Information Freshness Optimal Resource Allocation for LEO-Satellite Internet of Things

Low earth orbit satellite Internet of Things (LEO-SIoT) can provide services such as remote control for IoT devices in mountainous regions and oceans. In LEO-SIoT, time-sensitive applications such as natural disaster warnings and wildlife tracking often require the remote control center to receive the latest status updates from IoT devices in real-time. However, due to the long signal propagation distance and severe constraints on satellite computing resources, LEO-SIoT fails to meet the high information freshness requirements of time-sensitive applications. It is crucial to efficiently allocate resources in LEO-SIoT to improve its information freshness. To address this issue, firstly, a LEO-SIoT architecture for cloud-edge-end collaborative task processing is proposed by employing edge intelligence (EI) technology. Then, the introduction of peak age of information (PAoI) as the metric for the information freshness in the LEO-SIoT is followed by the proposal of a freshness-fairness-aware scheduling strategy. Moreover, an optimization model is developed under the terminal energy constraint with the objective of minimizing the average PAoI. A information freshness optimal resource allocation algorithm, utilizing convex optimization and search algorithm, is proposed to minimize the average PAoI in the LEO-SIoT. Experiments and results demonstrate that the proposed resource allocation algorithm and task scheduling strategy effectively improve information freshness and freshness fairness.