<![CDATA[This study proposes a Convolutional Neural Network (CNN)–based approach to enhance the intelligence of MIMO antenna systems in Internet of Things (IoT) environments, particularly for modeling the relationship between wireless channel characteristics and achievable communication capacity. Modern MIMO systems face complex challenges due to dynamic channel conditions such as noise, path loss, and multipath fading, which significantly affect data transmission quality. In this research, channel-related features are processed through a structured preprocessing stage before being fed into a CNN model to learn nonlinear relationships among channel parameters. The developed model is designed to predict achievable channel capacity accurately as part of an adaptive and intelligent wireless communication framework. Experimental results show that the proposed CNN model achieves a Test Loss of 0.0317 and a Mean Absolute Error (MAE) of 0.1267 on unseen test data. Visualization of actual versus predicted values indicates that the model demonstrates good generalization across most data ranges, although some deviations remain at extremely high capacity values. Compared to conventional approaches, the CNN-based method shows superior capability in capturing complex correlations among MIMO channel parameters. Therefore, this approach contributes to the development of adaptive and efficient intelligent antenna systems, supporting the growing demands of next-generation IoT communication networks.]]>
Copyrights © 2026
