Power management for cooperative localization: A game theoretical approach

Network cooperation among agents can significantly increase their position accuracy at the cost of power consumption. Current power management techniques aim at minimizing the total position estimation errors over all the agents subject to the power budgets. There are two main drawbacks for these approaches. First, the performance of a single agent may be sacrificed for the benefit of the whole network, and second, full power budget may be used for only marginal performance improvement on the position accuracy. This paper proposes a new type of power management strategies where each agent individually minimizes its square position error bound (SPEB) penalized by its power cost. The strategies are obtained as solutions to two power management games that are formulated under the knowledge of local information and global information, respectively. We show that agents are more likely to cooperate when global information is available or the channel quality is good. Analytical and numerical results show that the proposed strategies significantly reduce the energy consumption with only marginal performance loss in position accuracy.