Abstract
Abstract
The high demand for wireless communication and limited spectral power causes the conventional orthogonal multiple access approach ineffective for 5G communications. Thus, to specify the spectral inefficiency Multiple-input-multiple-output and non-orthogonal multiple access (MIMO-NOMA) were introduced. Here, MIMO and NOMA are integrated to earn to improve the channel capacity and spectral efficiency. However, the high Bit Error Rate (BER) and computational complexity in NOMA_MIMO due to successive interference cancellation (SIC) reduces the system performance for edge user. Thus, different channel estimation techniques are developed in the past to overcome these issues. But still, they face challenges in complexity and error rate. Hence, a novel hybrid Whale optimization algorithm with a Radial Basis Function Neural Network-based channel estimation method (WOA-RBFNN) was proposed in this article. The developed model estimates the path for data transmission for edge user and tunes the channel parameters till it attains their optimal value. The optimal fitness function in the proposed model offers the finest system performances in terms of Bit Error rate (BER), throughput, etc. Furthermore, the results are estimated and compared with the existing techniques for validation purposes. The comparative analysis proves that the developed model earned better performances than the existing ones especially for edge users.
Publisher
Research Square Platform LLC
Reference39 articles.
1. Parekh, N., & Joshi, R. (2022). Non orthogonal multiple access techniques for next generation wireless networks: A review. In Proceedings of the International e-Conference on Intelligent Systems and Signal Processing (pp. 171–188). Springer, Singapore. https://doi.org/10.1007/978-981-16-2123-9_13
2. Multiple access in cognitive radio networks: from orthogonal and non-orthogonal to rate-splitting;Gamal S;Ieee Access : Practical Innovations, Open Solutions,2021
3. Deep learning based successive interference cancellation for the non-orthogonal downlink;Luong T;IEEE transactions on vehicular technology,2022
4. Baek, H., Yun, W. J., Kwak, Y., Jung, S., Ji, M., Bennis, M., & Kim, J. (2022, May). Joint superposition coding and training for federated learning over multi-width neural networks. In IEEE INFOCOM 2022-IEEE Conference on Computer Communications (pp. 1729–1738). IEEE. DOI: 10.1109/INFOCOM48880.2022.9796733
5. Optimal power allocation for NOMA-enabled D2D communication with imperfect SIC decoding;Yu S;Physical Communication,2021