Decentralized activation in sensor networks – global games and adaptive filtering games

This paper considers two methodologies for decentralized sensor activation in wireless sensor networks for energy-efficient monitoring. First, decentralized activation in wireless sensor networks is investigated using the theory of global games. Given a large number of sensors which can operate in either an energy-efficient “low-resolution” monitoring mode, or a more costly “high-resolution” mode, the problem of computing and executing a strategy for mode selection is formulated as a global game with diverse utilities and noise conditions. We formulate Bayes–Nash equilibrium conditions for which a simple threshold strategy is competitively optimal for each sensor, and propose a scheme for decentralized threshold computation. The second class of results we consider is in a non-Bayesian context where sensors deploy simple adaptive filtering algorithms and the global behavior converges to the set of correlated equilibria.