Distributed Online Self-Localization and Tracking in Sensor Networks

Recursive maximum likelihood (RML) and expectation maximization (EM) are a popular methodologies for estimating unknown static parameters in state-space models. We describe how a completely decentralized version of RML and EM can be implemented in dynamic graphical models through the propagation of suitable messages that are exchanged between neighboring nodes of the graph. The resulting algorithm can be interpreted as an extension of the celebrated Belief Propagation algorithm to compute likelihood gradients. This algorithm is applied to solve the sensor localization problem for sensor networks. An exact implementation is given for dynamic linear Gaussian models without loops.