Achievable Throughput and On-board Buffer Sizing in RIS-UAV Relay-Assisted Satellite-Aerial-Ground Integrated Networks (SAGINs)

Satellite-Aerial-Ground Integrated Networks (SAGINs) and Low Earth Orbit (LEO) satellites are emerging as an essential network architecture for next generation mobile networking. Depending on the weather conditions, LEO satellites and SAGINs deploy the Free-Space Optical (FSO) and Radio Frequency (RF) links in tandem and can flexibly configure different network components, using High Altitude Platform Station (HAPS) and Reconfigurable Intelligent Surface (RIS) assisted Unmanned Aerial Vehicle (UAV). In SAGIN, since signal-to-noise ratio is majorly impacted by the weather, the interference plays an important role, and especially the co-channel interference from satellites to HAPS, or from HAPS to RIS-UAV. The interference management directly impacts the buffer sizing at the HAPS that stores data for scheduling transmissions. Practical dimensioning of the buffer at HAPS is an open, yet critical challenge due to the sizing constraints of the SAGIN system components. Motivated by this challenge, we analyze the throughput and buffer size in SAGIN systems. To this end, we develop a multi-dimensional Markov chain model and analyze varying weather conditions, including impact of past weather states, types of transmission links used (FSO/RF), and the interference nulling problem on RIS-UAV. Numerical results show that the model can derive practical throughput and buffer sizes for a range of weather conditions and link configurations.