Information Update for Multi-source Remote Estimation in Real-time Monitoring Networks

We study the remote estimation problem with a large number of wireless monitoring devices. As a first step, we model the states of multiple information sources as a multi-dimensional symmetric random walk process, where each source updates its status through costly transmission channel. We formulate the problem as an infinite-horizon optimization problem and consider threshold-based policy that achieves the optimal performance in many network settings. Depending on the situations, it might be possible to update each information source individually, or a group of sources must be updated together. To this end, we investigate the impact of different communication constraints on the performance of the threshold-based policy, and clarify the performance gain of the individual update.