A lightweight sensor scheduler based on AoI function for remote state estimation over lossy wireless channels

This article investigates the problem of sensor scheduling for remotely estimating the states of heterogeneous dynamical systems over resource-limited and lossy wireless channels. Considering the low time complexity and high versatility requirements of schedulers deployed on the transport layer, we propose a lightweight scheduler based on an Age of Information function built with the tight scalar upper bound of the remote estimation error. We show that the proposed scheduler is indexable and suboptimal. We derive an upper and a lower bound of the proposed scheduler and give stability conditions for estimation error. Numerical simulations demonstrate that, compared to existing policies, the proposed scheduler achieves estimation performance very close to the optimal at a much lower computation time.