Distributed estimators for sensor networks are discussed. The considered problem is on how to track a noisy timevarying signal jointly with a network of sensor nodes. We present a recent scheme in which each node computes its estimate as a weighted sum of its own and its neighbors ’ measurements and estimates. The weights are adaptively updated to minimize the variance of the estimation error. Theoretical and practical properties of the algorithm are illustrated. The results provide a tool to trade-off communication constraints, computing efforts and estimation quality.