Author(s): Nada Taher Malik

Abstract: There are problems in O-NOMA systems, which can be addressed using the method of deep learning called DLM, that result from overlapping and entangling of the system signals. The NOMA method increases the efficiency of the optical frame spectrum, allowing multiple people to use the same moment and frequency resources. However, the interference patterns Complex and non-stationary channel conditions in NOMA-DLM systems pose challenges to traditional detection methods, and by taking advantage of deep neural networks, these methods can surpass these challenges and as such improve their performance in detection. This article gives us a brief insight into the primary aspects and benefits of DLM detection methods when integrated with M-MIMO O-NOMA systems with underlying features like training procedures in addition to network design. DLM networks are designed to work with received codes or decoded data streams from the input signal, power allocation coefficients, additional information, and regression enhancement are also considered when updating the network parameters in the course of training this study also aims at examining the challenges faced by deep learning when applied in O-NOMA systems.

DOI: 10.33545/2707661X.2025.v6.i2b.147

Pages: 109-117 | Views: 156 | Downloads: 83

Download Full Article: Click Here