R. Setola

Exploiting routing information in Wireless Sensor Networks localization

In this work, an integrated solution to jointly solve the localization and data routing problems in sensor networks is proposed. It is based on a scalable “divide et impera” approach. The network is divided into 1-hop cluster and the localization is hierarchically performed inside each cluster, thereafter among clusters. The key advantage of the algorithm is that no additional communication is required to perform localization, which is obtained by a ranging technique, exploiting network topology provided by the routing algorithm. To this end, Shadow Edges, a novel class of links, are introduced to encompass the lack of communication among nodes. The proposed dual procedure to solve the localization and routing, represents a novel trend for Wireless Sensor Networks, where multiple problems are jointly solved by providing affordable and integrated solutions.

Assessing protein resilience via a complex network approach

In recent years the topological study of proteins is gaining momentum rapidly, and several studies are providing more and more insights on the structural and dynamical properties of proteins by exploiting topological indexes based on Complex Network Theory. To this end the amino acid residues play the role of graph vertices, while non-covalent contacts are the arcs. Topological structure of proteins can be imagined as resulting by folding a thread of pearls (primary sequence of aminoacids) in which amino acid (nodes) relatively distant along the sequence come into contact thanks to the folding process. The result is a configuration sharing some properties with Complex Networks. In this work we derive insights on the resilience of protein contact networks by evaluating the degradation in the size of the giant component with respect to iterated node removal. Specifically, several strategies based on topological indicators (e.g., removing nodes in descending order of clustering coefficient) are exploited, considering the human serum albumin as case study. The analysis of progressive giant component desegregation offered some interesting hints about protein folding principles and suggested some strategies to locate the amino acids most relevant for stability of the studied molecule.

Infrastructure Interdependencies: Modeling and Analysis

In this chapter some of the most well established approaches to model Critical Infrastructure Interdependencies are discussed. Specifically the holistic methods, where the interaction among infrastructures is seen from a very high level of abstraction, are compared with agent-based models, where the dependency phenomena that may arise among subsystems are considered both in terms of functional and topological relations. In order to better clarify the different approaches, the Input–Output Inoperability Model is discussed as one of the most representative Holistic methodologies; Agent-based methods are then discussed with particular reference to the Agent-Based Input–Output Inoperability Model, an extension of the Input–Output Inoperability Model, developed by the authors. The increased level of detail clashes with the lack of adequate quantitative data required to tune the models, which are typically assessed based on economic exchange among infrastructures; in order to partly overcome such an issue, an Input–Output methodology based on the theory of Fuzzy Systems is discussed. Finally, some conclusive remarks and open issues are collected.