Dynamic Beam Hopping Method Based on Multi-Objective Deep Reinforcement Learning for Next Generation Satellite Broadband Systems.
Xin HuYuchen ZhangXianglai LiaoZhijun LiuWeidong WangFadhel M. GhannouchiPublished in: IEEE Trans. Broadcast. (2020)
Keyphrases
- multi objective
- optimization method
- optimization algorithm
- reinforcement learning
- preprocessing
- high precision
- evolutionary algorithm
- probabilistic model
- trade off
- cost function
- high accuracy
- support vector machine svm
- dynamic environments
- computational cost
- clustering method
- distributed systems
- feature set
- learning algorithm
- significant improvement
- pairwise
- image sequences