Distributed estimation based on game theory in energy harvesting wireless sensor networks

For stochastic unstable problems of ambient energy harvesting in wireless sensor networks, a kind of distributed estimation algorithms based on perfect and complete information extensive game theory is proposed. A quantized distributed estimation problem is under investigation. Firstly, the ambient energy harvesting model and the energy consumption model of nodes are analyzed. Then, its extensive game theory model and game tree are both developed. Further, the existence and its solution of the refined Nash equilibrium of sub-game theory are discussed. Lastly, simulations show that the proposed algorithm can better deal with the random disturbance of ambient energy than the conventional distributed estimation scheme.