Questo sito utilizza cookie di terze parti per inviarti pubblicità in linea con le tue preferenze. Se vuoi saperne di più clicca QUI 
Chiudendo questo banner, scorrendo questa pagina, cliccando su un link o proseguendo la navigazione in altra maniera, acconsenti all'uso dei cookie. OK

Game Theory for Internetworking

This report offers a simple introduction to game theory, and presents possible application to Internetworking issues.
The game theoretical introduction is purged of historical references and mathematical formalisms to aid young researches approaching the discipline.
We depict principles of game theory, detailing non-cooperative and cooperative game modelling with examples and some related issues.
Then, we present applications of non-cooperative game theory to Internetworking, meaning with Internetworking the interactions between Autonomous Systems at the IP Internet network layer.
Looking for conflict situations in IP core networks, we apply non-cooperative game theory to engineer the bottleneck that nowadays stress inter-carrier peering settlements.
We define the peering non-cooperative framework and highlight coordinated routing strategies upon Nash equilibria and Pareto-efficient profiles.
We also propose a novel peering model for the next generation Internet, the Extended Peering Model, and show how the game theoretical non-cooperative framework can be extended for its implementation.

This research report is due to the minor research activity Stefano Secci carried out within the Ph.D. programme at the Politecnico di Milano, under the advisory of Fioravante Patrone, Professor of Game Theory at the Università di Genova.

Mostra/Nascondi contenuto.
Chapter 1 Principles of Game Theory Game theory is a mathematical theory, a discipline, that even a reader with scarce mathematical basis can adopt to analyse every day life problems: simple classical games, but also social behaviours, financial trends or telecommunication systems. In this chapter we depict principles of game theory, voluntarily avoiding math- ematical formalisms and historical facts for the sake of seamlessness. Our first references are, for an Italian reader, the book from Fioravante Patrone [1] and, for an English reader, the book from Osborne [2]. Game theory can intervene in all those situations in which more than one decision-makers have to plan their optimal strategy to solve a problem. Hence, it does not intervene in those situations with a single decision-maker, ground for operations research. In the following, before diving into the ways in which a player’s strategy is to be built and modelled, we highlight principles, concepts and assumptions of game theory, starting from the characterization of the decision-maker. 1.1 The decision-maker, i.e. the player The reader may think that life is a sentence to forced works, full of duties and difficulties. Others may, instead, think that life is a game, and that a person in charge of a decision, the decision-maker, is a player, as the game theory predicates. Not whatever player: an intelligent and rational player1. 1.1.1 Intelligence The starting hypothesis of game theory is that the player is infinitely intelligent, in the sense that he has infinite ability of deduction, computation and analysis. Besides being not really true for humans, it is a strong hypothesis especially for wide games, such as the chess, since just the simple enumeration of all the possible strategies may take a lot of resources. However, the vast majority of the problems 1From now on, we will refer to the decision-maker or to the player with the third male person for simplicity. 1

Tesi di Dottorato

Dipartimento: Dipartimento di Elettronica e Informazione

Autore: Stefano Secci Contatta »

Composta da 61 pagine.

 

Questa tesi ha raggiunto 601 click dal 22/10/2008.

 

Consultata integralmente una volta.

Disponibile in PDF, la consultazione è esclusivamente in formato digitale.