Optimizing Age of Information in Wireless Uplink Networks with Partial Observations

TL;DR

This paper develops a low-complexity scheduling policy for wireless uplink networks to optimize Age of Information, despite partial observations and stochastic update arrivals, by simplifying the belief-MDP framework.

Contribution

It introduces a simplified belief-MDP approach leveraging Bernoulli arrivals to design an efficient scheduling policy for AoI optimization.

Findings

The proposed policy outperforms baseline methods in simulations.

Upper bounds for AoI performance are established.

Performance guarantees are validated through numerical analysis.

Abstract

We consider a wireless uplink network consisting of multiple end devices and an access point (AP). Each device monitors a physical process with stochastic arrival of status updates and sends these updates to the AP over a shared channel. The AP aims to schedule the transmissions of these devices to optimize the network-wide information freshness, quantified by the Age of Information (AoI) metric. Due to the stochastic arrival of the status updates at the devices, the AP only has partial observations of system times of the latest status updates at the devices when making scheduling decisions. We formulate such a decision-making problem as a belief Markov Decision Process (belief-MDP). The belief-MDP in its original form is difficult to solve as the dimension of its states can go to infinity and its belief space is uncountable. By leveraging the properties of the status update arrival (i.e., Bernoulli) processes, we manage to simplify the feasible states of the belief-MDP to two-dimensional vectors. Built on that, we devise a low-complexity scheduling policy. We derive upper bounds for the AoI performance of the low-complexity policy and analyze the performance guarantee by comparing its performance with a universal lower bound. Numerical results validate our analyses.