Analysis of Machine Learning Based Noise Reduction in Cellular Mobile Communication

The cellular communication system is a mobile wireless communication that splits a huge geographic area into smaller cells or sections with less-power wireless transmitters. Recent analyses have focused on addressing issues such as network overfitting, local optima, and the need for large networks. In this study, different Machine Learning (ML) techniques are analyzed which are implemented in noise reduction over cellular networks. This study discussed significant assumptions, advantages, and drawbacks of analyzed ML techniques. ML-based techniques such as supervised, unsupervised and Reinforcement Learning (RL) are utilized for noise reduction in cellular communications. The existing methods' performance was estimated by utilizing various performance measures like accuracy, precision, recall, f1-score, energy efficiency, latency, packet delivery ratio, and error rates. This study concludes that the various noise reduction techniques have the potential to overcome the drawbacks like local optima and overfitting issues.