RETRACTED: Wireless communication channel modeling using machine learning

• Published in: Transactions on emerging telecommunications technologies 2024-10, Vol.35 (10), p.n/a
• Main Authors: Kamruzzaman, MM, Hossin, Md Altab, Alrashdi, Ibrahim, Alanazi, Saad, Alruwaili, Madallah, Alshammari, Nasser, Elaiwat, Said, Zaman, Roksana
• Format: Article
• Language: English
• ISSN: 2161-3915; 2161-3915
• DOI: 10.1002/ett.4661

Channel modeling is crucial in the development of wireless communication systems. Analyzing a large amount of data is typical practice before using statistical methods to construct appropriate channel models. Using channel estimate on top of channel modeling, physical layer transmission at huge bandwidth is a major challenge in current mobile communications. New 5G and diverse Internet of Things necessitate more effective channel modeling and estimation techniques. Machine learning (ETM‐ML) has been suggested in this research as a way to evaluate the transmission medium. With the arrival of 6G and the heterogeneous Internet of Things (H‐IoT), many challenging application scenarios will arise, necessitating the development of a more effective channel modeling and estimate approach. The suggested framework's trustworthiness is proven by the simulation analysis, which is based on correctness and efficiency. The initial parameters are set, and then the wireless physical layer model is designed with a machine learning model. The finally predicted results are plotted at the final display monitors. It applies a nonlinear matching approach in the situation, whereby it supposes a general case at which function Q^(w,t)$$ \hat{\mathrm{Q}}\left(\mathrm{w},\mathrm{t}\right) $$ is undefined. Nonlinear regression is applicable when the observed information is captured by a nonlinear system of the design variables. Several assumptions are used to suit the information. The idea has a benefit over the method, which assumes the PLM is just a black box. As a result, the approach is more general and adaptable, and it is utilized with a variety of PLMs.