Game-theoretic learning for activation of diffusion least mean squares

This paper presents a game-theoretic approach to node activation control in parameter estimation via diffusion least mean squares (LMS). The energy-aware activation control is a noncooperative repeated game where nodes autonomously decide when to activate based on a utility function that captures the trade-off between node's contribution and energy expenditure. The proposed two time-scale stochastic approximation algorithm ensures the parameter estimates weakly converge to the true parameter across the network, yet the global activation behavior along the way tracks the set of correlated equilibria of the underlying activation control game.