Event-triggered consensus on exponential families

The paper deals with discrete-time event-triggered consensus on exponential families of probability distributions (including Gaussian, binomial, Poisson and many other distributions of interest) completely characterized by a finite-dimensional vector of so called natural parameters. It is first shown how such exponential families are closed under Kullback-Leibler fusion (average), and that the latter is equivalent to a weighted arithmetic average over the natural parameters. Then, a novel event-triggered transmission strategy is proposed so as to tradeoff data communication rate versus consensus speed and accuracy. Some numerical examples are worked out to demonstrate the effectiveness of the proposed method. It is expected that eventtriggered consensus can be successfully exploited for bandwidthefficient networked state estimation.