greedy benchmark, both through analysis and simulation.

cs.NI↗

arXiv2007

Numerical evaluation of the upper critical dimension of percolation in scale-free networks

We propose a numerical method to evaluate the upper critical dimension $d_c$ of random percolation clusters in Erdős-Rényi networks and in scale-free networks with degree distribution ${\cal P}(k) \sim k^{-λ}$, where $k$ is the degree of a node and $λ$ is the broadness of the degree distribution. Our results report the theoretical prediction, $d_c = 2(λ- 1)/(λ- 3)$ for scale-free networks with $3 < λ< 4$ and $d_c = 6$ for Erdős-Rényi networks and scale-free networks with $λ> 4$. When the removal of nodes is not random but targeted on removing the highest degree nodes we obtain $d_c = 6$ for all $λ> 2$. Our method also yields a better numerical evaluation of the critical percolation threshold, $p_c$, for scale-free networks. Our results suggest that the finite size effects increases when $λ$ approaches 3 from above.

cond-mat.dis-nn↗

arXiv2007

Limited path percolation in complex networks

We study the stability of network communication after removal of $q=1-p$ links under the assumption that communication is effective only if the shortest path between nodes $i$ and $j$ after removal is shorter than $a\ell_{ij} (a\geq1)$ where $\ell_{ij}$ is the shortest path before removal. For a large class of networks, we find a new percolation transition at $\tilde{p}_c=(κ_o-1)^{(1-a)/a}$, where $κ_o\equiv < k^2>/< k>$ and $k$ is the node degree. Below $\tilde{p}_c$, only a fraction $N^δ$ of the network nodes can communicate, where $δ\equiv a(1-|\log p|/\log{(κ_o-1)}) < 1$, while above $\tilde{p}_c$, order $N$ nodes can communicate within the limited path length $a\ell_{ij}$. Our analytical results are supported by simulations on Erdős-Rényi and scale-free network models. We expect our results to influence the design of networks, routing algorithms, and immunization strategies, where short paths are most relevant.

cond-mat.stat-mech↗

arXiv2007

Graph Partitioning Induced Phase Transitions

We study the percolation properties of graph partitioning on random regular graphs with N vertices of degree $k$. Optimal graph partitioning is directly related to optimal attack and immunization of complex networks. We find that for any partitioning process (even if non-optimal) that partitions the graph into equal sized connected components (clusters), the system undergoes a percolation phase transition at $f=f_c=1-2/k$ where $f$ is the fraction of edges removed to partition the graph. For optimal partitioning, at the percolation threshold, we find $S \sim N^{0.4}$ where $S$ is the size of the clusters and $\ell\sim N^{0.25}$ where $\ell$ is their diameter. Additionally, we find that $S$ undergoes multiple non-percolation transitions for $f<f_c$.

cond-mat.stat-mech↗

arXiv2006

Bounding the Bias of Tree-Like Sampling in IP Topologies

It is widely believed that the Internet's AS-graph degree distribution obeys a power-law form. Most of the evidence showing the power-law distribution is based on BGP data. However, it was recently argued that since BGP collects data in a tree-like fashion, it only produces a sample of the degree distribution, and this sample may be biased. This argument was backed by simulation data and mathematical analysis, which demonstrated that under certain conditions a tree sampling procedure can produce an artificail power-law in the degree distribution. Thus, although the observed degree distribution of the AS-graph follows a power-law, this phenomenon may be an artifact of the sampling process. In this work we provide some evidence to the contrary. We show, by analysis and simulation, that when the underlying graph degree distribution obeys a power-law with an exponent larger than 2, a tree-like sampling process produces a negligible bias in the sampled degree distribution. Furthermore, recent data collected from the DIMES project, which is not based on BGP sampling, indicates that the underlying AS-graph indeed obeys a power-law degree distribution with an exponent larger than 2. By combining this empirical data with our analysis, we conclude that the bias in the degree distribution calculated from BGP data is negligible.

cs.NI↗

arXiv2006

Percolation theory applied to measures of fragmentation in social networks

We apply percolation theory to a recently proposed measure of fragmentation $F$ for social networks. The measure $F$ is defined as the ratio between the number of pairs of nodes that are not connected in the fragmented network after removing a fraction $q$ of nodes and the total number of pairs in the original fully connected network. We compare $F$ with the traditional measure used in percolation theory, $P_{\infty}$, the fraction of nodes in the largest cluster relative to the total number of nodes. Using both analytical and numerical methods from percolation, we study Erdős-Rényi (ER) and scale-free (SF) networks under various types of node removal strategies. The removal strategies are: random removal, high degree removal and high betweenness centrality removal. We find that for a network obtained after removal (all strategies) of a fraction $q$ of nodes above percolation threshold, $P_{\infty}\approx (1-F)^{1/2}$. For fixed $P_{\infty}$ and close to percolation threshold ($q=q_c$), we show that $1-F$ better reflects the actual fragmentation. Close to $q_c$, for a given $P_{\infty}$, $1-F$ has a broad distribution and it is thus possible to improve the fragmentation of the network. We also study and compare the fragmentation measure $F$ and the percolation measure $P_{\infty}$ for a real social network of workplaces linked by the households of the employees and find similar results.

cond-mat.dis-nn↗

arXiv2006

Optimal Paths in Complex Networks with Correlated Weights: The World-wide Airport Network

We study complex networks with weights, $w_{ij}$, associated with each link connecting node $i$ and $j$. The weights are chosen to be correlated with the network topology in the form found in two real world examples, (a) the world-wide airport network, and (b) the {\it E. Coli} metabolic network. Here $w_{ij} \sim x_{ij} (k_i k_j)^α$, where $k_i$ and $k_j$ are the degrees of nodes $i$ and $j$, $x_{ij}$ is a random number and $α$ represents the strength of the correlations. The case $α> 0$ represents correlation between weights and degree, while $α< 0$ represents anti-correlation and the case $α= 0$ reduces to the case of no correlations. We study the scaling of the lengths of the optimal paths, $\ell_{\rm opt}$, with the system size $N$ in strong disorder for scale-free networks for different $α$. We calculate the robustness of correlated scale-free networks with different $α$, and find the networks with $α< 0$ to be the most robust networks when compared to the other values of $α$. We propose an analytical method to study percolation phenomena on networks with this kind of correlation. We compare our simulation results with the real world-wide airport network, and we find good agreement.

physics.soc-ph↗

arXiv2006

Communication Bottlenecks in Scale-Free Networks

We consider the effects of network topology on the optimality of packet routing quantified by $γ_c$, the rate of packet insertion beyond which congestion and queue growth occurs. The key result of this paper is to show that for any network, there exists an absolute upper bound, expressed in terms of vertex separators, for the scaling of $γ_c$ with network size $N$, irrespective of the routing algorithm used. We then derive an estimate to this upper bound for scale-free networks, and introduce a novel static routing protocol which is superior to shortest path routing under intense packet insertion rates.

cs.NI↗

arXiv2006

Search in Complex Networks : a New Method of Naming

We suggest a method for routing when the source does not posses full information about the shortest path to the destination. The method is particularly useful for scale-free networks, and exploits its unique characteristics. By assigning new (short) names to nodes (aka labelling) we are able to reduce significantly the memory requirement at the routers, yet we succeed in routing with high probability through paths very close in distance to the shortest ones.

cs.NI↗

arXiv2005

Width of percolation transition in complex networks

It is known that the critical probability for the percolation transition is not a sharp threshold, actually it is a region of non-zero width $Δp_c$ for systems of finite size. Here we present evidence that for complex networks $Δp_c \sim \frac{p_c}{l}$, where $l \sim N^{ν_{opt}}$ is the average length of the percolation cluster, and $N$ is the number of nodes in the network. For Erdős-Rényi (ER) graphs $ν_{opt} = 1/3$, while for scale-free (SF) networks with a degree distribution $P(k) \sim k^{-λ}$ and $3<λ<4$, $ν_{opt} = (λ-3)/(λ-1)$. We show analytically and numerically that the \textit{survivability} $S(p,l)$, which is the probability of a cluster to survive $l$ chemical shells at probability $p$, behaves near criticality as $S(p,l) = S(p_c,l) \cdot exp[(p-p_c)l/p_c]$. Thus for probabilities inside the region $|p-p_c| < p_c/l$ the behavior of the system is indistinguishable from that of the critical point.

cond-mat.dis-nn↗