A Green Computing Business Aggregation Strategy for Low Earth Orbit Satellite Networks

Area of Science:

• Computer Science
• Aerospace Engineering
• Network Engineering

Background:

• Low Earth Orbit (LEO) satellite networks (LSNs) face dynamic, energy-constrained environments.
• Optimizing energy efficiency and minimizing delay are critical for sustainable LSN operations.

Purpose of the Study:

• To propose a novel green computing strategy for LSNs.
• To effectively balance energy savings and delay costs in dynamic LSN environments.

Main Methods:

• Integration of Markov Decision Process (MDP) with Double Deep Q-Network (Double DQN).
• Introduction of the Energy-Delay Ratio (EDR) metric for quantifying and balancing performance.

Main Results:

• Significant energy savings observed: up to 47.87% (low volume), 26.75% (medium volume), and 4.36% (high volume).
• Minimal delay increases: 0.0161 s (low), 0.0189 s (medium), and 0.0299 s (high) across business volumes.
• Demonstrated adaptability to varying traffic loads.

Conclusions:

• The proposed strategy effectively balances energy efficiency and delay in LSNs.
• The approach is suitable for sustainable operations in dynamic LEO satellite network conditions.