QoE-Aware Flexible Capacity Allocation Planning for Multi-Service Satellite Communication Systems

Satellite operators worldwide are in a race to deploy and enhance connectivity supporting diverse 5G applications and services, with success depending on the ability to deliver superior Quality of Experience (QoE) tailored to each service, despite limited network capacity. However, this effort is challenged by unpredictably fluctuating traffic demands, distinct packet arrival distributions across services, and evolving stochastic user QoE expectations. This paper addresses these challenges by formulating a statistical optimization problem that minimizes allocated capacity (intending to accommodate more users) while satisfying specific QoE requirements, such as queuing delay. To achieve this, we leverage packet queuing analysis within the buffer system of the SatCom gateway's forward link. Given the complexity of solving the problem directly, we first approximate its constraints using probabilistic analysis. Then, we propose a multi-agent Double Deep Q-Network (DDQN) algorithm that enables a more accurate representation of queue-length states and facilitates better decision-making by the agents. The approach leverages episodic training to ensure agents are well-prepared and optimized through simulations before being deployed in a real-time environment. Extensive simulation campaigns validate the effectiveness of our method, demonstrating clear improvements over benchmark algorithms.