Learning-Based Adaptive IRS Control With Limited Feedback Codebooks

Mathematics

• Limited Feedback 100%
• Codebook 77%
• Learning 46%
• Reflection Coefficient 21%
• Channel Estimation 20%
• Adjacency 12%
• Policy 11%
• Time-varying Channels 10%
• Neural Networks 10%
• Multipath 8%
• Random Design 7%
• Propagation 5%
• Angle 4%
• Optimization 4%
• Evaluation 4%
• Design 3%

Engineering & Materials Science

• Control surfaces 59%
• Atoms 35%
• Channel estimation 34%
• Deep neural networks 33%
• Feedback 32%
• Conveying 19%
• Base stations 14%