Wireless Power Allocation for Hybrid TDMA-NOMA System Using Deep Learning Recurrent Neural Networks

The recurrent neural networks (RNNs) are the type of a specialized artificial neural network that is designed to be operated with time series and sequential data. This technology is increasingly being used in various industries including finance, healthcare, and manufacturing, to name a few. In the hybrid system where a total time available for the passing is split into the multiple another time slots in the designated sub-time slot are assign to service the cluster of the customers. This allocation ensures efficient utilization of the available transmission time and enables the system to cater to multiple customers simultaneously. The total Sub-time slots are created from the broadcast duration, the system can allocate resources to clusters of customers, ensuring that each cluster receives the required level of service. The implementation of this approach facilitates optimal resource management and system performance, thereby enhancing the overall customer experience. Through the channelling of im- proved resource management, system performance are optimized. Customer experience is enhanced. Using NOMA technique, each cluster's users receive signals. This project deals with the selection of users in a collaborative NOMA frame-work that includes a central station. A far user, and nearby users in order to assist a far user, The chosen individual who is in close proximity needs to receive a message and then pass it on according to the stated requirements. The challenge at hand is to develop a strategy that maximizes the chosen close to the user's data rate while maintaining the required level of service quality requirements of more user. We request you to propose such a strategy that takes into account the aforementioned concerns. When using real-world communication systems, users' channel state data (CSI) can be imperfect I.e. This study examines the outage probability of the NOMA- Non-Orthogonal Multiple Access system under imperfect CSI, according to the consumer's selected method, and derives A term in closed form. Theoretical examination reveals Under faulty CSI, the NOMA system's diversity order is zero, indicating that the multiuser diversity order is lost. To address performance degradation the system under imperfect of CSI, we propose an approach that employs the Deep Learning Recurrent Neural Network method to classify the channels with enhance the accuracy of CSI and rectify any errors. Implementation of the proposed approach is carried out using the MATLAB software.