A Comparative Study to Classify and Predict the Throughput of Fifth Generation Wireless Technology Using Supervised Machine Learning Algorithms

In the modern era, the demand for 5th generation (5G) communication technology is increasing day by day due to the increased data rate, higher bandwidth, and lower delay time of 5G. To find the throughput range or its expected value in a particular slot, the classification and regression models are used. The present research applies three machine learning algorithms to predict and classify the throughput of 5G. The data for this study is obtained from the internet repository. Two classification models and two regression models are tested to predict the throughput of the millimeter wave (mm-wave) 5G dataset. The performance of classification algorithms is verified using precision, recall, F1 score, overall classification accuracy, and speed. It is observed that the random forest (RF) classifier achieves better values of all the performance parameters as compared to the support vector machine (SVM) classifier. The performance of the regression models is checked using root mean square error, correlation, R-square, and execution time. The experimental results show that the random forest model achieves better values of these parameters as compared to the generalized linear regression model (GLM). In addition, the observations show less execution time of the generalized linear model than the random forest model.