Adversarial control in a delay tolerant network

  • Authors:
  • Eitan Altman;Tamer Başar;Veeraruna Kavitha

  • Affiliations:
  • INRIA, Sophia Antipolis, France;University of Illinois, Urbana, IL;INRIA, Sophia Antipolis, France and LIA, Avignon University, France

  • Venue:
  • GameSec'10 Proceedings of the First international conference on Decision and game theory for security
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

We consider a multi-criteria control problem that arises in a delay tolerant network with two adversarial controllers: the source and the jammer. The source's objective is to choose transmission probabilities so as to maximize the probability of successful delivery of some content to the destination within a deadline. These transmissions are subject to interference from a jammer who is a second, adversarial type controller, We solve three variants of this problem: (1) the static one, where the actions of both players, u and w, are constant in time; (2) the dynamic open loop problem in which all policies may be time varying, but independent of state, the number of already infected mobiles; and (3) the dynamic closed-loop feedback policies where actions may change in time and may be specified as functions of the current value of the state (in which case we look for feedback Nash equilibrium). We obtain some explicit expressions for the solution of the first game, and some structural results as well as explicit expressions for the others. An interesting outcome of the analysis is that the latter two games exhibit switching times for the two players, where they switch from pure to mixed strategies and vice versa. Some numerical examples included in the paper illustrate the nature of the solutions.