Kalman Filter in Lossy Networks

Bruno Sinopoli and Luca Schenato
(Professors S. Shankar Sastry and Michael I. Jordan)

We are considering the problem of evaluating the performance of Kalman filtering techniques on a sensor network. Communication delay and loss of data cannot be neglected in this new scenario. Starting from the discrete Kalman filtering formulation, making use of graphical models, we have made a first attempt to generalize the Kalman filter to take into account a model for the communication accross the network sensor.


Send mail to the author : (sinopoli@eecs.berkeley.edu)


Edit this abstract