Search

Cooperative Non-Orthogonal Multiple Access (Cooperative NOMA) is a communication technique used in wireless networks to enhance spectral efficiency and overall system performance. Non-Orthogonal Multiple Access is a multiple access scheme that allows multiple users to share the same time-frequency resources by allocating different power levels to each user. This paper presents a comprehensive performance validation of NOMA empowered by optimal relay selection and deep learning-based power allocation techniques. To achieve effective data transmission to far users, the optimal relay selection process employs the Maximum Energy Harvested Relay (MEHR) selection approach, which intelligently selects relays based on their energy harvesting capabilities. Furthermore, an enhanced Deep Convolutional Optimized Neural Network (DCOnN) is utilized for optimal power allocation among users in a cooperative NOMA system. To efficiently optimize the power allocation process with the DCOnN model, the Grey Wolf (GW) optimization is utilized to fine-tune the network weights. The presented approach significantly increases the energy efficiency in cooperative NOMA networks. The combination of optimal relay selection and deep learning-based power allocation enhances outage probability performance, ensuring better coverage and communication reliability while reducing the bit error rate for users in challenging channel conditions. Finally, the integration of optimal relay selection using DCOnN with GW optimization and cooperative cognitive NOMA provides a powerful framework to achieve superior communication performance in wireless networks. The proposed scheme achieves higher performance in terms of energy efficiency (519.257 Bits/J/Hz) and CPU time (0.012 seconds). The experimental results proved that the proposed performance is enhanced than the different existing schemes. [ABSTRACT FROM AUTHOR]