A Game Theoretic Load-Aware Network Selection

Majed Haddad (1) , Piotr Wiecek (2) , Oussama Habachi (3)

1 LIA - Laboratoire Informatique d'Avignon
2 Wroclaw University of Science and Technology
3 UNILIM - Université de Limoges

This paper presents a load-aware network selection model to help users to determine whether or not to connect to a macro cell (MC) or a WiFi access point (AP) in a vertically integrated cellular WiFi system. The problem is formulated as a game theoretic model in which users selfishly maximize their throughput. Unlike most existing work, we do not assume that users have complete information about the other users' dynamics, which is more realistic in a communication network with distributed users where conveying such information would create enormous overhead. To do so, we provide a simple procedure that introduces a full hierarchy among the users reflecting in an accurate way their channel quality and then allowing them to sequentially choose their preferred network. Interestingly, the result for each user in the payoff is not far from his optimum.

Domaines: Réseaux et télécommunications [cs.NI]; Intelligence artificielle [cs.AI]; Théorie de l'information [cs.IT]

Majed Haddad, Piotr Wiecek, Oussama Habachi. A Game Theoretic Load-Aware Network Selection. 2018 IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), Apr 2018, Honolulu, United States. pp.577-582, ⟨10.1109/INFCOMW.2018.8406908⟩. ⟨hal-03485183⟩