Traffic-Prediction-Based Dynamic Resource Control Strategy in HAPS-Mounted MEC-Assisted Satellite Communication Systems

Satellite communication is increasingly essential and widely used, especially with the rapid development of the Internet of Things (IoT) and networks beyond fifth-generation (B5G), providing ubiquitous coverage. However, the current reactive approaches to optimize resources have become inadequate due to the massive rise in IoT traffic with varying patterns and limited resources of satellite networks. These approaches fail to predict dynamic traffic and its requirements. To efficiently allocate satellite communication resources as necessary, there is a need to proactively predict traffic demand. We propose utilizing mobile edge computing-enabled high-altitude platform stations (HAPSs) to predict traffic from ground users to satellite networks at HAPS. However, resource control based on traffic prediction in satellite networks faces challenges, such as the wastage of resources or insufficient resource availability, due to misaligned traffic variations and resource control timing. To overcome these challenges, we propose a dynamic scheduling strategy for resource control based on traffic demand prediction. This strategy aims to reduce resource wastage in satellite communication systems. Our proposed approach can predict traffic with high accuracy and allocate resources with minimal difference between achievable throughput and required throughput, demonstrating high-resource utilization. We evaluated the effectiveness of our scheduling strategy through simulation analysis by comparing it with periodic resource control.