Random Topologies Research Articles

Efficient Schemes for Optimizing Load Balancing and Communication Cost in Edge Computing Networks

Edge computing architectures promise increased quality of service with low communication delays by bringing cloud services closer to the end-users, at the distributed edge servers of the network edge. Hosting server capabilities at access nodes, thereby yielding edge service nodes, offers service proximity to users and provides QoS guarantees. However, the placement of edge servers should match the level of demand for computing resources and the location of user load. Thus, it is necessary to devise schemes that select the most appropriate access nodes to host computing services and associate every remaining access node with the most proper service node to ensure optimal service delivery. In this paper, we formulate this problem as an optimization problem with a bi-objective function that aims at both communication cost minimization and load balance optimization. We propose schemes that tackle this problem and compare their performance against previously proposed heuristics that have been also adapted to target both optimization goals. We study how these algorithms behave in lattice and random grid network topologies with uniform and non-uniform workloads. The results validate the efficiency of our proposed schemes in addition to the significantly lower execution times compared to the other heuristics.

The asymptotic convergence based on marked locators for time-varying nonlinear multi-agent systems with the protected random switching topologies against stochastic communication disruptions

The asymptotic convergence based on marked locators for time-varying nonlinear multi-agent systems with the protected random switching topologies against stochastic communication disruptions

Performance of qpAdm -based screens for genetic admixture on admixture-graph-shaped histories and stepping-stone landscapes.

qpAdm is a statistical tool that is often used in exploratory archaeogenetic studies for finding optimal admixture models of population history. Despite its popularity, qpAdm remains untested on histories in the form of admixture graphs of random topology or stepping-stone landscapes. We analyzed data from such simulations and found that while for admixture-graph-shaped histories there exist simple solutions (temporal stratification) for minimizing false findings of gene flow, in the case of stepping-stone landscapes the method generates results that do not appear suspect but are misleading: feasible qpAdm models are either accurate but simplistic in the context of landscapes, or highly inaccurate in the case of multi-component models. This is largely is due to two reasons: 1) because of complex migration networks that violate the assumptions of the method, there is poor correlation between qpAdm p -values and model optimality in many sections of the parameter space; 2) admixture fraction estimates between 0 and 1 are largely restricted to symmetric source configurations around targets, hence popular [0, 1] model plausibility criteria confound analyses of landscape-type demographies, unless their interpretations are explicitly spatial. For many species/regions/periods archaeogenetic sampling is very sparse and may be random with respect to population density of ancient individuals. In this situation only a specific combination of landscape properties and feasibility criteria allows to efficiently reject highly asymmetric non-optimal models most abundant in random deme sets. This problem may obscure useful signal (rare optimal models) and might be responsible for some claims about rapid long-distance migrations in the literature.

Exploring the limits of the law of mass action in the mean field description of epidemics on Erdös-Rényi networks

The manner epidemics occurs in a social network depends on various elements, with two of the most influential being the relationships among individuals in the population and the mechanism of transmission. In this paper, we assume that the social network has a homogeneous random topology of Erdös-Rényi type. Regarding the contagion process, we assume that the probability of infection is proportional to the proportion of infected neighbours.We consider a constant population, whose individuals are the nodes of the social network, formed by two variable subpopulations: Susceptible and Infected (SI model). We simulate the epidemics on this random network and study whether the average dynamics can be described using a mean field approach in terms of Differential Equations, employing the law of mass action. We show that a macroscopic description could be applied for low average connectivity, adjusting the value of the contagion rate in a precise function. This dependence is illustrated by calculating the transient times for each connectivity.This study contributes valuable insights into the interplay between network connectivity, contagion dynamics, and the applicability of mean-field approximations. The delineation of critical thresholds and the distinctive behaviour at lower connectivity enable a deeper understanding of epidemic dynamics.

Advances in random topology

Advances in random topology

Tensile deformation and failure of AlSi10Mg random cellular metamaterials

Cellular metamaterials are an emerging class of porous materials, in which the topology of the solid is precisely-engineered to achieve attractive combinations of properties. This work specifically examines the tensile response of a novel type of cellular metamaterials, in which pores of arbitrary elliptical shape are randomly-dispersed into an aluminum alloy matrix. Their porous mesostructure is generated numerically via a random sequential absorption algorithm, and is fabricated by laser powder bed fusion from AlSi10Mg powders. The results – obtained by means of digital image correlation combined with X-ray tomography – highlight the advantages offered by a random cellular topology. They are notably a low sensitivity to geometric imperfections (which inevitably result from manufacturing), coupled with the ability to delay long-wavelength strain localization, which is in turn responsible for failure. Structural disorder also leads to highly heterogeneous deformation patterns, which result from the interaction between geometric pores and promote void growth and coalescence during plastic straining. The presence of manufacturing defects exacerbates void interaction and promotes early plastic flow localization. Interestingly, experiments reveal that this class of porous solids effectively displays features of the ductile fracture of metals at the mesoscale. For example, void-sheeting is observed with increasing pore aspect ratio and is accompanied by large geometric distortions of the voids upon coalescence. Measured data for the pore strains are highly scattered and show a departure from McClintock’s model predictions for the voids within the fracture band. Collectively, this study highlights the potential offered by random porous metamaterials, which can be harnessed to revisit the complex mechanisms of ductile fracture at the mesoscale.

Random Contrastive Interaction for Particle Swarm Optimization in High-Dimensional Environment

In high dimensional environment, the interaction among particles significantly affects their movements in searching the vast solution space and thus plays a vital role in assisting particle swarm optimization (PSO) to attain good performance. To this end, this paper designs a random contrastive interaction (RCI) strategy for PSO, resulting in RCI-PSO, to tackle large-scale optimization problems (LSOPs) effectively and efficiently. Unlike existing interaction mechanisms for low-dimensional problems, RCI randomly chooses several different peers from the current swarm to construct a random interaction topology for each particle. Then, it lets the particle interact with the selected peers based on their current evolutionary information instead of their historical evolutionary information. Within the topology, RCI only propagates the evolutionary information of two contrastive dominators with the largest difference in fitness to direct the evolution of the particle. Therefore, particles with no more than two dominators in their topologies are not updated. Furthermore, a dynamic topology size adjustment scheme is devised to gradually enlarge the interaction topology. In this way, the swarm gradually switches from exploring the immense search space dispersedly to exploiting the found optimal regions intensively as the evolution continues. With these two strategies, RCI-PSO expectedly compromises search diversity and search convergence well at the swarm level and the particle level. At last, extensive experiments executed on two public LSOP suites verify that RCI-PSO performs competitively with or even much better than totally 40 state-of-theart large-scale approaches and preserves a good capability and scalability in tackling complex LSOPs.

EVALUATING ROUTING ALGORITHMS ACROSS DIFFERENT WIRELESS MESH NETWORK TOPOLOGIES USING NS-3 SIMULATOR

Wireless Mesh Networks are gaining traction as a solution for delivering reliable connectivity without centralized infrastructure. They operate through wireless node interconnections, forming self-configuring networks ideal for scenarios where wired networks are impractical. Routing is crucial in Wireless Mesh Networks to ensure efficient communication among nodes. However, the suitability of routing algorithms for Wireless Mesh Network’s topology requires further investigation. This paper proposes an investigation into the effectiveness of routing algorithms like AODV, DSDV, and OLSR across various Wireless Mesh Networks topologies using NS-3 simulation. It also aims to determine the optimal number of nodes and protocols to maximize throughput and minimize packet loss within a limited area. Through rigorous NS-3 simulations, the study demonstrates that AODV, DSDV, and OLSR exhibit differing effectiveness across random, mesh grid, and Fruchterman-Reingold topologies. These results emphasize the importance of considering topology-specific factors when selecting and optimizing routing protocols for Wireless Mesh Networks. In summary, Wireless Mesh Networks offer decentralized connectivity, but the effectiveness of routing algorithms in different topologies remains understudied. This investigation addresses this gap by evaluating routing algorithms across various topologies, shedding light on their suitability and performance in Wireless Mesh Networks.

A Comparative Analysis of Flooding Methods on Random and Real Network Topologies

The Internet of Things (IoT) is revolutionizing industries by connecting everyday objects, known as smart devices, via the Internet. These devices, embedded with sensors and communication technologies, gather and share data. For the guaranteed gathering of information, the devices share global knowledge with each other, by using dissemination mechanisms in order to broadcast information. This study evaluates four flooding methods for broadcasting information across network nodes, namely: (i) blind flooding; (ii) probabilistic flooding; (iii) m-probabilistic flooding; and (iv) scoped probabilistic flooding, the latter to be introduced here. The evaluation considers random networks that are based on the Burr Type XII distribution and seven real networks. The evaluated flooding methods are studied on three different metrics: (i) coverage achieved; (ii) number of messages exchanged; and (iii) a metric that is based on binomial approximation. The latter is introduced to provide deeper insights into the particulars of the under-evaluation flooding methods. The results show that, under certain conditions, m-probabilistic flooding outperforms probabilistic flooding in terms of coverage, while requiring significantly fewer messages. Additionally, the study revealed that the scoped probabilistic flooding achieves coverage comparable to that of the probabilistic flooding while reducing the number of exchanged messages.

Multi-objective topology optimization method for multi-axis random vibration based on hybrid cellular automata

Considering the lack of effective solutions for multi-axis random vibration topology optimization problems, and recognizing that multi-axis random vibration excitation is the most common loading conditions experienced by structures during their operational life, this paper proposes a multi-objective topology optimization method for multi-axis random vibration. By combining Hybrid Cellular Automata (HCA) and Analytic Hierarchy Process (AHP), this method aims to enhance the vibration resistance of structures and extend their fatigue life during the topology optimization phase. To validate the effectiveness and engineering practicality of this method, two engineering cases were designed: a cantilever beam case and an automotive steering knuckle case. Multi-objective topology optimization were conducted with objectives including mass, stiffness, and vibration intensity metric (root-mean-square stress). The two cases demonstrate that the control group's topology model exhibits weaker resistance to random vibrations and shorter fatigue life. In contrast, the validation group's topology model introduces an additional load path near the excitation point, effectively dissipating the impact of vibration excitation. This enhances the structure's vibration resistance and significantly extends its fatigue life.

Distributed l2−l∞ filtering over wireless sensors based on adaptive event triggering

This paper investigates the design problem of a set of distributed filters. These adaptive event-triggered (AET) filters have the [Formula: see text] property and are suitable for sensor networks characterized by random measurement data loss and dynamic topology switching. In a distributed filtering network, each local filter estimates the system’s state not only relying on the node’s individual information but also using information from neighboring nodes in the network topology. The adaptive event triggering condition is determined by the combination of the latest release data of the filter itself, the estimated value of the current moment, and the latest release data of the neighboring nodes. The event triggering mechanism adopts a threshold adaptive adjustment scheme, which dynamically changes the threshold parameter according to the filtering error under the premise of guaranteeing the filter performance, thus maximally saving the network communication resources. Initially, we used a binary sequence to describe the random loss of sensor measurements, and a Markov chain is used to describe the switching of the filtering network topology. Secondly, a Lyapunov function is created to examine the [Formula: see text] performance index of the filtering error system and investigate its mean square exponential stability. Subsequently, the distributed filter is formulated, and its parameters are determined using the linear matrix inequality approach. Ultimately, the validity of the proposed method is substantiated through numerical illustrations.

Minimum spanning tree analysis of EEG resting-state functional networks in schizophrenia

Schizophrenia is a serious and complex mental disease, known to be associated with various subtle structural and functional deviations in the brain. Recently, increased attention is given to the analysis of brain-wide, global mechanisms, strongly altering the communication of long-distance brain areas in schizophrenia. Data of 32 patients with schizophrenia and 28 matched healthy control subjects were analyzed. Two minutes long 64-channel EEG recordings were registered during resting, eyes closed condition. Average connectivity strength was estimated with Weighted Phase Lag Index (wPLI) in lower frequencies: delta and theta, and Amplitude Envelope Correlation with leakage correction (AEC-c) in higher frequencies: alpha, beta, lower gamma and higher gamma. To analyze functional network topology Minimum Spanning Tree (MST) algorithms were applied. Results show that patients have weaker functional connectivity in delta and alpha frequency bands. Concerning network differences, the result of lower diameter, higher leaf number, and also higher maximum degree and maximum betweenness centrality in patients suggest a star-like, and more random network topology in patients with schizophrenia. Our findings are in accordance with some previous findings based on resting-state EEG (and fMRI) data, suggesting that MST network structure in schizophrenia is biased towards a less optimal, more centralized organization.

Design of a load balancing Objective Function for RPL

Routing protocols for Internet of Things (IoT) play a major role in the performance of the network. The standard Routing Protocol for Low-Power and Lossy Networks (RPL) suffers from a number of limitations including congestion of higher-level nodes and unbalanced topology. This paper proposes a novel Objective Function called Load Balanced Minimum Rank with Hysteresis Objective Function (LB_MRHOF), which assigns child nodes to the most suitable parent in the topology. The Objective Function utilizes a weight of the Expected Transmission Count (ETX) and number of children to calculate the Composite ETX and Number of Children (CENOC) which estimates the load on each node. The attained CENOC is used to select the optimum parent for each node in the topology, where nodes with high CENOC are avoided in the parent selection process. The proposed Objective Function has been evaluated under random and hierarchical network topologies. In addition, the evaluation has investigated the influence of the number of nodes by testing for small, medium and large-scale networks. Results have shown that the proposed Objective Function outperforms MRHOF, OF_FUZZY and OF-EC in terms of Packet Delivery Ratio (PDR) and reduces nodal hop-count under all tested scenarios, with no compromise in energy consumption. They have also revealed that the best performance achieved by LB_MRHOF is attained under large-scale networks. The resulting network topology which is formed by the proposed Objective Function has shown improved balance and more depth.

Brain topology improved spiking neural network for efficient reinforcement learning of continuous control.

The brain topology highly reflects the complex cognitive functions of the biological brain after million-years of evolution. Learning from these biological topologies is a smarter and easier way to achieve brain-like intelligence with features of efficiency, robustness, and flexibility. Here we proposed a brain topology-improved spiking neural network (BT-SNN) for efficient reinforcement learning. First, hundreds of biological topologies are generated and selected as subsets of the Allen mouse brain topology with the help of the Tanimoto hierarchical clustering algorithm, which has been widely used in analyzing key features of the brain connectome. Second, a few biological constraints are used to filter out three key topology candidates, including but not limited to the proportion of node functions (e.g., sensation, memory, and motor types) and network sparsity. Third, the network topology is integrated with the hybrid numerical solver-improved leaky-integrated and fire neurons. Fourth, the algorithm is then tuned with an evolutionary algorithm named adaptive random search instead of backpropagation to guide synaptic modifications without affecting raw key features of the topology. Fifth, under the test of four animal-survival-like RL tasks (i.e., dynamic controlling in Mujoco), the BT-SNN can achieve higher scores than not only counterpart SNN using random topology but also some classical ANNs (i.e., long-short-term memory and multi-layer perception). This result indicates that the research effort of incorporating biological topology and evolutionary learning rules has much in store for the future.

Investigating the Impact of Technology Scaling on Stability Performance of Sub-threshold Static Random Access Memory Bitcell Topologies

Static random-access memory (SRAM) is increasingly being used in VLSI circuits as a result of the development of portable devices. As the demand for low-power devices increases, sub-threshold operation of SRAMs has become a popular method to reduce power consumption but at the cost of reduced stability. The work aims to propose design parameters for SRAM bit cell topologies- 6T and 10T with proper device sizing and to investigate the impact of technology scaling on its stability performance in the sub threshold region. Moreover, analysis demonstrates the impact of variations of device sizing in terms of Cell ratio and Pull up ratio on stability margins of designed SRAM topologies. Simulations are carried out in the HSPICE environment using 16 nm and 22 nm CMOS process nodes. We find that the stability margins of 10T SRAM cell outperform the 6T SRAM cell; however stability margin decreases with decreasing technology nodes. The Read Static Noise Margin (RSNM) and Write Static Noise Margin (WSNM) of 10T SRAM cell is improved by 48.38% and 308.64% respectively in 22nm and 53.04% and 243.93% respectively in 16nm process technology as compare to 6T SRAM cell.

Vulnerability Analysis of Power Transmission Grids Subject to Cascading Failures

Cascading failures are a major threat to interconnected systems, such as electrical power transmission networks. Typically, approaches proposed for devising optimized control strategies are demonstrated with reference to a few test systems of reference (IEEE systems). However, this limits the robustness of the proposed strategies with respect to different power grid structures. Recently, this issue has been addressed by considering synthetic networks randomly generated for mimicking power transmission grids’ characteristics. These networks can be used for investigating the vulnerability of power networks to cascading failures. In this work, we propose to apply a recent algorithm for sampling random power grid topologies with realistic electrical parameters and further extend it to the random allocation of generation and load. Integration with a realistic cascade simulation tool, then, allows us to perform thorough statistical analyses of power grids with respect to their cascading failure behavior, thus offering a powerful tool for identifying the strengths and weaknesses of different grid classes. New metrics for ranking the control and mitigation effort requirements of individual cascade scenarios and/or of grid configurations are defined and computed. Finally, genetic algorithms are used to identify strategies to improve the robustness of existing power networks.

M-Voronoi and other random open and closed-cell elasto-plastic cellular materials: Geometry generation and numerical study at small and large strains

The present study deals with a numerical design strategy of a novel class of three-dimensional random Voronoi-type geometries, called M-Voronoi. These materials comprise random, non-quadratic convex void shapes and non-uniform intervoid ligament thicknesses, and can span high-to-low relative densities. The starting point for their generation is a random adsorption algorithm (RSA) construction with spherical voids embedded in an incompressible, nonlinear elastic matrix phase. The initial RSA geometry is subjected to large elastic volume changes by prescribing Dirichlet boundary conditions. Due to the incompressibility of the matrix phase, the externally imposed volume changes lead to significant void growth. The numerical growth process may be stopped at any desired porosity. The proposed M-Voronoi process is general and allows the formation of isotropic (or anisotropic) designs. As a byproduct of the developed approach, we also present a novel remeshing technique allowing to read arbitrary geometries of one or multiple phases. The elasto-plastic properties of the M-Voronoi porous materials are numerically investigated at small strains as well as large compressive and shear loads. Their response is assessed by comparison with other well-known random and periodic porous geometries such as polydisperse porous materials with spherical voids (RSA), classical TPMS Gyroid geometries and random Spinodoid topologies. The results show that M-Voronoi and RSA (with spherical voids) geometries exhibit the stiffest elastic and highest flow stress response compared to the other two geometries. This study shows unambiguously that randomness may or may not lead to enhanced mechanical response such as higher stiffness or flow stress.

Identifying the perceived local properties of networks reconstructed from biased random walks.

Many real-world systems give rise to a time series of symbols. The elements in a sequence can be generated by agents walking over a networked space so that whenever a node is visited the corresponding symbol is generated. In many situations the underlying network is hidden, and one aims to recover its original structure and/or properties. For example, when analyzing texts, the underlying network structure generating a particular sequence of words is not available. In this paper, we analyze whether one can recover the underlying local properties of networks generating sequences of symbols for different combinations of random walks and network topologies. We found that the reconstruction performance is influenced by the bias of the agent dynamics. When the walker is biased toward high-degree neighbors, the best performance was obtained for most of the network models and properties. Surprisingly, this same effect is not observed for the clustering coefficient and eccentric, even when large sequences are considered. We also found that the true self-avoiding displayed similar performance as the one preferring highly-connected nodes, with the advantage of yielding competitive performance to recover the clustering coefficient. Our results may have implications for the construction and interpretation of networks generated from sequences.

Minimal-Overlap Centrality for Multi-Gateway Designation in Real-Time TSCH Networks

This article presents a novel centrality-driven gateway designation framework for the improved real-time performance of low-power wireless sensor networks (WSNs) at system design time. We target time-synchronized channel hopping (TSCH) WSNs with centralized network management and multiple gateways with the objective of enhancing traffic schedulability by design . To this aim, we propose a novel network centrality metric termed minimal-overlap centrality that characterizes the overall number of path overlaps between all the active flows in the network when a given node is selected as gateway. The metric is used as a gateway designation criterion to elect as a gateway the node leading to the minimal number of overlaps. The method is then extended to multiple gateways with the aid of the unsupervised learning method of spectral clustering . Concretely, after a given number of clusters are identified, we use the new metric at each cluster to designate as cluster gateway the node with the least overall number of overlaps. Extensive simulations with random topologies under centralized earliest-deadline-first (EDF) scheduling and shortest-path routing suggest our approach is dominant over traditional centrality metrics from social network analysis, namely, eigenvector , closeness , betweenness , and degree . Notably, our approach reduces by up to 40% the worst-case end-to-end deadline misses achieved by classical centrality-driven gateway designation methods.

An improved localization approach based on Sybil attack for WSN

Recently, the precise location of sensor nodes has emerged as a significant challenge in the realm of Internet of Things (IoT) applications, including wireless sensor networks (WSNs). The accurate determination of geographical coordinates for detected events holds pivotal importance in these applications. Despite the DV-Hop gaining popularity due to its cost-effectiveness, feasibility, and lack of additional hardware requirements, it remains hindered by a relatively notable localization error. To overcome this limitation, our study introduces three new approaches of localization that combine the DV-Hop with the chicken swarm optimization approach. The primary objective is to improve the precision of DV-hop-based approaches. Through extensive simulations, we compare the efficiency of our novel localization algorithms with competitive approaches considering random network topologies. The results derived from the simulation validate the efficiency of our proposed algorithms. The proposed algorithm, the HW-DV-HopCSO, achieves a considerable improvement regarding positioning accuracy.