1: \begin{abstract}

2:   In this paper we consider a network of agents monitoring a spatially

3:   distributed arrival process. Each node measures the number of arrivals seen at

4:   its monitoring point in a given time-interval with the objective of estimating

5:   the unknown local arrival rate. 

6:   % 

7:   We propose an asynchronous distributed approach based on a Bayesian model with

8:   unknown hyperparameter, where each node computes the minimum mean square error

9:   (MMSE) estimator of its local arrival rate in a distributed way. As a result,

10:   the estimation at each node ``optimally'' fuses the information from the whole

11:   network through a distributed optimization algorithm. Moreover, we propose an

12:   \emph{ad-hoc} distributed estimator, based on a consensus algorithm for

13:   time-varying and directed graphs, which exhibits reduced complexity and

14:   exponential convergence.

15:   % 

16:   We analyze the performance of the proposed distributed estimators, showing

17:   that they: (i) are reliable even in presence of limited local data, and (ii)

18:   improve the estimation accuracy compared to the purely decentralized setup.

19:   % 

20:   Finally, we provide a statistical characterization of the proposed estimators.

21:   In particular, for the ad-hoc estimator, we show that as the number of nodes

22:   goes to infinity its mean square error converges to the optimal one.

23:   %

24:   Numerical Monte Carlo simulations confirm the theoretical characterization and

25:   highlight the appealing performances of the estimators.

26: \end{abstract}

27: