Scholarlink

It is expected that mobile traffic will increase rapidly in the coming years due to high demand of real-time communication. The fifth generation (5G) is seen as a new generation of technology that can improve the quality of services (QoS) and enable user satisfaction by providing higher data rates and low latency. An advance prediction of 5G throughput improves QoS. The authors of this paper propose an ensemble machine learning scheme to predict 5G throughput to improve the QoS. The support vector machine (SVM), decision tree (DT), and K nearest neighbor (KNN) machine learning models are trained separately in this ensemble machine learning approach. The output of each model is combined and the final predicted throughput of 5G is calculated based on the maximum voting rule. The proposed ensemble model approach obtains optimal results with an accuracy of 88.80%, a precision of 89.19%, a recall of 88.71%, and F1-score of 88.71%. The ensemble model obtains an area under curve (AUC) of 94.90%.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

5G Wireless Technology Throughput Prediction Using Ensemble Machine Learning Approach

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Abhilasha Sharma,
  • Salman Raju Talluri,
  • Shweta Pandit

摘要

It is expected that mobile traffic will increase rapidly in the coming years due to high demand of real-time communication. The fifth generation (5G) is seen as a new generation of technology that can improve the quality of services (QoS) and enable user satisfaction by providing higher data rates and low latency. An advance prediction of 5G throughput improves QoS. The authors of this paper propose an ensemble machine learning scheme to predict 5G throughput to improve the QoS. The support vector machine (SVM), decision tree (DT), and K nearest neighbor (KNN) machine learning models are trained separately in this ensemble machine learning approach. The output of each model is combined and the final predicted throughput of 5G is calculated based on the maximum voting rule. The proposed ensemble model approach obtains optimal results with an accuracy of 88.80%, a precision of 89.19%, a recall of 88.71%, and F1-score of 88.71%. The ensemble model obtains an area under curve (AUC) of 94.90%.