Dynamic Offloading and CPU Clock Scaling for Edge Computing in LEO Satellite Networks

Low Earth Orbit (LEO) satellite communication is an effective solution for addressing global coverage challenges in terrestrial networks. As we enter the 6G era, the rise of complex applications has increased device computational demands, driving the adoption of techniques like offloading and Edge Computing for efficient processing. To achieve seamless global 6G service integration, active research is underway in Satellite Edge Computing (SEC). SEC involves deploying computation-capable servers on LEO satellites, enabling edge computing. However, many studies extend terrestrial Edge Computing techniques to satellite networks without adequately considering dynamic channel conditions due to physical topology changes. Our research proposes an Edge Computing framework tailored to LEO satellite networks, effectively addressing dynamic channel conditions. We introduce optimization algorithms to minimize system power consumption and propagation latency. Simulations in our proposed SEC scenario confirm our algorithm’s adaptability to changing channel conditions and service requests, resulting in lower power consumption compared to conventional methods for the same processing latency.