Deep Q-Network Enabled RIS-CNOMA System with Receiver Diversity Techniques under Nakagami-m Fading channel
This paper presents an intelligent optimization framework for a Reconfigurable Intelligent Surface supported Cooperative Non-Orthogonal Multiple Access System operating under Nakagami-$m$ fading environment. In traditional approaches, key system parameters such as RIS phase configurations and power allocation coefficients are typically predetermined, which restricts adaptability in dynamic wireless conditions. To overcome this limitation, a Deep Q-Network (DQN) based strategy is proposed to enable real-time adaptation of these parameters by interacting with the channel environment. In addition, receiver diversity methods, including Maximal Ratio Combining , Selection Combining , and Equal Gain Combining are incorporated to further enhance signal reception and combat fading effects for both the nearby user and the distant user. The system performance is evaluated with respect to outage probability across different signal-to-noise ratio (SNR). The simulation results show that the proposed DQN-enabled RIS-CNOMA system offers up to 95\% reduction in outage probability and improves system reliability by 90\%--95\% compared to conventional non-learning-based schemes. Overall, the proposed learning-driven framework offers an adaptive and efficient solution to increase the effectiveness of RIS-assisted CNOMA systems in realistic fading scenarios, making it a promising solution for future wireless communication networks.
Authors
- Kiran Dasari (ORCID: https://orcid.org/0000-0002-3680-8058)
- Ranjeeth Mamidi (ORCID: https://orcid.org/0000-0002-1133-8563)
- Ramadevi Maguluri
- Dhanalakshmi Narra
Institutions
- Manipal Academy of Higher Education (IN)
- Vallurupalli Nageswara Rao Vignana Jyothi Institute of Engineering and Technology (IN)
Publication Details
- Journal
- Research Square
- Published
- 2026-11-18
- DOI
- https://doi.org/10.21203/rs.3.rs-11293767/v1
- Primary Topic
- Advanced Wireless Communication Technologies
- Type
- preprint
Funders
- National Institute of Technology Warangal