This thesis presents an approach to the design and management of core networks where the packet transport is the main service and the backbone should be able to respond to unforeseen changes in network parameters in order to provide smooth and reliable service for the customers. Inspired by Darwin's seminal work describing the long-term processes in life, and with the help of graph theoretic metrics, in particular the "random-walk betweenness", we assign a survival value, the network criticality, to a communication network to quantify its robustness. We show that the random-walk betweenness of a node (link) consists of the product of two terms, a global measure which is fixed for all the nodes (links) and a local graph measure which is in fact the weight of the node (link). The network criticality is defined as the global part of the betweenness of a node (link). We show that the network criticality is a monotone decreasing, and strictly convex function of the weight matrix of the network graph. We argue that any communication network can be modeled as a topology that evolves based on survivability and performance requirements. The evolution should be in the direction of decreasing the network criticality, which in turn increases the network robustness. We use network criticality as the main control parameter and we propose a network management system, AutoNet, to guide the network evolution in real time. AutoNet consists of two autonomic loops, the slow loop to control the long-term evolution of robustness throughout the whole network, and the fast loop to account for short-term performance and robustness issues. We investigate the dynamics of network criticality and we develop a convex optimization problem to minimize the network criticality. We propose a network design procedure based on the optimization problem which can be used to develop the long-term autonomic loop for AutoNet. Furthermore, we use the properties of the duality gap of the optimization problem to develop traffic engineering methods to manage the transport of packets in a network. This provides for the short-term autonomic loop of AutoNet architecture. Network criticality can also be used to rank alternative networks based on their robustness to the unpredicted changes in network conditions. This can help find the best network structure under some pre-specified constraint to deal with robustness issues.

The network representation of the interactions between proteins and genes allows for a holistic perspective of the complex machinery underlying the living cell. However, the large number of interacting entities within the cell makes network construction a daunting and arduous task, prone to errors and missing information. Fortunately, the structure of biological networks is not different from that of other complex systems, such as social networks, the world-wide web or power grids, for which growth models have been proposed to better understand their structure and function. This means that we can design tools based on these models in order to exploit the topology of biological interactomes with the aim to construct more complete and reliable maps of the cell. In this work, we propose three novel and powerful approaches for the prediction of interactions in biological networks and conclude that it is possible to mine the topology of these complex system representations and produce reliable and biologically meaningful information that enriches the datasets to which we have access today.

Recently, network theory has been used to analyze the formal syntactic and semantic properties of written texts to explain the development of language (Solé et al. 2005). While foundational, this approach neglects the social and cultural pressures affecting language in interaction, a central focus of sociolinguistics and linguistic anthropology (Hymes 1974, Goffman 1981, Gumperz 1982, Goodwin 2006). The influential work of M.M. Bakhtin (1981) frames speech as an emergent social process inflected by shifting patterns of negotiated meanings. As Hill (1986) observed "the enormous impact of Bakhtin's work, already felt with earthquake strength in literary studies...[is] now beginning to appear with equal force in the anthropology of language" (1986: 89).The aim of this research is to test the conjecture that by expanding the frame of language network analysis to include the social context of speech, the emergent properties of heteroglossia predicted by Bakhtin will be clarified. This analysis builds on prior research on language in interaction, drawing from sociolinguistic analysis (Sacks et al. 1974, Atkinson & Heritage 1984), word frequency (Nelson et al. 1998, Mendoza-Denton 2003), and network analysis (Bearman & Stovel 2000, de Nooy et al. 2005, Solé et al. 2005, Mehler 2010).According to Bakhtin, heteroglossia emerges as speakers "appropriate the words of others and populate them with one's own intention" (1981:428). This multi-sited doctoral research investigates the speech of butchers through participant observation, work place interactions and interviews, with a focus on references to blood. Some of the semantic features that become affixed to blood are due to historical and popular culture understandings of this signifier, while other salient features derive from subject positionality and community of practice (Lave & Wenger 1991). This work provides a snapshot of all of these processes at work in the speech of an occupational community of American butchers. The results of this analysis show that including the social context has significant effects on the conceptualization of both semantic and social networks, in comparison with networks derived exclusively from written texts.

This research presents a structural model of the effect of the organization of military units upon their capability. This research is oriented towards a more complete understanding of military capability and policy decisions about the structure and development of military forces. We identify the types of national and military policy decisions that claims of military capability inform, and find that there are five distinct types of capability claims relevant to military policy. We show how these types of capability claims are logically related to each other, but have different premises, predicates, and standards of proof. We find that one of these types of claims, General Organization Capability Claims, ties together the various military policy decisions. The remainder of this research shows how these capability claims can be formally structured based on military doctrine and structurally evaluated using a network-science based model. The interaction between the structural elements of a military organization (personnel, materiel, and information) and the things it is supposed to do (military tasks) can be represented and analyzed with network science methods, and represents a type of general organization capability claim. We present a method for representing policy decisions about unit structure and tactical doctrine. We then develop two versions of a structural model of capability–one that links the individual elements of an organization to the tasks it performs; another that considers the capacity of a set of organizations to meet a set of requirements. We show that network statistics of organizations represented off of authoritative, rather than observational, data are still consistent with network science findings but require interpretation. We also show how alternate methods of aggregating organizations can expand the utility of the capability measurement. This research presents five new contributions to the fields of military policy analysis and network science–(1) a taxonomy of military capability claims, (2) a meta-network model of doctrinal organization and task data, (3) a structural model of organization capability, (4) a structural model of organization capacity, and (5) a network-based method integer programming method.

Since 2008, more than half of the world population lives in cities. To cope with this rapid urbanization in a sustainable manner, transit systems all around the world are likely to grow. By studying 33 networks in the world, this thesis identifies the properties and effects of metro network designs by using a graph theory approach. After the literature review, a new methodology was introduced to translate networks into graphs; it notably accounts for various transit specificities (e.g., presence of lines). Metro networks were then characterised according to their State, Form, and Structure; where State relates to the development phase of metros; Form investigates the link between metros and the built environment; Structure examines the intrinsic properties of metros, by notably looking at their connectivity. Subsequently, the complexity and robustness of metros were studied; metros were found to possess scale-free and small-world features although showing atypical topologies; robustness emphasizes on the presence of alternative paths. Three network design indicators (coverage, directness and connectivity) were then related to ridership (annual boardings per capita), and positive relations were observed, which suggests that network design plays an important role in their success. Finally, these concepts were applied to the Toronto metro plans announced by the Toronto regional transportation authority, Metrolinx; it was found that the grid-pattern nature of the plans could hinder the success of the metro; seven possible improvements were suggested. Overall, the topology of metro networks can play a key role in their success. The concepts presented here can particularly be useful to transit planners; they should also be used along with conventional planning techniques. New transit projects could benefit greatly from an analysis of their network designs, which in turn may play a relevant role in the global endeavour for sustainability.

abstract: Our ability to understand networks is important to many applications, from the analysis and modeling of biological networks to analyzing social networks. Unveiling network dynamics allows us to make predictions and decisions. Moreover, network dynamics models have inspired new ideas for computational methods involving multi-agent cooperation, offering effective solutions for optimization tasks. This dissertation presents new theoretical results on network inference and multi-agent optimization, split into two parts - The first part deals with modeling and identification of network dynamics. I study two types of network dynamics arising from social and gene networks. Based on the network dynamics, the proposed network identification method works like a `network RADAR', meaning that interaction strengths between agents are inferred by injecting `signal' into the network and observing the resultant reverberation. In social networks, this is accomplished by stubborn agents whose opinions do not change throughout a discussion. In gene networks, genes are suppressed to create desired perturbations. The steady-states under these perturbations are characterized. In contrast to the common assumption of full rank input, I take a laxer assumption where low-rank input is used, to better model the empirical network data. Importantly, a network is proven to be identifiable from low rank data of rank that grows proportional to the network's sparsity. The proposed method is applied to synthetic and empirical data, and is shown to offer superior performance compared to prior work. The second part is concerned with algorithms on networks. I develop three consensus-based algorithms for multi-agent optimization. The first method is a decentralized Frank-Wolfe (DeFW) algorithm. The main advantage of DeFW lies on its projection-free nature, where we can replace the costly projection step in traditional algorithms by a low-cost linear optimization step. I prove the convergence rates of DeFW for convex and non-convex problems. I also develop two consensus-based alternating optimization algorithms --- one for least square problems and one for non-convex problems. These algorithms exploit the problem structure for faster convergence and their efficacy is demonstrated by numerical simulations. I conclude this dissertation by describing future research directions. / Dissertation/Thesis / Doctoral Dissertation Electrical Engineering 2017

Real-world networks are dynamic in that both the state of the network components and the structure of the network (topology) change over time. Most studies regarding network evolution consider either one or the other of these types of network processes. Here we consider the interplay of the two, specifically, we consider how changes in network structure effect the dynamics of the network components. To model the growth of a network we use the specialization model known to produce many of the well-known features observed in real-world networks. We show that specialization results in a nontrivial equitable partition of the network where the elements of the partition form clusters that have synchronous dynamics. In particular, we show that these synchronizing clusters inherit their ability to either locally or globally synchronize from the subnetwork from which they are specialized. Thus, network specialization allows us to model how dynamics and structure can co-evolve in real-world systems.

No description available.

A step towards renewable energy is the smart grid, i.e. a power grid that is capable of smart distribution and load managing. This requires the participating components, nodes, to be able to communicate in a reliable fashion. The work of this project was to take on a mathematical approach by using the study of network science, which resulted in the design of a communication network topology for the power grid. The main focus of this undertaking was to quantify how a node in the power grid reacts to change in power flow, which was denoted as its sensitivity. By using Matpower, a tool programmed in Matlab, and our developed method, a communication network was able to be designed using a threshold algorithm. This method worked for the small scale power grids considered in this project, but a different approach might be needed for larger grids.

Grasslands cover 40% of the earth’s land area and provide numerous valuable ecosystem services. However, climate change, global land use change and increasing intensive anthropogenic interventions make grasslands to one of the most endangered ecosystem types in the world. Effective protection in the future requires a fundamental understanding of the dynamics of grasslands and their major drivers. Field experiments have been conducted for impact analyses, for example, with different management intensities, plant community composition and altered climatic conditions. Complementary, ecological models allow to extend the analysis to long-term effects of changes as well as to a deeper understanding of the underlying ecological processes. In this thesis, an individual-based grassland model and network science were applied to understand the community structure and dynamics emerging from individual plant interactions – in relation to plant traits, ecological processes, environmental and anthropogenic impacts, and the small-scale spatial distribution of plants. In the first study, an individual-based process-oriented grassland model was parameterized to simulate field data of a local biodiversity experiment using the concept of plant functional types. The influence of various functional plant traits and ecological processes on grassland productivity and functional composition were analyzed. Different functional plant traits showed partly contrasting effects on plant growth. With regard to the modeled ecological processes, competition for space between plants affected grassland productivity more than shading of plants. In the second study, the parameterized grassland model was used to analyze the impact of functional diversity, mowing frequency and air temperature on ecological processes that lead to changes in grassland productivity. The model reproduced the increase of biomass yields with functional diversity as observed in the field experiment. Modeled plant competition for space showed to be the dominant process and was responsible for an increase in biomass yields in more frequently mown grasslands. In the third study, an approach to generate a regionally transferable parameterization of the grassland model is presented. The impact of management, environment and climate change on productivity and functional composition of grasslands was analyzed within a German-wide scenario analysis. Management intensity had more influence on grassland productivity than environmental factors and correlations of productivity with environmental factors become stronger in less managed grasslands. Climate change showed to have only a minor influence on simulated vegetation attributes. In the fourth study, network science was applied to forest megaplots to quantify the spatial neighborhood structure of species-rich ecosystems. Networks at the individual-tree and tree-species levels revealed similar structures at three investigated forest sites. Tropical tree species coexisted in small-scale networks and only up to 51% of all possible connections between species pairs were realized. A null community analysis showed that details on the tree position and tree size have no major influence on the network structures identified. In summary, this thesis presents the development of advanced methods and analysis tools as well as their application to vegetation ecosystems with high diversity. Thereby, complex structures and dynamics of ecological systems could be systematically explored by combining ecological models with extensive field measurements.