Beamforming for Massive MIMO Aerial Communications: A Robust and Scalable DRL Approach

Keyphrases

• Aerial 33%
• Aerial Communication 100%
• Airborne Platform 33%
• Average Sum Rate 33%
• Beamforming 100%
• Beamforming Vector 33%
• Channel State Information Imperfection 33%
• CNN-based 66%
• Communication Overhead 33%
• Computational Efficiency 33%
• Conjugate Prior 33%
• Decomposition Method 33%
• Deep Deterministic Policy Gradient 33%
• Deep Reinforcement Learning (deep RL) 100%
• Distributed Beamforming 33%
• Entropy-based Approach 33%
• Fourier Neural Operator 33%
• Learning-based 33%
• Local Channel State Information 33%
• Long-range Dependence 33%
• Low-rank Decomposition 33%
• Massive multiple-input 100%
• Multi-agent Deep Reinforcement Learning (MADRL) 33%
• Multiple Input-output 100%
• Multiple Output 100%
• Network Size 33%
• Network Users 33%
• Non-terrestrial 33%
• Non-terrestrial Networks 33%
• Policy Gradient Methods 33%
• Reinforcement Learning Algorithm 100%
• Size-density 33%
• Sum-rate Maximization 33%
• Terrestrial Base Station 33%
• Transfer Learning 33%
• User Density 33%
• User Mobility 33%
• Weighted Minimum Mean Square Error (WMMSE) 66%

Computer Science

• Channel State Information 66%
• Communication Overhead 33%
• Computational Efficiency 33%
• Conjugate Prior 33%
• Convolutional Neural Network 66%
• Deep Reinforcement Learning 100%
• Entropy 33%
• Frequency Domain 33%
• Gradient Method 33%
• Learning Approach 100%
• Multi-Agent Deep Reinforcement Learning 33%
• Neural Operator 33%
• Range Dependency 33%
• sum rate 66%
• Terrestrial Network 33%
• Transfer Learning 33%