An improved deep learning framework for enhancing mimo-Noma system performance

• Published in: Multimedia tools and applications 2024-03, Vol.83 (8), p.22581-22608
• Main Authors: Prabakaran, N., Devi, R. Prameela
• Format: Article
• Language: English
• Subjects: Algorithms; Computer Communication Networks; Computer Science; Data Structures and Information Theory; Deep learning; Heuristic methods; Multimedia Information Systems; Nonorthogonal multiple access; Optimization; Performance enhancement; Random noise; Special Purpose and Application-Based Systems
• ISSN: 1573-7721; 1380-7501; 1573-7721
• DOI: 10.1007/s11042-023-16259-z

Non-orthogonal multiple access (NOMA) is an effective mechanism based on the multiple access technique. Here, the MIMO-NOMA system is considered to deal with problems especially energy and spectral efficiency. To improve the performance based on energy and spectral efficiency, the proposed technique is incorporated with a deep learning-based technique with a hybrid Meta-heuristic algorithm. The algorithms including Harris Hawks Optimization (HHO) and Mayfly optimization (MF) are considered for enhancing the MIMI-NOMA system. Moreover, with this proposed model, the automatic encoding, decoding, and channel detection in an Additive White Gaussian Noise (AWGN) channel. The deep learning technique incorporated with DNN is based on conventional user activity and data detection techniques. In specific, the user activity and data in the environment are in a non-linear form that can be approximated by the proposed model. The proposed HHMF-DNN-NOMA is implemented into the platform of MATLAB 2020a and the proposed model is evaluated with the existing model to prove the performance is better than existing.