LATENCY OPTIMIZATION IN POWER-DOMAIN NOMA USING AI-ENHANCED RSSI FEEDBACK
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Nonorthogonal multiple access (NOMA) is a key multiple-access technique for 5G and beyond, enabling simultaneous multiuser transmission via power-domain multiplexing. However, conventional power domain NOMA often relies on fixed or slowly adaptive power-control updates. It increases power-adjustment latency and leads to inefficient interference handling and throughput loss under time-varying channels. This paper proposes a latency time improvement (LTI) framework that utilizes received signal strength indicator-driven feedback to guide adaptive multiphase power updates and reinforcement learning for long-term power control decisions. Users are grouped via K-means clustering; a two-stage coarse-fine adjustment reduces overshoot and correction delay, and tabular Q-learning selects power actions to accelerate convergence to signal-to-interference-plus-noise ratio targets. Simulation results under fading channels show that LTI reduces power-adjustment latency by 20%-30% while improving spectral efficiency, interference behavior, and bit error rate compared with classical power domain NOMA baselines.
Fields of Study & Subjects
Advanced Wireless Communication Technologies
PAPR reduction in OFDM
Advanced MIMO Systems Optimization
Publication details
- Journal / source
- Telecommunications and Radio Engineering
- Volume
- 85
- Issue
- 5
- Pages
- 73–93
- Publication date
- 1 January 2026
- Language
- EN