AoI and Data Rate Optimization in Aerial IRS-Assisted IoT Networks

The unmanned aerial vehicle (UAV) with intelligent reflecting surface (IRS) mounted, namely, aerial IRS, has the potential in improving the information freshness (IF) and transmission data rate for wireless networks, where age of information (AoI) is generally utilized to characterize the IF. In this article, we optimize both the AoI and transmission data rate in aerial IRS-assisted Internet of Things (IoT) networks through the joint transmission scheduling, UAV location, and IRS phase shift matrix design. We formulate a multiobjective optimization problem, which simultaneously minimizes the system average AoI and maximizes the overall transmission data rate. The optimal solutions for the two objectives in the formulated problem are not always consistent with each other. Besides, the optimization problem with either objective is nonconvex and difficult to tackle directly. An effective three-step scheme is developed in this article to solve the formulated problem. To be more specific, firstly the UAV locations are optimized through a
$Q$
-learning-based scheme to maximize the overall data rate while guaranteeing the signal-to-noise ratio (SNR) constraint of each IoT device; then the IRS phase shift matrices are determined through a low-complexity Tabu-search-based scheme to further improve the overall data rate given the SNR constraint of each device; finally, the transmission scheduling is performed to optimize the system AoI based on a deep
$Q$
-network algorithm. Simulation evaluation demonstrates that the proposed scheme outperforms existing ones.