Impact of Transmission Delays Over Age of Information Under Finite Horizon Scheduling

Abstract

Sensor communications in real-time systems may be required to report status updates to minimize the so-called age of information metric, which quantifies the freshness of exchanged data. This situation can be heavily impacted by the delays in data transmission, as some updates may reach the destination when their information content is already stale. In this paper, we consider the problem of scheduling sensing updates over a finite time horizon, and discuss the impact of their transmission delay on the resulting data freshness. We tackle the adjustments required for an offline schedule aimed at minimizing the average age of information which, as long as they do not cause the updates to go off the boundaries of the finite horizon, are only dependent on the average transmission delay. We derive a closed form expression for the average age of information, also verified through simulations, and the resulting performance is evaluated under different system conditions. This can be used to further explore the task of delivering timely system updates under general scenarios, e.g., when the statistics of the delay is not known a priori, or under other non idealities.