Deep Learning-Based Prediction Of Channel Quality Index For Enhanced 5G Network Performance

The rapid evolution of 5G wireless communication has created new opportunities for high-speed connectivity, but it also presents significant challenges in ensuring accurate channel prediction and efficient resource allocation. This work proposes a machine learning-based framework that utilizes Convolutional Neural Networks (CNN) and Support Vector Machines (SVM) to predict the Channel Quality Index (CQI) in 5G New Radio (NR) systems. By integrating deep learning with optimization techniques such as Particle Swarm Optimization (PSO), the approach enhances CQI accuracy and reduces system-level delays caused by outdated feedback. The model was trained and tested using MATLAB simulations under both single-user and multi-user mobility scenarios. Results demonstrate that CNN outperforms SVM in terms of prediction accuracy and downlink scheduling efficiency, ultimately improving transmission reliability. This study highlights the potential of deep learning to strengthen 5G resource management, offering a scalable solution for handling high user density and diverse traffic demands.