PhD Alumni


Malanchini Ilaria

Present position: Postdoc at Politecnico
http://home.dei.polimi.it/malanchini/
 

Thesis title:  Game Theoretic Models for Resource Sharing in Wireless Networks
Advisor:  Matteo Cesana
Research area:  Networks
Thesis abstract:  
Wireless communications have been recently characterized by rapid proliferation of wireless networks, impressive growth of standard and technologies, evolution of the end-user terminals, and increasing demand in the wireless spectrum. New, more flexible schemes for the management of the available resources, from both the user and the network side, are necessary in order to improve the efficiency in the usage of the available resources.

This work aims at shedding light on the performance modeling of radio resource sharing/allocation situations. Since, in general, the quality of service perceived by a system (e.g., user, network) strictly depends on the behavior of the other entities, and the involved interactions are mainly competitive, this work introduces a framework based on non–cooperative game theoretic tools to characterize radio resource sharing/allocation situations.
Non–cooperative game theory is suitable in distributed networks, where control and management are inherently decentralized. In detail, we propose non–cooperative game theoretic models for network selection and spectrum sharing by considering different scenarios where both users and networks take part in the game.

First, we consider wireless networks, where many users have to make decisions on which access point to connect to. In this scenario, the quality perceived by the users mainly depends on the number of other users, i.e., congestion, choosing the very same accessing opportunity. In this context, we also consider two–stage games where network make decisions on how to use the available resources, and users react to this selecting the network that maximizes their satisfaction. Then, we refer to the problem of spectrum sharing, where users directly compete for portions of the available spectrum. Finally, we provide a more complex model where the users utility function is based on the Shannon rate. The aim of this second part is to provide a better representation of the satisfaction perceived by the users, i.e., in terms of achievable throughput. Due to the complexity of the game model, we first provide a complete analytical analysis of the two–user case. Then, we extend the model to the N–user case. We mainly analyze this game through simulations. Finally, inspired by the results obtained numerically, we introduce stochastic geometry in the analysis of spectrum games in order to predict the performance of the game in large networks.


Curriculum: