Power constrained distributed estimation with correlated sensor data

We consider distributed estimation of a random scalar parameter in a power constrained wireless sensor network (WSN), where the measurements are sent from the sensors to the fusion center (FC) over noisy wireless channels by employing an analog transmission scheme. We study the power allocation problem with generally correlated sensor observations that can accommodate nonlinear measurement models and spatially correlated observation noise. An effective solution is developed by utilizing a tractable lower bound of the objective function. The proposed algorithm is also extended for random field estimation. Simulation results are presented to illustrate the effectiveness of the proposed algorithm.